V=®-~4I=3fc: ) 4£ =00 for a survey of the route, with the expectation that the Federal Government would aid in the construction. This act contemplated a waterway from the Hudson to Lake Ontario and a canal thence around Niagara Falls to Lake Erie. In 1811 a commission, consisting of Gouverneur Morris, S. Van Rensselaer, I)e Witt Clinton, William North, Thomas Eddy, and Robert R. Livingston, was appointed to consider and report on all matters relating to the inland navigation of the State. This commis¬ sion made an appeal to Congress and to the State government for aid, but without success, and in April, 1812, submitted a report, the recommendations of which were made the basis of legislation for the construction of the canal on the line as finally built. In their report it is stated that the project for a canal suitable for vessels of 50 to 60 tons capacity, capable of navigating Lake Ontario, was impracticable for the reason that sufficient water was not avail¬ able for the supply of the summit level, and inadvisable for the reason that freight for export, once having reached Lake Ontario, could be delivered at Montreal for less cost for transportation than at New York, an argument which is still an important factor in the location of a waterway from the lakes to the sea. At that time a large amount of the commerce of New York City was from the central and western part of the State, making a canal through the State much more advan¬ tageous than a route through Lake Ontario, passing around the west¬ ern half of the section to be benefited, a condition which has been greatly modified by the enormous increase of through traffic and the development of railroad transportation lines by which local commerce is economically handled. In 1812 the legislature passed a law appropriating #5,000,000 for the construction of the canal, but owing to the war with Great Brit¬ ain the act was repealed before any work was done. On April 10, 1816, a law was passed for an estimate for the canal, and the following year the construction was authorized. The first ground was broken July 1, 1817, at Rome, and in October, 1819, the first boat made the trip from Rome to Utica. During 1821 the entire canal was placed under contract and was completed in October, 1825. It was recognized before the canal was completed that its dimen¬ sions were inadequate for economical transportation, and in 1834 a law was passed authorizing the enlargement of the system and pro¬ viding for a second set of locks of the enlarged plan from Syracuse to Albany. In May, 1835, a law was passed approving the policy of enlargement of the entire canal, which was not completed until 1862. 32 DEEP WATERWAYS. CHAMPLAIN CANAL. The law of 1817 providing for the construction of the Erie Canal also provided for the Champlain Canal, which was completed in sec¬ tions, as follows: 1810. From Fort Edward to Whitehall. 1820. From Whitehall to Fort Miller. 1821. From Fort Miller to Stillwater. 1822. From Stillwater to Waterford. • 1823. From Waterford to Albany. This canal was gradually enlarged and the locks replaced until 1860, when the legislature directed that the depth be made 5 feet and the bottom width 35 feet. OSWEGO CANAL. This canal was authorized in 1825 and completed in 1828, with the same dimensions as the original Erie Canal. The report of the State engineer for 1854 states that the business of this canal was greatly increased, at the expense of the western portion of the Erie Canal, by the completion of the Welland Canal, indicating that the opinion that freight having reached Lake Ontario would not be locked over the divide to the Mohawk River was not well founded. The following table gives the dates of construction, dimension of prism, size of locks, and cost of these canals as originally built, and for the enlargements completed in 1862: 1 N ame of canal. When au¬ thorized. When com¬ pleted. Length. Width on surface. W id th on bottom. Depth of water. Number of locks. Length of locks. Width of locks. Original cost of the canal. Cost, in¬ cluding enlarge¬ ment and land dam¬ ages. Total cost, including interest on loans. Miles. Ft. Ft. Ft. Ft. Ft. Ft. Erie Canal_ 1817 1825 40 28 4 83 90 15 17,143,790 Enlargement.. 1835 18(52 35(11 70 56 7 71 110 18 838,977,831 852,491.916 Oswego Canal 1825 1828 38 40 24 4 18 90 15 565,437 Enlargement.. 1847 18(52 38 70 56 7 18 110 18 3.077,430 3,612,825 Cham plain Canal.. 1817 1822 06 50 35 5 20 100 18 921,011 1.746,063 2.647,002 Total cost of the Erie Canal in detail. Original cost of 363 miles of canal.$7,143, 790 Enlargement of 3504 miles of canal. . 31.834,041 Total cost, exclusive of feeders, structures, and land damages _ 18,439,849 Cost of feeders.. .. ...... 794,618 Cost of structures for old canal and enlargement.. .. 16, 494,218 Cost of land damages for old canal and enlargement. 3.249,146 Total interest on loans during construction and enlargement. 13,514,085 Total cost, including interest on loans.... 52,491,916 Total cost of repairs, 1827 to 1862...... 10,995,333 1 Compiled from the report of State engineer and surveyor of New York for 1862. DEEP WATERWAYS. 33 In 1895 $9^000,000 were appropriated by the State of New York to make the locks twice their present length and deepen the canal to 9 feet. After the expenditure of most of the money the amount appro¬ priated was found to be inadequate and the work discontinued. PROPOSED SHIP CANALS. In 1835 Capt. W. G. Williams, United States topographical engi¬ neers, made surveys for live different routes from Schlossers, on Niagara River, above the Falls, to points on the river near Lewiston, and from Tonawanda to Olcott, on Lake Ontario. The estimates were on a basis of a canal prism 10 feet deep, 110 feet wide at the surface, and with locks 200 feet long, 50 feet wide, and 10 feet deep, and varied from $2,538,900 for the Lewiston line to $5,041,725 for the Tonawanda-Olcott line. In 1853 surveys were made under the direction of a State commis¬ sion from Tonawanda to points on Lake Ontario for a canal 130 feet wide on bottom, 14 feet deep, with locks 300 feet long and 70 feet wide. The estimates varied from $10,290,500 for canal with single locks to $13,169,600 for canal with double locks. In 1863 these lines were again surveyed by Mr. Charles B. Stuart for the General Government and an estimate for a canal 105 feet wide at the surface and 12 feet deep, with locks 275 feet long by 45 feet wide, reported at $6,007,000 for a canal with single locks and $7,680,600 for canal with double locks. In 1867 Lieut. Col. C. E. Blunt, Corps of Engineers, United States Army, made surveys of 3 routes from Schlossers, on the Niagara River, about 3 miles above the Falls, to the river near Lewiston; 1 from Schlossers to Lake Ontario at mouth of Four-mile Creek, 1 to Wilson, on Lake Ontario, and 1 from Tonawanda to Olcott, at the mouth of Eighteen-mile Creek. The estimates were based on a canal 90 feet wide at the bottom, 125 feet wide at the surface, and 14 feet deep, with locks 275 feet long and 36 feet wide, with lifts of 15 feet and 16 feet, and varied from $11,032,000 for the Lewiston line to $12,893,000 for the Olcott line. In 1884 Mr. E. Sweet, engineer and surveyor of New York State, read a paper before the American Society of Civil Engineers, outlining a project for a down-grade canal 18 feet deep from Tonawanda to Utica, and thence by locks and dams on the Mohawk to Troy. The project called for a high aqueduct 2 miles long, at the crossing of the Seneca River, where the natural conditions are not suitable for safe foundations for such a structure. It is, however, probable that, by adopting a low-level canal from Oneida Lake to the Mohawk, and diverting the Seneca River east of Baldwinsville so as to carry that level to Oneida Lake and passing over the outlet of Onondaga Lake, such a down-grade canal could be safely constructed, the only ques¬ tion being whether such route would not be much more expensive and H. Doc. 149-3 34 DEEP WATERWAYS. would take 30 per cent more time to navigate than the route via Lake Ontario. In compliance with the river and harbor act of August 11, 1888, a revision was made of the surveys of the Wilson and Olcott routes by Capt. C. H. Palfrey, Corps of Engineers, United States Army, and an estimate for a canal 100 feet wide at the bottom, 20| feet deep, with locks 400 feet long, 80 feet wide, and 18 feet lift, was reported, as follows: Wilson route. 18.25 miles: With single locks.. ..- $24,201,600 With double locks ............ 29,347,900 Olcott route, 25.28 miles, single locks .... 23, 6i7,900 Under the provisions of the sundry civil appropriation act of March 2, 1895, the President appointed James B. Angell, John E. Russell, and Lyman E. Cooley members of a commission to meet and confer with a similar commission from Canada, and report whether it is feasible to build such canals as shall enable vessels to pass to and fro from the Great Lakes to the Atlantic Ocean. After a year’s investigation and study of available data the com¬ mission reported, “That it is entirely feasible to construct such canals and develop such channels as will be adequate to any scale of naviga¬ tion that maybe desired between the Great Lakes and the seaboard,” and recommended that complete surveys be made on which to base projects for ship canals from Lake Erie to Lake Ontario and from Lake Ontario to the Hudson River via the Oswego and Mohawk rivers and via the St. Lawrence River and Lake Champlain. In compliance with the river and harbor act of June 3, 1896, Maj. T. W. Symons, Corps of Engineers, United States Army, made an estimate from existing data for “cost of construction of a ship canal by the most practicable route, wholly within the United States, from the Great Lakes to the navigable waters of the Hudson River, of suf¬ ficient capacity to transport the tonnage of the lakes to the sea.” Major Symons interpreted tonnage to mean the commerce of the lakes, without regard to the size of vessels in which transported on the lakes, and recommended a barge canal 12 feet deep, 82 feet wide at the bottom, and estimated to cost $50,000,000. The history of the Erie Canal shows the amount of freight carried to be as follows: Tons. 1837.. 667,151 1850.. .. 1,635,089 1860.. . 2,253,533 1870. 3,083,132 Tons. 1880. 4,608,651 1890. 3,303,929 1899. 2,419,084 Since 1880, when the traffic of the Erie Canal reached a maximum, the rate per ton mile on the railroads from the lakes to the seaboard has been about double that on the Erie Canal, yet during that time the business of the canal has diminished about one-half, showing i DEEP WATERWAYS. 35 beyond question that, the volume of freight which will be shipped by any given route does not depend entirely upon the relative cost of transportation, and that, unless the conditions which have produced the decline of traffic on the canal are changed, any increase in the carrying capacity will produce no material increase in the volume of freight carried. CANADIAN CANALS. Since 1821 the Government of Canada has expended upward of $80,000,000 on the construction and improvement of canals, with results somewhat similar to those developed by the waterways through New York. The first Canadian canals constructed were those to overcome the Cedar and Coteau rapids, which were commenced in 1779 and com¬ pleted in 1781. The locks were of cut stone, with a chamber G feet wide, 24 feet deep, and designed for boats carrying 30 barrels of flour. These locks were enlarged in 1804 and 1817, and abandoned in 1845. A canal with locks to overcome the rapids at Sault Ste. Marie was constructed by the Northwest Company in 1797 for passing loaded canoes. One of these locks is still in a fair state of preservation. In the St. Lawrence River canal system— The Lachine Canal was begun in 1821 and opened in 1825. The Beauharnois Canal was begun in 1842 and opened in 1845. The Cornwall Canal was begun in 1834 and opened in 1843. The Farrans Point Canal was begun in 1844 and opened in 1847. The Rapid Plat Canal was begun in 1844 and opened in 1847. The Galops Canal was begun in 1844 and opened in 184G. The Welland Canal was begun in 1824 and opened in 1829. The Welland Canal was enlarged between 1841 and 1850, and in 1873 was made 12 feet deep, and in 1887 was deepened to 14 feet. In 1871 it was decided to enlarge the canals of the St. Lawrence River system to afford a depth of 12 feet throughout, which project has since been modified and all the locks on the route made 45 feet wide and 14 feet deep on the sills. The locks (where not made for fleets of barges) are 270 feet long, and are intended to pass lake steamers of 2,000 tons capacity. The completion of this system of waterways from Lake Erie to tide water, capable of passing vessels of twice the capacity of the proposed New York State Barge Canal, does not indicate that the causes which have produced the decline in water transportation between the lakes and the Atlantic have been overcome. LAKE HARBORS AND WATERWAYS. The entrance to Niagara River from Lake Erie was originally a wind¬ ing channel about 17 feet deep at mean lake level through Horse Shoe Reef, which, in accordance with a project approved in 1888, was deep¬ ened to 18 feet for a width of 400 feet. This cut through the natural 36 BEEP WATERWAYS. barrier to the discharge of the lake undoubtedly had the effect to slightly lower the natural levels. Previous to 1819, the mouth of Buffalo Creek, which constitutes the interior harbor at Buffalo, was closed most of the time by a gravel bar. In 1820 and 1821, two piers about 200 feet apart were built across the bar by the State of New York, which were transferred to the United States in 182(3. From 1830 to 1807 various projects were submitted for the improve¬ ment of the harbor entrance and construction of breakwaters, but none of them were carried out until 1868, when it was decided to repair and extend the piers, dredge to a depth of 15 feet between them, and construct a breakwater 4,000 feet long in 27 feet of water outside of the harbor entrance. The project lias since been gradually enlarged until the present, and consists in extending the detached breakwater to Stony Point, about 4 miles from the entrance of the harbor, and to deepen the harbor for 20-foot navigation. The 21-foot curve in the lake is about 1,200 feet outside of the outer end of the piers, and the 30-foot curve about 14,000 feet outside of the entrance. The entrance to the harbor and deep waterway channels, if less than 23 feet in depth, will be protected by the breakwater as now con¬ structed ; but for greater depths the channel excavation will be in the open lake, across which at times there will be very heavy currents, which will probably produce rapid deterioration of the channel from the material carried by the currents and ice. The harbors on the south shore of Lake Erie had original depths of from 3 feet to 6 feet, which have been generally improved by building piers about 200 feet apart and dredging between to depths of 16 feet to 19 feet. The 30-foot curve in lake is from 3,000 to 4,000 feet from shore in front of the harbor entrances. At the mouth of the Detroit River, with the exception of a few sand bars, there is a natural 21-foot channel into the lake at mean stage, but for a 30-foot channel the bed of the lake would have to be deep¬ ened for a distance of about 10 miles, from which points the route through the islands, about 35 miles, varies in depth between 30 feet and 40 feet. The Detroit River, with the exception of a reach of about 5 miles near its mouth, has a natural 30-foot channel. At the Limekiln cross- * ing, about 1 mile above Amlierstburg, the original depth was only 13 feet, above and below which for a total distance of about 4 miles there were reefs of bowlders and rock, with depths of 15 to 18 feet. Through this reach of the river a channel has been excavated having a width of 440 and a depth of 18 to 20 feet. The river currents are at places across the channel, making it dan¬ gerous to navigate in its present condition. Previous to the improvement of the channel across Lake St. Clair a DEEP WATERWAYS. 37 large shoal off Grosse Point obstructed navigation for vessels drawing over 16 feet at mean stage, through which a channel 20 feet deep lias been dredged, Lake St. Clair lias at present a depth of 20 feet at mean stage from the head of the Detroit River to the delta of the St. Clair River, a distance of about 15 miles, and in order to establish 30-foot navigation between Lakes Huron and Erie a channel 10 feet deep would have to be excavated in the bed of the lake for this distance. From the South Channel of the St. Clair River into the lake there was originally a narrow winding channel from 9 to IS feet deep. In 1871 a straight channel 13 feet deep between piers 300 feet apart was completed from the deep water in the river to the 13-foot curve in the lake. This channel was deepened to 16 feet in 1873, and to 20 feet in 1894, for which depth the dredged channel extends 1 mile outside of the piers into the open lake. The St. Clair River has a natural depth of over 30 feet, except across the middle grounds at Port Huron, Marysville, and St. Clair, where the depths are from 21 feet to 24 feet. The maintenance of a 30-foot channel through these reaches will require a rectification of the channel banks or dredging at frequent intervals. The general depth of the foot of Lake Huron, 14 miles above the head of the St. Clair River, was originally about 21 feet to 27 feet, over which were scattered numerous shoals with only 16 to 18 feet of water. A channel 2,400 feet wide and 21 feet deep at mean stage has been cut through these shoals. At the time of tin 1 last complete survey of the head of the river, in 1867, the depth across the bar over which the lake discharges into the St. Clair River was only 27 feet, and through the gorge at the head of the river the central depth was 48 feet. Investigations made during 1898 and 1899 show that a channel has been scoured through the bar 75 feet deep, and the depth in the gorge at the narrowest place increased from 48 feet to 66 feet. There is now a channel over 40 feet deep from the lake into the river, the increased outflow through which has lowered the general level of Lake Huron and Michigan about 1 foot. From the foot of Lake Huron to Milwaukee and Chicago there are no natural obstructions to 30-foot navigation outside of harbor entrances, although the marking or removal of some shoals at the foot of Lake Michigan may be found desirable. Chicago River is a small stream, which, before improvement, was closed most of the time by a bar across its mouth. Its entrance has been improved by piers and dredging, and the river widened to a width of from 200 feet to 300 feet, and dredged about 18 feet deep. For 30-foot navigation on the lakes the business of the port would have to be transacted from slips constructed in the outer harbor, where the depth of water is now only about 14 feet, or by deepening the river to 30 feet. 38 DEEP WATERWAYS. The St. Marys River is the only waterway from Lake Superior to the lower lakes, and before improvement was obstructed in many places by bowlders and rapids, the principal fall being at Sault Ste. Marie. The fall of the river from Lake Superior to the head rapids was about 0.5 foot, in the half mile of rapids 18 feet, and from foot of rapids to Lake Huron about 2.3 feet. Since the improvement of the river and the enlargement of the head of the St. Clair River, the gen¬ eral level of Lake Huron and of the St. Marys River has been lowered so that the fall in the rapids is now about 19 feet. The governor of the State of Michigan, in 1837, recognizing the immense importance of the timber and mineral resources of the Lake Superior region, called the attention of the State legislature to the advisability of constructing a canal around the rapids at Sault Ste. Marie, and three years later the importance of the project was dis¬ cussed in the United States Senate. Nothing, however, was done until 1852, when a grant of 750,000 acres of public land was made to the State of Michigan, from the proceeds of which the canal was to be built. Work was commenced in 1853 and the completed canal turned over to the State in 1855. The canal was 5,400 feet long, 100 feet wide at water surface, and 12 feet deep, with locks 350 feet long, 70 feet wide, 11^ feet deep on the miter sills, and 9 feet lift. The branch of the St. Marys River flowing to the north and east of Sugar Island was considered best adapted for the improvement required by the depth of water in the locks, and a project was approved in 1856 for removing the obstructions to navigation by dredging to a depth of 14 feet through Lake George and the East Neebish Rapids, which work was not completed until 1871. It was recognized soon after the completion of the State locks that the dimensions of waterway and facilities afforded were inadequate to accommodate the growing commerce of Lake Superior, and in 1870 the General Government commenced the improvement of the canal, comprising the deepening to 16 feet and the construction of a new lock 515 feet long, 80 feet wide, 16 feet deep on miter sills, and 18 feet lift, which was completed in 1881. The lowering of the level of Lake Huron, which has occurred since the completion of the work, has increased the lift to about 19 feet and diminished the depth on the lower miter sill to about 15 feet. In connection with the lock construction at the rapids, the obstruc¬ tions in the river channel between Lake Superior and Lake Huron were dredged to a depth of 16 feet, which work was completed in 1883. The commerce of Lake Superior reached such proportions in 1884 that a project for a larger lock with 21 feet of water on miter sills was proposed, and in 1886 the project was modified and work commenced. This lock, 800 feet long, 100 feet wide, and 21 feet deep, was com¬ pleted in 1896, and in the meantime a channel 300 feet wide and 20 feet deep was dredged through the Hay Lake and the Middle Neebish KM*** JULIUS BltN & CO PHOTO LiTH — State Pocks opened. 1855 1- 1 1 1856 I 1857 1 1858 1 1 v. ft ^ > o > ® S u => §. 1 s'! * ^ a- Q «o $*> lJ) ^ ^ 5 § 8 if 5 Vh ^ 8 ^ § > 5 ^ ^ $ § $ c£ pi 1 C$ § N 1859 -i- \ 1 1860 1 1861 ! c? 1862 “\°9 l (a' s 5> $ 1863 -V 4 1864 l C\> * CN 1865 — V.— S vi 1866 “i'O' 1 N 1867 Hu - V \ . 1868 \ § L'l £i Co b N H !| 1869 1 1 hi 1870 1 l t n 1871 \ — 1872 \ 1873 / 1 — 1874 t l 1875 \ \ \ 18 76 r i i ^ ^ q > ^ g 1M .V 1 1877 \ \ 1 1878 i UP > > 1879 i 1880 \ <• cui'n U -CO ^ ^ >i Weitzel Pock opened. 1881 \ \ v _ _ 1682 \ / $ Oci §• Co \ o t - ' \J H i\ p \3 tt 1883 1884 1 1885 v\ > 1886 \\ O \ 1887 L— 1888 ,y„. > V > . --S biv — 1889 > Hu 1890 i / ! A LiL ->v o r f\ 1 1891 u 1892 ✓ s' ■' / 7 c® 1893 *•-. V Co _ 1894 S'- Canadian Lock opetied. 1895 \ \ Poe Pock opened. 1896 \ — — 1897 —V ~ ■' W **i S 1898 — 1899 _ 1 1900 si 3* cj a Is I S3 2 , £1 £| cS*j <« H gill 2 0 _g ?: a _c "-4-J s 03 3 t S o o •13 -J3 03 03 © - 4 J c '$ 2 b ® (3 C 13 5 o: B 13 2 S 8 11 $ 38 o I © > i a >, S .2 ■s 33 £ 3 £ fa © .5 E-t a Si • ■3 “3 2SS 21 (4 S 2 m 2 3 <•-» O U 3 cS •S _o *2 ■§ a 1 © •c a © a i * 1 2 (U o A © £ ©cocoooocc £._£'C r T r ^ r C r © r d r ©! f ©' r © Ji. © © •r eg ® O o c o o o o ococccocj^cgo ^ r ^ r ©? r ^ r © r © r d r ©"w r C r C r ^T; r u r ©r© t 2 ~ ~ r© r ”' 1*1 loo ' 2 • 2 ■ 2 EH3S£HSH$eh3 S 3 'S 3 '33 *s 3 ■ • S 3 'S 3 ■ S 3 • £ S 3 £ V- %2 -.2 §1 3 M -*-< £ a e :fc ! •. » = is : = 55:223 S 22 TO i — s 2 8 «s gi2SS?i i Tan gent: |r; c\i • ; r-I oi i-H o X s •I o Eh o H s tl © > 5 CO £ >3 03 «a + + 1313 ?5 'B lsBSSI8i8S1138?SliSlSlil!ISIfi§il£ISS ; p - • B o' 33 ■ 4 J 12 55 + fc3i3 + + B b B® M cr. f ® 5& sjSgg. II © 35 C O 3| S 3 c F? IlddSlillllllillgilllllsillBlIlllBll Jj Table of distances, alignment, width of channel, sailing time , and water levels —Continued. [21-foot channel.] 90 DEEP WATERWAYS. a O o •g a iS o w . a 44 44 72 72 ?a 3 1 ? «§ S cS p Oh 33 c 3 ■33 1 a 4-3 ‘a 5

ift p p o s 4 8 5 § a a aJ r=i 44 £ 44 44 ©©ooooooooooooooooooooo ££^r 3 ^^'CH 3 ^^^^'d'd'c'd' 0 'd'a'd'd X t© ooooooooo ' r C *w *d T 3 r G *d TO o o ■O'G o 'd © © © K«w 3°S a o ' a _ gxv a‘- a' a ^ ar-.T' n ' a- a a Eh © © £ . 44 V) © a *“ o ©44 2 a 44 a » 44 73 73 © . PX) co 00 01 1 •—( —< X 1-H .p t- o co i- X X -2 T—( • 1—1 r—H ,-H r-H y—i -H 1 1 £ 2 1 1 x 1 H 1 1 S3 .2 44 a 44 72 ©H CO X O' X -r X OICOC^ CO CO CO — -4 >.o • x o> x -h x1-»o ©■ —> ».© i.O 1- l-*X C5 c »C X »© t- 01 0> Cl I'-l-J— X-— 1 ©3 01 -—< 0© CO CO »© x co 01 co x >© x x co x co • © x ot x. — ».© j»o-ri*T-arHio ©CHCD^-i'JNNNXXCJaC©- 01 2 > co of — — — -( r-i — —4 — —t ^4 ^ —. 01 01 02 ©1 ©1 fl ©l 01 01 ”4 01 ^1 CO P ^r 1 Zo io X O *.4 -s ■* la -J 71 £r £r 26 rr* — * * — -* -e - ► ■-. « . ^ ^ 0 © X i© CO X 0 © X CO »-0 X 01 x. — — *o — x — x. — >.© C C - 0 ! 1C -i X N DD X X c. C. * “- 0 ! 0 ! co ©5 Table of distances, alignment, width of channel, sailing time, and water levels —Continued. DEEP WATERWAYS. c8 a o o A J-4 jg O -4-i © X cS J *+ 5? :£ > gSa 3 O Z O fci rfl-S £ Q + N M N CO £ 3 o.2 i 4-> © =* S-ji -Ji > s« +J I § ' -» — t »w S3 ’■3 n 5'3 xa sg • go — y .2 © -g* Xt-3 tfleS 2 oc \ 2-2 3 g = 3*?s 02 £ ■« 2 5 O (« co g 5 S ;^h :=a 1 1 • 30 ! GO I 40 •40 1 • m S • * 40 • 1 « ' 4 O • ^ • ■ ^ • i— • . t- 1 l 1 • N 1 1 1 - ■ t- • t- • »o • I 40 i 1 1 '40 1 1 ' 40 1 4.0 ' 40 4^ .855 3 o o o o o r ^ p C r O r d r OJr o o o o o 000000000^0 -M , © ! :o 2 3 © © i£ O *- © d Q « © o I £ • . CO 1 © d ■O o ©+3 © c$ s -te cS V I -*-' 2 ! S o cs d • ' 2 ? —4 ■ -H *o • CS t- ! GO • 4 ^ •3 ■» CS ■ 40 40 ' X. 40 ~s ; i g « : ■ t- ^ ; • CO ^ • 1 1 1 ■ i i 1 1 • 1 ■ 1 ! ■ ajt- ^8, r? • f? i — . . —. . : — ?l 4.0 N«»» o H 3 4? cg o h -© »0 ».C O r-4 «-H i-4 H—I—h ZC U I" 40 ‘O 40 iO >0 40 4.0 40 r: — -r ot zi ri co 4.0 O 40 l-"? 40 4.0 >0 i© i + + + + +'- *0 40 >0 40 «5 40 tO G© GC —H 1—H ++++++ ..._^-xpN?!xxN:!^:iN“C“rci'«r. ;_, . - c t- c: o- *t -*• t- cs ?i i- x —i t- © co >o c: c* — c: — 5* — — co d c * co o :© t- cs —< 1 - CSS 40 40 »o H—I—h -O —< -4 01 40 >o io >o »o »o >o »o »o »o »o »o »o »o »o »o »o >o »o *2 *o »o ++++-4-4-T ++4-+ 4- . 4 ' -i .- - -t ?» 3 :? -:ci* r. -t ci-t 1 !' r; :»i* x ^--ri. ~ “Ti* - r. —■ tc — ■—^t ^O 13 *3 d hQ d a 91 92 DEEP WATERWAYS. a Elevations of water surface are those for a regulated stage of Lake Erie and a standard low water of Lake Ontario, and are given immediately above and below locks unless otherwise noted. Elevations referred to Buffalo bench mark. DEEP WATERWAYS. 93 •g 3 a ! s § 'tj 3 o s I i — H 5 D 8 O I S? © c s i | s Si 1 ■s “3 3 I > s — 1 ! 5 | g j o •a & P PO p p a a v o XJ is co HH i? a j :§ S 3 j o •43 cS A Eh a 1 3 c Eh 2 a <0 iH S=T £« "fc TJ • I §|.2 8 £3 £ * 0 5 * T . .. s -,40 2 £.a£ g'S © a ® g g ! ^ H I I OC =5 'P £ •- I© 8 C » = 4 -H 5 is •x .2 000000000000.30:0 “ 1 1 r C r C r C , C r O p S r G r C ^ r C r C r 3 . $ ' ■si oc > a *a . h ~ a * * so - • c ■ c a a „ 1 cs , acs „ eSs s - , 88a«v=0v {S HE-^-H-^E-i-iE-* —Eh-^-^Hcc^Eh^E - 1 0= s? 3!S I :3 SS ee 33 fc. _ S3 © fl 3 ss OO ,3 i C ♦ 8 . s SS 3 2 i: a 2 :s — X + + a sssssssg^s ~ ^ * “ — — — -J-L -1- -I— -i- -i-U + + + 4 -Z-- ■ p c 33 o p ' G t-J o 1 4-H G = = •=« ■ i+ ._. . ;+•+- - w ii~o = .cj«! s ° -36 + + M§ + + -. = « ’++ '£3 3 3 3 SS2S*Ssiig858Il^l?HISS»Sia Table of distances, alignment, width of channel, sailing time, and water lei'cls. OSWEGO-MOHAWK ROUTE, LOW LEVEL. [21-foot channel.] 94 DEEP WATERWAYS. >> £ c ® i, ® ^ r» S b S-2 % “--=5 M-i ^ c tx c > a q Sts c £ C P ;§® X — c 03 O (. i c. fc h o Ill C P o a T ] a a c a ee q c « £ o > © ^ x £ o cr C C~ci c eft a-' a £■ o' - 2*2 © & O $ o u o . ci ~ O a) ~ > ft V ftift *.5*0 t. SC ® M ® ffl £ c C O p* to c Sc c cr. o c W c o5 5 P . CL,iO a • • a & 2= P c c ft ft if? a its r~* £ g 6 c si X> Ol to Q ! . c: s oo*- CO £ © ft ci © 5 CC c o o o o o CC ®oocoocoooc c£'O r C r w r e r C r C r e p ©^ © a) ® P cm > ti£ C S- ® 52 o p fl • 0 ■ C cSa Pa Pa ci H c~ H » E-* 5 Eh . a a • ce 1 ci 1 ci 1 ci ci ® ® £ . +i C/2 © P ft O © -4-^ © ci P ci « -p cr. .co CC Th o H C -4-S ci -*-> cz: © ft ci + +lO I } } . _L_L"r — -r + + + CC± tT * 1 T 1-CC^^?1XC-.?P I I U't- X c I I »* rc 'r w » cr. — — ?! oj ci o-i ci oi co ft c~ X »! 4- + + 4- + 4-H- + + 4- — ~ ».o — >o c r: >o CO CC ' - - ^ X ^ T X x CO -rh TJ- uo *-0 LO to ci ?fc £ © : X X C3 4J 01 5£ ^ 4 .3 I il^ 5SSS?5S55 + + + 4- + +js. l f;c: OlCO'^r^^C:^-'-"' + LO + - +++++++++- co t o> x c: oico x lo --ic-icc r.cc *T l »CNN3Jwi-- o r. - - o x - o : oi oi oi oi 55 c^ co co co co ir> uo l a Elevations of water surface are those for a low stage of rivers, and are given immediately above and below the locks unless otherwise noted. OSWEGO MOHAWK ROUTE, LOW LEVEL—Continued. DEEP WATERWAYS 95 r3 °| {, § © 5 C 3 g ? rt ° ^ CTcS •—I w HE £ © © © fc£C ~ O p Cl © S3 © © & . 4-3 CO Jg P S3 44. c3 co •g co CO g o *-C o3 4 - GO © >• 6 g+s «3 p Sh © M fcJD A E-i o s- £ © eg C *4 ft ft c5 gP ft & © u V ^ c o ~ . u ft co ft ? 9 g £> « s o ft c3 u 6 6 |QQ o io s'? ® £'u I.S , rn ^ £ © r-2S> P S3 ^ c3 > © - + > _~ 4 ? —H ® c; ’“T 03 c > c 8 ft co r T r > 44 ) O © c 3 co i£ ^ © t 23 >> © © & oft 5 -f- ft *43.. . 2 t- SR 44) CO to c3 . rp o G^ - 44) X c3 & 5 O ftft a ? _ _ £ cS x S?5^c: £8 O r?> £ © ~ £ © ©-5* MJ © c3 C3 4- o © o © . W « CO 1 3 0 0 0 0 0 0 2 r O r C r O , C r © ss © '*3 4-3 & o o o o © o ©^ o o G • G Sv ci^. <£ . ft 55 H 55 H 55 H - ’ o •' ' ! 1 o grp I C • G'G a* ■*■ ^ «- c3 1 1 c2 1 cS Sn^ggc^ + o © r— © — «M r- O G'O cS ' . 5 CO c* © N ._ II 3SSxS?»SS52ccSSSSS5SS^3^ + + -f- + + + + + + + + + + 4- -(- 4" + + + 4“ + *-T »0 OXiCCiT. XH)ONMM35»5 CNpH-+ -ft~ «o--r x < i ri cc tz »cf i- r-< cs co j? *-© 51 co co CO if b + + 4 - 4 - ’ ft 5? ?* *r x + + -, •“ - i, ij.S ijj ^ Table of distances, alignment, width of channel, sailing time, and water levels —Continued. OSWEGO-MOHAWK ROUTE, LOW LEVEL—Continued. [21-foot channel.] 96 DEEP WATERWAYS. A -a © 5 6 g ' 43 ' !* P a A A p, a v a P © oj ufc M | S 5 ci o 3 a £ Si'r £ a CO a g I •P « 2 I S 73 ® O O -2 Hi >occccooco 2 r c r 3 r a r c r c r c‘ r c r d r c coooooo?o § 5 O Crp o i-O- P'3 s a • a a ■ a ■ a ■ a ■ a <8i,Si.cSaSa:S^cSaS s 3 > £■» u s o a'13 a| ii p * S 3 .2 Q © ■ p > p p > > p da Pa. P P v a. P Eh A H ^ Eh Eh ~ ~ Eh s • ~ gS P a, a p _ E- 1 AEhEh A £ i£H Joi £8 3 SB 3 55 :© gg'AS J 32 A EH © + A e© g B3 ? g = 3 ?: V: A V: 3 g ff g ~ U 3 g g rt g £ g § £ g A g 5? £ g g g g ?: -f—h H-—I—h—I—h H~ —I—4—h—I—I—(- -f- H—I—|— £ + - s ~ >1 r*\ — AT .-r , , r ^ ^ ^ «S3383 8 A 5 $ x g A 3 5 g g » fc 3 8 g “ A 3 g g g A A 3 A $ g g § P 55 + +++ + -t--t- + + +++-f + 4- + + + + + + + -i-+ + + +++ + + + + + + + sllllllllllllllllllllllllllllllilllll OSWEGO-MOHAWK ROUTE, LOW LEVEL—Continued. DEEP WATERWAYS. 97 0) a *-3 a o O I o CO q5 o -o 9 **o CO •-o rO g G eg A O c o d u 5 ' r 4-< d Q 5 £ 35 S ^ 3 O O k B ^ 3-.il £ dj? d © _m © 5-o £ ~o o 03 G3 O d d o o 2dS2 gp-ss 5 3,3 rG o c3 O £u & bC O ~ © 3 Q W d ^ d d ^ d c*H5*EH£' , xxH © I g £ -4J ; © ; r °*' © j G -I d t- G o I CD G •£ eg G Eh © © o o o © o o ^'d rd <2 H3 r0 ■'‘s'’ CO eg ?3' ' ^ ;S ; . -*f —T ' 1 1 1 < 1-0 • i£0 •-■< •i'-O • • ' 1 OC ' 01 CO ' CO •—1 1 ,1,1 • I • 1 0 • >0 1 >0 ' CO CO •!*•*'*•**' j 1 • • ' 1 • 1 • 1 • 1 1 1 > 1 • 1 • 1 1 1 1 I t 1 1 • 1 1 • 1 1 • 1 1 • 1 1 1 1 1 1 1 1 1 « « I 1 ©oocoocc £ r G p O r Cr0 p dn0r0 © o o o o ! r 3 r G d di^d d —* “ d ^ gi d be 10 l— a ot ! t»rH ; OHHWO *01 I ! ! <-h X ^ «C0 »C ' CO : ' t'- ' ohqco In !x ' co 01 h co 01 • x •a 1 • 1 • 1 1 • II • 1 • r—1 iri • ; oi . o H d 4-1 gq a o CH ! OI 01 Ot O 05 ce 01 X' CO oi co 01 X © OlOXHOl^OlCi 0 4 cr.- i - OS »Z0 05 05 05 X 01t- i- t'- CO 01 35 X CO CO 05 02 JCCOJOHOO + 4-4-4-4-4-4-4- ^sssafess' 01 01 XX -* -**O»f0 »o »o uO iO >C lO o >o. +++++ + + + 8SSfe§ ».0 Ad !o ? -+• 01 05 t- + + CO^H X X C5 »o ssg + + + CO 01 01 S83 $22 4" 4*4* 4- xiHOia X X *D 05 e5 S § S 4-4-4- X 05 1-1 01 CO r-i *■+ »Q X X -H _l ^ 01 Cj *0 X to CO CO CO -H< "4 -H< V' co co co co co c H. Doc. 149- •/ appr 98 DEEP WATERWAYS. CD a o O I CO C- <£> T3 Q O O I J > w 2* s ^ e j CO ^“H ^ £xj §> £ S £> 5 o Js « w £ s H* 6 £ 3 ^ w s S § £ &. s ft° o A A U a? +3 CO s r ^; p P •P CO w H P Cc o f+ i 02 C3 ‘S3 Eh ® tie I o C crx X O H- C> C! H H- H ? I r. 31 X M X ^ 1 W C X QQ H C ^ M ^ C5 OOH^p^ 05 W^OlC VOrH CO O CO L* Iff H C C lC ^co^wJowccco^ ■ + + ;.o©w mu: co 05 »oici—icr. -hco»oco> o ichnos tlCOHNNC5®aCOHOOW:__-__ _. __..._ .. _. COOONOHf/jC © -t* CO 1 05 —I —I 7 * Cp -h © i- 05 C CO i- X C CC »c '-H *o C cc cc cc i - T" i ' CO' 05 05 a * »—I *—I —(|>1 CQ S'* C'l W .J S £ •+^ 3 Oi CO w 5 ■K- GO O o p H Eh L) O « W £ 05 fc,& 1. O S^s £Q ft® P fto o O P u . P> 71 P,0 . ® 2 ® g;®* ftc o io ^ , - H -5 * «.s cS fB m o - © s 22 ^ s § p P3 o tn>o ^ © p P © © ■sg ^ o © 2 p p eg CO +3 CO so © > f-1 p o ; © ’© rQCOOOOOOOOOOO p f d r d r d r u r o r o , d r d r o , d r o r o P CO P * P < P » P « P C?X ^ A P _ P X P „ P Eh^EhSH^^SH^Eh ?5 it rrt ■ r r ^ d Tl 2 ; >> ®3^ ci t. in tr'+ Jt ci 'a o _, _, 0 “'d g' 3 s ®^23 g co 'OOffiKco p p^,5u p p p p HHgecen H P P H ce ■ a 00 1 -/ P 'O p'd , p ~ +* p ’p’p^f WHiaQ 3 8 p -M CO' H © © © ©5 »-0 lOHC©-*©Xt^*-H© CD CD © l - CO X . H^N-OX*X?l-H w3J '!THi.'ONNrLOrH--H + + + + + + + + + + + 4 - +++++++++++ + ce ^ »C r-H © »0 © *—' l- © 01 -+< -H - 00 © 01 CO »0 © O'} -+- U3 © © -M © 30 CO ^ CO 40 GO 00 C'i CO © © ? t 01 Oi 30 'X NNNcocoao5©oocr-iH?iwf:'+'+^'+HfHf-+■ l- L- 1- t- t- t- l- i- X X X GO X 00 3C X GO X X X X X X X H OOhSI?! »C ut »0 00 X X X CO *•. <- V. — ^ . [ CO CO -+> © t - X 1.0 »o »o >0 »o o X X X X X X X X © CD X Oi © © t- to to © ++- © X -f *0 •— - + 4*-f + + + + + + 4-4--f + - -f 1.0 -t* © »0 CD ^ t- © 01 ^ -H ?Mh X © 0! CO »o © 3> -t* —' t- X *•+ >o i- X © ti *0 *D © Oi _ i - X X -+■ X i-O X X OJ X © © ? * 0 1 Oi -P X X © © — 01 0» CO CO -t- *D E i-Nxxr.o.r.cro-'-:if:c^*t'+-r-t'r-t' -t* -h »c *.o ».o »o io ».o uo »o »: -f ^ »C *0 1-0 *0 iO 1-0 1-0 »o »o X XXXXX X XXXXX 100 DEEP WATERWAYS. 'O D 0 O U ■e 8 sF ■ H J H &h b o pH ~ O '♦-j s 50 QJ o . o 0 J •g £ -i *o ^ zJ °1 3 3 © 1 c3 c5 c3 Cl, c« Cw Ch :wtntntdo®wo ■i rt ^ -'C'W'W Me t _(C4_i £ 54_| i —i Jr-irH QJ 'J >c£a5puPP_f2PPp ;/} © d ^ o © 1j g =e 3 +H> c« CQ © > p V §§8 3 g fl Hi ce c H P bn sS CO o Eh o £ X in W X Cl i CJl O Oi O CO O 01 O CO 01 05 a o os coc 0) >0 X b- X -. w. — — — ... 01 CSO^NHHHHO®.....-- 4~4 _ + 4~ 4- H - 4—h 4—h 4" t 4" + t 4-4* 4* co ^ a ox ci ic Cl -H »0 »0 CO i' X 00 00 XX X XXX x-x v- X N CJ OJ ,-, ,-, w v . X X X i- I- i-t— b iO c: 01 XXX 4- + + X X X X X XXX S 3 xSs§ 4- 4-4-+ 4- a? 1 - X X X X X xxxx iX'-* 05 N «5 CO DO? O 05 Ol f-H H 01 05 X -* Ol rH *- ».■ X X X X X X 05 X 05 X C5 f-H f-H X f-H 4- 4-+4- 4- 3 S£3 5 i- i-t-t— fr¬ ee XXX X 7T 4 + 4 4~ 4-4-4-4- H 01 h H lC »o i- X 05 * ?t -f i- i- i- t- i-XXX XXX X xxxx OSWEGO-MOHAWK ROUTE, LOW LEVEL—Continued. [21-foot channel.] DEEP WATERWAYS. 101 ^3 a> t3 A a o © I 50 CO o* © cS Sgb PJJ ^3 P ?l © ©iS' c3 O On X fn5 oS a <3 rH S q s a P x ^ O *3^0 iff tff r*i : £2 x P ■21! cc 5 s 0 S a © 5 s-g p C - 2 > “<5 m c5 w 2 o . © w o p. O © © 4 4-1 > b£0 ^ © g 02 © £ P © 0) r§ £ O'& £ O 2 r c *-i4-* 2 (Dr-I-I ® 3 P.c2 P Ph «OWMo S4 7T LE-i © j . © © > ^ . +-» CO r 2 ® p O rP 0 ©43 © 02 i-L- fi QC H ® r Iff $2 TO 2 § 's- Cfi lu r-.-*-< "^3 O " C ai f—i (—1 1 (A 41 f | . © S ' ©^Hf— p,2 ' a c3 c3 Oc£ i_i Vi Ui CffiS Sin d p cS*- 1 c3 oi ShSH CO 0 ) 0 ! ooooococ i r a r C' r :'Gr = '?^£r:r=r 0 „ ej ~ cS cS c2 ct ^ £- Ai ce P H ® bl O H o3 43 fe + + + + + *—I CO CO 05 CO to CD£-l-i- 05 O 05 os 05 ci c. c. X X X X X X t- © tff >■ t- -rH * 1 ? & X P X __ „ .... ?\ ++ 4 - 4 - 4 - * + + + : + _ w w f - v - ^ w S SR ^ C: O ~ — Ml-t-XXX 05 05 05 05 05 05 HH iff iff iff iff iff Oi C 05 O. 05 *0 — co ++++ ~ iff 05 CO + + *0 CO iff i- x x to iff +++++++++ + + + + + + iff >0 i-CO 1—I CO CO 05 CO iff iff — —'C5i- a a v. ^ ^ ci jo *0 Jo Z- g* Q 2^ g XXXX X XXXXXXXXX X 05 05 3s 05 3s ■+■ 4 DUXS'-ICC “ CT *—« 'C "/. 1!’t i-h 4 -' 4 - x 7+- th u" 4 ” 4 " 4 “ + AI-iCSOOXJDOl-tC X -f — CO fliANMcang^cc^s x 2J JT ( , i , i - X X XXX X 05 05 05 05 o M»J if? »0 iff iff iff iff iff iff iff iff iff Iff iff 50 5 i - T- ^ o *3 4 4 - 4 - 4 - 4 - 4 - c3 4-i A E-* cOOTCOCOCOiO^'^'-^i-H^frltbcOlOiO 05 O'.' QMOHTlIdc^H ^ - . - . - I.'I ? I a O O © to CO CO O CO ri CC tH J - 1 C O .c W N © O ic »o C H + 4-4-4-+ !4-4- + ++4-4- + 4- + + + + + + + + 4- + + + + 4-4-4- + +^j CO ~+< r-( 05 CO CO r-. -H Q t- CO l- CO CO lO © CO t— 01 -H -f <—< 05 05 05 © ^ *£ iz 1 - — — orH-xowxoi'Nori^iooc: ?! r» w wi—;•—i .. . t .4.4cc -«-+ 0 ^ o « - o n i -i-i -go xc: r. oo;oO' — ^ to to to to to •© to to to to to to to to to to to to to to to to to to to to to to to to t-t-1- ffl 44 3 V* £ a) p **£ £0 s2 >o x >o ^ + rt or n «© ’-o »o »o ©aortOH^oc^H to Ol © © »-0 © to 00 CO © e o o 3 30 .30 -3 3 c6 P3 0 cC O % i • g © *4 © JT -H 4^ r- O Oim jS; 3-2 * ” * jC H (2 o i-. H o3 P 3 y CM o A -«-> s © © a? ?-• aw > fcfi O *H © 3 A « r© O ©'-£ o ce to- 5 c 3 £ O P3 P3 _ • to © a/-- 1 y ,—1 «P43 O a « S’S 2 o^ 1 © to j2 r o *jr* -*— 1 ^ © . o^5 P H to P 0) 'P tx O rC .a © o c3 o o Sm . iw P< ’”' P a ® ^ ;-. o^< *h 2P*g P co r r V 5 c6 £ 0 C 0 4-* ^ . P** P to P .9 *3 U gi cS £ p-c ° Ph3 J «B^ >3 . u r j u v 5 S-o cc ^ gE^^H 0 O 1 1 1 I 1 1 1 1 1 1 • • 1 • 1 1 1 ■ 1 1 1 • III! ' u \ • u • i • > ' • ' • i i i • i i i 1 © © ! 1 1 1 1 I 1 » • 1 1 \ © © j ; • © • ^ i ■ ! © ^3 > J 1 1 1 1 1 ■ lll> > \ : > ! ; : > ■ y_, i ■ 1 *>■ e8 •jH | 4-3 iiii 43 l • dr 41 -lt 1 4 -» » »T* ?„ s oo 4_j *rH ^ 0 50000 © o 1 • M 0 © ' h © O O ^ © P-< aw £«♦-. P~Q PO Oo^CS •50 tO • o o © e ‘ H s ^ ^ P o§£ ° Hi 03 c5 © C *+-* gj So ce os E-»H s P H ? to I H o J4 O P 4 -J © P> r© H -m p P .2 4-3 c3 4-3 CO 5? ++++++++++ ^ o-H^tcNwc^i.':- to I — _ ~ ~f h -f T f -I- X C iC_ iff oto to tC ?'-t—t—I — <«t— X X X X ff cocccccococococococorccccccocoo a p *“5 .2 *C a3 0 o • : : p ' 1 r3 cd o ^ : ti P c3 £© cq J N:-J Hlffff ■^Xfl 30 •“ to n oi + + : : ++++ :+++++ O CQ to © * i - x -t c nd x »ff t tr •- •- «- * -1 -1 - <" I- x x x x r.r ~r'M:i:»t!t!x::xxx ccccrccococff:rofftccrocccococ , trocC'^'^ , » , t H ’^ , '^''^‘'^''^"'^'' , t''^''^'^ + + + + -4 X X X + + -+- + a M ffl QO ' a Elevations of water surface are those for a low stage of rivers, and are given immediately above and below locks unless otherwise noted. Table of distances, alignment, width of channel, sailing time, and water levels —Continued. CHAMPLAIN ROCTE-Oontinued. 104 DEE! WATERWAYS CHAMPLAIN ROUTE—Continued. DEEP WATERWAYS. 105 © P O O w SO eg A O 43 o o CO .-O 5 CO O' O **— o o -1 £■« r-\ © 5© c *?? $-1 2 Q.^ 4 04 lit © M 0) * c-c o ? 0 ffl *. © Hb* ■“ o* "c5 Its Time. . a . © o W CJ (h fa Q§ a rd ?} »o|o -®' r ’ n’O S .cc © © © be© ~ © a Q o £ £ C _ ce a a o^E-«EhEh © © * ■*3 < © i < © *-] © ( I roOOCOOCOOOOOOOOOOOCOOCCOO £ f O r C r £ r C r O r C r O r C r C r O r O r C r £ r O r C r C r C p C r C r O r O r C r £ r w cfi m o c c c r C r £ r C f O C3oS ' £ • £ ' a * £ • £ < £ — £ • £ ■ £ c3-c3^.ri £ - £ ;■ £ v, c3 ' ^ cS £ ci'T d? si ' £ x £ : *2 •» • es a £ a ! op :8 It- ' 05 01 : ^ •' i- ! -f. cc 1 -~t ~+ 01 r-“ — •—'i i Ci ' »o \ i—i ; 1 i- ! 71 ; co |OI • ?o • £4 0< 1 1 Ol ■ i i •td i i • ' 4 1 1 1 1 • • CO • 1 1 «oi • • • ■ • • ' • l • i i • • • • i i ;;;;;;; • 1 t i i • • • • i • i I o Eh a o fa 4- + + + + + + 4- + + 4- + 4* 4- 4- + + 4- + + + + 4- + 4 + + 4- 4 4 4- -I- 4 4 4- 4- 4 COOCOO'-'l-COOUftCO^^COCO—'' »-0 —'CINCXOJHCCJ-.'f ^CXC?l©i*5?©COC ci- :! c i' i • x r t r- - •- r e r: - r ^ -r u: ~ i^ r. rc >2 - ~ x v ! x • ! ^ -f -+■ »o >o i*i.:^»j:cn3Cx ->+ -r •* ci-i - x x r. c :i n^r::c - -t it i- t; r. i-Ni -i-i-i-i--i-i-»i-Nb»cox x X' xx x x x x x * esc. 3*. c: n csc: j; oci cj © 106 DEEP WATERWAYS. 33 a o O i 3 -to e 00 © a a 4-5 a o O I w Eh 1 § s « -I 5 £ <5 o* J £ & "3 <1 S K . ° 3 .Cs '-o oT o PI a A v -4-5 o o © 3 £ d © O £ ° 3 t. g W-~£ .£3 o Ci o £T o ci U& u 8*5 % O'© > &c o a o5 O a ©.a ^ a cSO? c3 CQg 3>® a '-T sJ >» u a © « o £ 73 w S o u s © a a a ,a © <*-4 O -a -+-> na £ C/2 © © © tto £. © a Q « a H © © £ . 4J CO © £ r2 O ©'43 © d a c$ 72 -*-> w 5 C/3 5 C/2 co © > rj © * "3 u r§ O O C a'tf'O'C d C/2 5$ © -U i ,_ ,_ © o^ ©®©©©©©©©©©©©©©©®©®©©©©©©© <2^3 S'^t2 , ® r ®'®^ :i, 33'C n 3'33 r S r 3'C r O , ^'33t£'®=2 r C r: 5 , C r 3'^ :,l:3r: 3 5 U • =. ci' ci ci ci' EH 1C H E -1 E - * C£> cj 1 ?3 < a ■ a «g ci 1 cS' ci' ci’T 1 ci £3 - £3 ' £3 £3 ' • £3 ■ £3 '£3 ; £3 ci' civ ci ci'' civ civ ci ' ci EhSEhc*EhEh~^Eh~5E-i3EhAEh »o o> lO X *o —< o X CO X) o 05 l- CO M O X t- CO H r—1 • CO a -*^ a a i o Eh t'« < — r-^ tO iiO CO CO 05 CO i i ' i . i i i i i COCOCOCOiiiiii ' :D05C0<0'*©'M I ' 'OXNOl.COtOCOM C! X 'f O CC »c X © -H CO *H J_ ■+" Ni-X O — O O --i' —i 0 J a i X; X X X- O O © ©; © ' O ^ ‘_' s§ © £ x*Lr i ~’a^^^ lC, 305C0 +++++iii±ii C.:!00'rOCDCXC^CO Ct'D- CC " C5 -• - '*i - .'vLAij - .cj u'j gj .j 1 ' CO X tJh 01 X OC N i — 00 O N 05 r , .OOCOO^OC3C30XClONCOCi X 01 CO. 05 05 05 lC 05 05 ©CCi-XO-_.COCOCOX ' I r-. X - X X X 1 C O 01 'C - C CO »C »0 »0 05 0! *0 05 CO ^ g^2§S£g3^^232o*^2Zt^?f2^- :: t 0 6ol/^3rtSSiS52?iriSSr53 4 ^i5 CHAMPLAIN ROUTE—Continued. DEEP WATERWAYS 107 nd 0 ) 3 o O =0 5^ "2 S4 24 -c o P 43 © -U 3 44 v o o ~ c5 O 2,-0 §*0 ll O © Jh © p-g 5 8 O £ a> P J3 3 *3 p CQ g o © g 3 43 o o o . © 03 a © g o 43 O o 43 © O Qrnn s rfi O O O © O O © —IfHT- .©r—I «+-( P — P ac 21 CO o o o o o r c r P r p'c r P' 5 O O O O O O 3 rp n$ *3 T3 r3 'P r O 0 O O O O O O O G P'P'PJ'P'P r: r © © © tl'o *■« © p Q © p • a p 50 f3 p p . d . p "P+P.PPi^P;^ c8 P ^ P w P i p ’-45 p p T 1 _p Pi., P7 - p © © Bx © P 4^0 ©'£ © p gS CO CO © > p O o H P o &H . 2 ? s 26 P P ?S3 H«C?rJ2i!300 1 35 --2 i- t- 2i O: 21 — 2? N ^ OS W rH JJ cr © ^ H N P O X rH ? > -H i - i - p. 2! pi —1 JO lO '?tPPH?lilpLOX'+p5r-(-iXp4cOWXOrH'f NX2.2 + 4- +++++++ : + + +++++++ + + + + + +++++ + +++++++ c: 22 -f i- 22 t- j- to t—< :c t- 21 :o r: :p to- 2 i 05 ci x> t- to 22 t- -+< -+• co oi t- i-h c? x o: o o cr. ct r. ip p r. 3i •.-: i - :c *r p p. p ^ - p i - r. p ^ -p x r. >: i - r. ^ 4* x — r. c r. c © © t- I - X M X P 2. 53 C O 21 21 2* 22 CO 2* CO ££ 22 -H -* ^ «2 »P »2 t- <- J- ©2' CP 3 — - rirHHHHHr-irHHM5lWN2lW7l2l25N 2{ 21 21 21 21 2» 2! 21 21 21 2i 21 2i 2 I 22 22 22 22 + + - + + . CO i- i — 2* 05 21 05 2 + + + + -+ :+++++++++++++ ++++++++- C 2 ^2 4* t - r- r: i - i - 2 -i 2 i - 2 ► 2 2. 2 2 -H -H 21 P 2 X N 2 C2 I--rHp X2 C *2 2. ?2 2 >C 2 2. 3! i2 i' P 2 2 CP. P> 22 ~ - P. P. 22 25 X C1 *1 - i > X 05 _ C .-h ^ r-1 r-H r-i , —H —. 21 21 21 21 2! 2> 21 21 21 2* M 21 2i 21 21 21 21 21 21 21 2t 2i 21 22 22 22 (gu; Table of distances, alignment, width of channel, sailing time, and water levels —Continued. * CHAMPLAIN ROUTE—Continued. [21-foot channel.] 108 DEEP WATERWAYS. cs o r £ a A ; A c ~ ; wc ? c5 Es fc. o 2 3 5 i .2 p. 1 ' i c ... « & o o *5 M i a, T. C — ooofoccooooo g-cl-TI-rl® .-s 1 CO ® O O O^C^C^C O C O O C C 0 ti'M r" 0®o ^ cfi^iwCwCw c^^Cw 7i ' 7i 1 ci c3 c5' d 1 c3 0 > Jh 6 ^ ; A Z • cc C +2 a2 £ cS -M Cfc rf, Cl CC —I—h—I—f~ -f—-j- ~+—f—I—I—f—I—I—I—1—I—h—I—I—I—I—!—I—h—I—h—f - - + + + + 0MfO-^«rH*XiQ®Na^NCOcOO:C:C:'+CO^QOQr^CC0QH»r Cl -+■ »C i - CT. ~r ~ C 1 CC C * -e X> CC r— -H X Cl C* CC —< Cl CC CC X CC CC *-C X C“ — ^ 5{ ~ ,+ x .^ — — 1-1— 1-1— i-i- x X X X£»03 0-.i;w«-r-H -CCCCCCCCCCCCCCCCCCCCCCCCC0CCCCCCCCCCCCCCC0CC'^ l ' ! +''rt1^'^ , '^ Cl C’-t- — C C! CC X X — — i - 1 - X — IT l- JC -+• cr CC CC C f ---t- CC •' X ~ CC -4. i- »^c — I- cr. X ‘C — C X i- C ”f i-C C. —~ •—< —r Cl 1 — 1 — CC >.C CC i— 1 — X X 1- l— i~!©CC-fi.CCCCC — 1- 1.C + + + + 4--f + -h+-f — ^-4- + -f4--r + + 4- + + 4- + + + +-)- + + 4- + 4-"i" + + -|- ic pClCO-^Xn C c C. N C. C N X iC C. C. c: -t CC C o X-hM ^C-h X-i C — C C X X -4 ci -t tc i - c. c r. :> c ci c x r-»• x ' - - r - es c c x cc cc -c x c - c c i - C D *t --— — — C * C l CC — C* — -*■ »C *C * C «C •— *C 1- 1 - 1 - 1 - t- i - X XX XC.C.CCCr-HH CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC CC -r -r ^ ^ CHAMPLAIN ROUTE—Continued. DEEP WATERWAYS. 109 V 3 2 O O s TS 3 © p c3 P3 V 4-3 o o £ r-3 .Si *"-o &T o5 © o s o (J i = 3 > --t> ® 5 g W33 * jg W B s o !» fc Ql g .2 p,x P. cs,3 S* o v p Ro & Sb j p) ~t 5 2 s 5 © o ft ci ft ft -ft .jg f S jr ft o o 6 PJ ce ^ 6' c3 O ft ft oS rP v a . o O P H Q« . . ^ O OH H o gPP-g |Q ft o 5 £ h2-3 s- ^ i. ® o||!o o|o o o|o •S'® a<2 c'O r a<2'r' a '° q-o cS ,-P* P V2Z* r Jl © © © C--4-. > be o ^ © a P O 2 pi p p cs ce 7i a rJl c3 p © © £ . -*J CQ © P rP O ©Xa © c3 p -te -» 71 £ > Sh o • ® p-tf c« P Eh © tx 5 Vl ' S • S S i ®,§ 3^ =c X?5 •3 C8 CIS- ' ® o o o o a a - „ ce cs^ cs & ® o o B'a'a j q - H ^ 58 55 c« o Eh **- 55 tH ?i«i2ii522 L ?‘ l Q l o?i^-HXccMWO?ic©©w^^-H^^MX05©*occxNo s o u fo rH ^ 4-4- <© 3 b c c © . B O k - © 5 B © s|5 B S c3 (5 g p. cp *S «s © S3 B B ^3 O rB © 0/ © P=^ > be o P © £ Q © B © © j! t: © £ rO O P'S £ O i © tuo X o rB ,B © o B o3 O . O P — P ft^ ft ft ft |ol P. 0 o &JJ JS o B O 33 EL 6 & p d^ £< o . ®Q^ £p & ft v > o ft o O f. P PP fn « B O P ft B ■♦? 1 • • 1 eg ^OOO 0 + £ ' :jS X o o B 'BP B ' B ' B B ^ B Eh£hH^EhEh + H S «c iO C* CO - $0000000 ^na-C'C-O-O'O'O B 1 B • B 1 B :o ft cSi. b^ S' Sr 1 ? 1 00000000000 © «g r d'o r dT3'd f B r d f d r d'or! © £ 1 £ I £ *0 £ I £ •££ B ■*• B ^ B ' B ' BB S? 55 r-* + 15 *b *1 £* »o *-h x co r> ci ft —«—- xi cc ft ft o> ft ft 1 - o -+• — o* cc t- ft 01 *-h x »o -+* ft ft ©®W©»CH»3o5)aif50HX®H*5l»OOl'«W , ^i-'«OC5rtMCONHOW©»?5 ++++++ +++++++++++++++++++++++++++++++ C2‘2^HQO»-HO$^O^M ft ft O* — l-» X Ol 01 010101 o 1 ci ci cc cc cc —t* '-+* -+ 1 »o »o »c ».o * ‘ O i.C lC ».o »o ».o iC iC O O uO iC lO iC iC iC lC u'j OO X CO00CSO5CJ_ S "B*-5 *-x ji ic h x x 01 c C.OOBtOXCH»COXX »." 4- + + + 4- + + + + 4-4-4--1 fo OJ r-H O CO ft + + + 4-4- + + + + + CD CO WJ '-+ X ^-1 ft ft c: X 01 ft cc o» — t- X o> co «o CO CO ft CD CO CO Q »o 00 CO 03 ^ W Q —* -+ ft ft 01 0» ft x o: r i ro -* ci co co ft ft ft ft —< 01 + -H ft i- co ft t- ft »o *o ft 01 -* coxxccxftftftfto 1 r r ~ ~-—. — — ci 01 o* oi 0^ oi ci o* c^ co co -t- -+• -# uo >0 »o ^ rt r)< 7t •?}• Tt rt L* LC 1C 1 .* c LC »C C LC 1.C c »C lC O iC c c C C iO O C iC lO LC c o - + + + + + + - DEEP WATERWAYS 111 S3 S3 O Q e CO T3 © P a o O I w p £ t—t < fc 3 <) ffi o &> .so ^>o o £ © g S a3 © C c p c3 <*-i Qf rP O ■S « §? Strl P CO C G *_£ G a s -4—1 W CO ? ° ® i ® © 5^3® Eg-2 £ --S a c5 o £ *8 Q§ 4^ © g 3 © «W o Si -4-S 'd £ © © © u'S 1 © a Q © c8 a o © o o o o t ^ r d r d r d r d • co a -*-> a P Eh © ^ tUD 05 t + + iff »o 5 S3 + + CC 71 1— co ■*+• o x i-i- iCOO CO H CO T O ‘ O 05 l - CO r-H + + + + + S? a © a * a to 5° q_, CO' o * a ,a:a o ■H Cj ^ 3-SAI s-s& a ^ a © a 5 ^ c a CO © r*-< -*-> a -h* a cs ffiCQ c3 fSo A! O t* £ o £ c r—I a *s m 112 DEEP WATERWAYS Estimated cost of excavation and construction. 21-FOOT CHANNEL. Duluth to tide water. Chicago to tide water. t t Lasalle-Lewis- tonCham- ton-Oswego- nlain route Mohawk low- plam i oute. Ievel route . Lasalle-Lewis- ton-Cham- plain route. Lasalle-Lewis ton-Oswego- Mohawk low- level route. Excavation. Retaining walls, slope walls, and back fill... . Embankment .... .. Railroad and highway changes. Right of way..... Bridges and ferries.. By-passes..... $128,368,611 6,544,621 1,853,154 707,443 5,551,765 4,296,481 43,385 20,396,732 1,884,702 1,800,000 $105,100,457 10,822,948 155,014 1,560,054 12,478, 442 5,392,010 $123,372,812 5,544,621 1,853.154 707,443 5,551,765 4,296,481 43,385 20.396, 732 1,884,702 1,800,000 $100,104,638 10,822,948 155,014 1,560,054 12,478,442 5,392,010 Locks..... Cribs at lock approaches... Lock-operating plant. Breakwater.... 36,328.417 5,629,389 3,200,000 721,380 64,371 206,413 3,122, 610 796,923 2,062,295 30,000 36,328.417 5,629.389 3,200,000 721,380 64,371 206.413 3,122,610 796,923 2,062,295 30,090 Entrance of streams...... Dikes.. Dams.. Regulati ng works... W a ter supply.. 225,096 206,413 551,819 1,687,167 225,096 206,413 551,819 1,687,167 Buffalo waterworks tunnel. Engineering, superintendence, and contingencies, 10 per cent a. Total... .. 30,000 30,000 173,147,389 17,235,047 187,670,723 18,687,380 168,151,590 16,735,467 182,674,924 18,187,800 190,382,436 b 206,358,103 184,887,057 c200,862, 724 30-FOOT CHANNEL. Excavation. Retaining walls, slope walls, and back fill... Embankment. Railroad and highway changes_ Right of way.. Bridges and ferries.. By-passes.... $231,874,094 7,294,119 1,793,961 707,443 5,576,765 4,6:35,724 52,115 32,114,558 2,422,340 1,900,000 $178,970,590 15,567,970 157,015 1,560,054 12,494,442 5,947,106 $217,601,194 7,294,119 1,793,961 707,413 5,551,765 4,635, 724 52,115 31,191,065 2,004,350 1,800,000 $164,697,690 15,567,970 157,015 1,560,054 12,469,442 5,947,106 Locks. Cribs at lock approaches.. Lock-operating plant. Breakwater. . . 56,975,392 6,067,266 3,300,000 1,190,317 64,371 206,413 3,122,610 796,923 2,062,295 30,000 56,051.899 5.649,276 3,200,000 1,190,317 64.371 206,413 3,122.610 796.923 2,0:>2,295 30,000 Entrance of streams. Dikes... Dams.. . Regulating works.. . Water supply. 225,096 206,413 551,819 1,687,167 225,096 206,413 551,819 1,687,167 Buffalo waterworks tunnel. Engineering, superintendence, and contingencies 10 per cent a . Total.... 30.000 30,000 291,071,614 29,027,469 288, 512,764 28,771,584 275.332,231 27,453,531 272,773,381 27,197.646 320,099,083 6317,284,348 302,785,762 c 299,971,027 a Estimate for Lake Erie regulating works already contains cost of engineering, superin¬ tendence, and contingencies. b See note, p. 113. cSee note, p. 114. DEEP WATERWAYS 113 Estimated cost of divisions. LAKES MICHIGAN AND SUPERIOR TO TIDE WATER. 30-foot chan¬ nel. 21-foot chan¬ nel. Lake Superior to Lake Erie: 1. With regulated surface ._. $33,539,809 41,554,104 10,220,548 23,287,911 73,435,350 75,084,453 75,572,250 77,221,353 $0,961,818 9,123,923 1,466,439 2,979,072 42,393,203 43,214,344 48,453,753 49,274,894 2. Without regulated surface .. Lake Michigan to Lake Erie: 1. With regulated surface . .. 2. Without regulated surface . Lake Erie to Lake Ontario: Via Lasalle-Lewiston route— 1. Witn regulated surface ... ._ . . _ 2. Without regulated surface._. Via L’o.iawanda-Oicott route— 1. With regulated surface .... .. 2. Without regulated surface..... Lake Ontario to tide water: Via Oswego-Mohawk route (high-level plan)— 1. Western division a._.. 83,395,975 112,474,lln 10,383,408 05,076,050 85,488,414 4,159,904 2. Eastern division.... 3. Hudson River division. Total a . Via Oswego-Mohawk route (low-level plan)— 1. Western division a..... 2. Eastern division......- 3. Hudson River division... 206,253,553 155,324,968 87,451,551 112,474.110 10,383,408 07,354,704 85,488,414 4,159,904 Total a . ........ 210,309,129 157,003,082 Via Champlain route (with movable dams)— 1. St. Lawrence division ..-. 35,514,438 64,008,981 103,210,977 10,383,408 21,027,113 47,912,078 67,927,720 4,159,904 2. Northern division... 3. Hudson River division, upper ..... 4. Hudson River division,tidal.. Total.. 213,123,804 141,027,415 Summary of cost of divisions. DULUTH TO TIDE WATER. 30-foot chan¬ nel. 21 -foot chan¬ nel. Via Lasalle-Lewistoti (regulated surface): Oswego-Mohawk (high-level plan)— 1. Lake Superior to Lake Erie... $33,539,869 73,435,350 206,253,553 $6,961,818 42,393,203 155,324.968 2. Lake Erie to Lake Ontario. 3. Lake Ontario to tide water a__... Total a ........ 313,228,772 | 204,679,989 Oswego- Mohawk (low-level plan)— 1. Lake Superior to Lake Erie... 33,539,869 73,435,350 210,309.129 6,961.818 42,393,203 157,003,082 2. Lake Erie to Lake Ontario...... 3. Lake Ontario to tide water a _____ Total ci ______ 317,284,348 206,358,103 Champlain (with movable dams)— 1. Lake Superior to Lake Erie. 33,539,869 73,435,350 213,123,8*54 6,961,818 42,393,208 141,027,415 2. Lake Erie to Lake Ontario.... 3. Lake Ontario to tide water. Total....... 320,099,083 190,382,436 a Tlio estimated cost of the Oswego-Mohawk route is based on the use of swing or bascule bridges at all railroad crossings. If an overhead railroad crossing be adopted for the New York Central Railroad near Utica, with a fixed span over the waterway at a clear height of 85 feet above the water's surface, the estimated cost of the route would be increased $788,000 for the low-level project and $953,000 for the high-level project. With this modification, the total esti¬ mated cost of the 21-foot waterway would be $207,140,103 for the low-level project and $205,032,989 for the high-level project. For the 30-foot waterway, the corresponding amounts would be $318,072,348 and $314,181,772. II. Doc. 149- •8 114 DEEP WATERWAYS. Summary of cost of divisions —Continued. CHICAGO TO TIDE WATER. 30-foot chan¬ nel. 21 -foot chan¬ nel. Via Lasalle-Lewiston (regulated surface): Oswego Mohawk (high-level plan)— 1 Lake Michigan tol.ake Erie ... . $16,226,548 73,485,350 206,253,553 $1,466,439 42,393,203 155,324,968 2. Lake Erie to Lake Ontario...... 3. Lake Ontario to tide water ci _-__ 295,915,451 199,184,610 Oswego-Moliawk (low-level plan)— 1 Lake Michigan to Lake Erie .. .... 16,226,548 73,4:35,350 210,309,129 1,466,439 42,393,203 157,003,082 2. Lake Erie to Lake Ontario____ 3. Lake Ontario to tide water a ... Total ct __________ 299,971,027 200,862,724 Champlain (with movable dams)— 1. Lake Michigan to Lake Erie _____ 16,226,548 73,435,350 213,123,864 1,466,439 42,393,203 141,027,415 2. Lake Erie to Lake Ontario.-.... 3. Lake Ontario to tide water..... Total_____ 302,785,762 184,887,057 a The estimated cost of the Oswego-Mohawk route is based on the use of swing or bascule bridges at all railroad crossings. If an overhead railroad crossing be adopted for the New York Central Railroad near Utica, with a fixed span over the waterway at a clear height of 85 feet above the water’s surface, the estimated cost of the route would be increased $788,000 for the low-level project and $953,000 for the high-level project. With this modification, the total esti¬ mated cost of the 21-foot waterway would be $201,650,724 for the low-level project and $200,137,610 for the high-level project. For the 30-foot waterway the corresponding amounts would be $300,759,027 and $296,868,451. THE RELATIVE ADVANTAGES OE THE 21 AND 30 FOOT WATERWAYS. In the preceding divisions of this report the various routes surveyed and investigated by the Board for waterways of different depths from the Great Lakes to the Atlantic tide waters have been described and discussed from a physical point of view and estimates of their cost of construction have been given. The Board now submits a statement of the relative advantages of the 21 and 30 foot waterways, as required by the provisions of the sundry civil act of July 1, 1808. The benefits to be derived from a waterway from the lakes to the seaboard result directly from low rates and ample facilities for the transportation of freight between the terminals and from the develop¬ ment of new industries and new commerce. At the commencement of the investigation of this subject an effort was made to collect sta¬ tistics showing the effect of better transportation facilities on the com¬ merce of the country, with especial reference to determining whether the indirect benefits to the producers of the country would justify large expenditures in the increase of transportation facilities in excess of the actual requirements for the lake traffic. It was found, how¬ ever, that a satisfactory solution of this problem would require more time and money than was at the disposal of the Board, and that the results, when obtained, would not be sufficiently conclusive to war¬ rant making them a basis for estimates. Statistics of lake and ocean commerce are collected, arranged, and discussed in the publications of the Bureau of Statistics of the Treasury Department, and from these publications all the information can be obtained which can be DEEP WATERWAYS. 115 employed to advantage in the investigation of the problem of relative advantages submitted by Congress to the Board. This problem is of a speculative character, since it involves the consideration of condi¬ tions which will exist after a deep waterway has been established— conditions which will be very different from those existing at the present time. The Board therefore considers it useless to seek for great accuracy and detail in the statement of existing commercial conditions. An exhaustive study of the relative advantages of routes of differ¬ ent depths has been made by Lieut. Col. C. W. Raymond, member of the Board (Appendix No. 5), from which it appears that the benefits to be derived from the respective waterways are so well defined that fur¬ ther statistical inquiry would add but little to the force of the conclu¬ sions. This investigation was made by Colonel Raymond in consul¬ tation with the other members of the Board, and his report full}’ expresses their views on this subject. The details of the investiga¬ tion are given in Appendix No. 5. In this report it is only necessary to give an outline of the method of investigation followed, and a summary of the conclusions at which the Board has arrived. A waterway connecting the Great Lakes with the sea maybe regarded as an instrument of commerce for the purpose of increasing the value of merchandise by transporting it from one point to another. Waterways of different dimensions and cost may be compared from this point of view, and their relative direct advantages as compared with each other, and with the amounts required for their construction may be approximately determined. Such comparisons, however, are too narrow and limited for a satisfactory solution of the question of relative advantages. Indirect benefits may result from the establish¬ ment of a great transportation line, which maybe of such importance in their influence upon production, commerce, and the general pros¬ perity of the people that the question of a greater or less return of direct value may become comparatively insignificant. Accordingly, the investigation of the question of relative advantages has been con¬ sidered under two heads, viz, relative direct advantages, and relative indirect advantages. In determining the relative direct advantages of the waterways com¬ pared, an attempt has been made to express the relations existing between the various elements entering the problem in a mathematical formula from which may be obtained numerical quantities represent- ing approximately the relative values of the waterways considered. These elements are the traffic capacities of the waterways, the costs of construction and maintenance, and the cost of transport proper in the most economical type-carriers. The board believes this method of stating and discussing the problem has the great advantage of exhibiting clearly the elements upon which it depends and the uncer¬ tainties unavoidable in such investigations. 116 DEEP WATERWAYS. The determination of relative direct benefit has been limited to the consideration of through traffic conducted in the most economical car¬ riers. The waterways extend from Duluth to New York and from Chicago to New York. For each route these lines of through traffic are considered separately, both for foreign and domestic traffic. The routes from Duluth and Chicago to New York which are com¬ pared are as follows: 1. Thirty-foot waterway via La Salle, Lewiston, St. Lawrence River, and Lake Champlain. 2. Thirty-foot waterway via La Salle, Lewiston, and the Mohawk Valley, high-level plan. 3. Thirty-foot waterway via La Salle, Lewiston, and the Mohawk Valley, low-level plan. 4. Twenty-one-foot waterway via La Salle, Lewiston, St. Lawrence River, and Lake Champlain. 5. Twenty-one-foot waterway via Lasalle, Lewiston, and the Mohawk Valley, high-level plan. G. Twenty-one-foot waterway via Lasalle, Lewiston, and the Mohawk Valley, low-level plan. The traffic is supposed to consist of the movement of bulky freight, such as grain, coal, lumber, and ores, to domestic and coast markets or to markets beyond the sea. For each depth of waterway a tj^pe carrier is adopted which is believed to furnish the most economical transportation. The data for the comparisons are determined prin¬ cipally from the estimates of the Board. The formula gives for each waterway a measure of its relative annual return of value upon each $100 expended in construction, after the payment of the costs of maintenance, operation, and transport proper, as compared with a standard waterway for which the return of value is assumed as unity. The results obtained by this method of investigation are briefly as follows: 1. The return of direct benefit from the 30-foot waterway via the St. Lawrence River and Lake Champlain is less for both foreign and domestic traffic than the return from either of the Mohawk Valley routes. The two Mohawk Valley waterways give practically the same returns. 2. All the 21-foot waterways give practically the same return of direct benefit. 3. The return of direct benefit from the 21-foot waterway is much greater than the return from the 30-foot waterway. In order to form an estimate of the relative indirect advantages of the waterways compared it was found necessary to consider the amount and character of the existing lake traffic and its past and probable future development, to point out the distinguishing peculiarities of transportation lines of different character and capacity, and to indi- DEEP WATERWAYS. 117 cate the objects which the proposed waterways are intended to sub¬ serve. The following condensed discussion of these subjects is taken from Appendix No. 5. THE HAKE TRAFFIC. The demand for increased facilities and diminished rates of trans¬ portation from the region of the Great Lakes to the interior of the country and to the sea is based upon facts which are believed to be established by the history of the development of the productive resources of this part of our territory. The commodities forming the bulk of the traffic for which provision is desired are grain (including flour), iron ore, lumber, coal, and manufactured products. The movement of the four leading commodities above mentioned comprises about 90 per cent of the total freight movement on the lakes. As will be shown hereafter, the greater part of this traffic goes to the domestic markets of our country, but still an important part is destined to foreign markets. The volume of these products has increased rapidly with every increase in the facilities of transpor¬ tation and with every permanent decrease in transportation rates. It is claimed that further increase in facilities and reduction in rates is absolutely necessary if we would hold our place in foreign markets in competition with the products of other countries. The following table 1 shows for the year 1898 the traffic for each of the four leading commodities referred to above, the eastward traffic (that is, the traffic east of Detroit), and the quantities destined to domestic and foreign markets, respectively: Commodity. Total traffic. Eastward traffic. Total. Domestic. Export. Grain (including flour). Iron ores.-. Net tons. 12,086,013 13,650,788 4,540,000 8,722,667 Net tons. 12,036,013 11,028,321 2,531.180 Net tons. 2.888,829 11,028,321 2,531,180 Net tons, a 9,147,184 Lumber -...__ .. Coal..... Total,____ 38,949,468 25,595,514 16.448,330 9.147, 184 a Exports from Montreal, Boston, New York, Philadelphia, and Baltimore. To indicate the magnitude of the past development of this com¬ merce, it is only necessary to say that the total lake traffic for the year 1871 has been estimated at 14,283,000 tons. Since that time transportation facilities by rail and water have been greatly increased, new locks around the falls of St. Marys River have been constructed, the Welland Canal has been deepened, the lake harbors and channels have been improved, steam vessels have taken the place of sailing ! The figures for grain (including flour) are compiled from a report entitled “The Grain Trade of the United States.” published by the Bureau of Statistics of the Treasury Department. January, 1900. The other figures are based upon data obtained irom the admirable tables which accompany the report of the committee on canals of New York State, 1899. 118 DEEP WATERWAYS. vessels, and the population of the country has about doubled. These are the principal causes of this enormous expansion of the volume of traffic. FUTURE DEVELOPMENT OF LAKE TRAFFIC. The population of the country will surely continue to increase rap¬ idly, and this must be accompanied by an increase in the volume of the lake traffic. It must not, however, be inferred that the eastward traffic will develop in direct proportion to the increase in population of the country, for about one-half of our population is situated in the great Mississippi basin, where the rate of increase is much greater than in our Eastern territory. The future demands of this part of our country upon the products of the lake region will doubtless reduce the relative amount of Eastern traffic. Nevertheless it does not seem unreasonable to believe that the ratio of demand to supply will con¬ tinue to be as great as it is at the present time even should the facili¬ ties for transportation be very largely increased. It appears from the table given above that only about one-third of the east-bound lake freight is exported to foreign countries, the remain¬ der being distributed to domestic markets. Practically the entire exports of commodities transported on the lakes and received from the lake region consists of grain and other food products. GRAIN. As regards the future development of the production of grain in the region tributary to the lakes, it is only necessary to point out that the rapid increase of our population will imperatively demand the utiliza¬ tion of all our food-producing areas in the near future for the supply of our own markets. It has been stated In Hon. John Hyde, Chief Statistician of the Agricultural Department, that within the short period of thirty years more than the entire wheat production of the country will be required for consumption by our own people, to the entire exclusion of our export trade. 1 Even should this view not be accepted by all, it must be admitted that the ratio of the export trade to the domestic trade in food products must rapidly diminish. IRON ORE. The movement of iron ore, which forms at the present time so large a proportion of the lake traffic, is principally from Lake Superior to Lake Erie ports, from which the ore is sent by rail to the great coal and iron region of which Pittsburg is the center. As the undeveloped resources of the Lake Superior region are enormous, this traffic may increase greatly under the demands resulting from increased popula¬ tion. Should adequate facilities for water transportation be provided, ’“America and the wheat problem.” Published in The Wheat Problem, by Sir William Crookes, F. R. S. DEEP WATERWAYS. 119 it is possible that a considerable part of these products may be carried to points within the interior of the State of New York, where conven¬ ient limestone and the saving in cost of transportation both of the crude material and finished product, may compensate for the advan¬ tage of the Pittsburg district in its greater proximit y to coke and coal. 1 None of this ore is exported at the present time, nor is it probable that much of it ever will be except in the form of finished material. LUMBER. Of the four leading commodities considered, lumber forms the smallest proportion of the lake traffic, and its movement is rapidly diminishing. The reasons for this rapid decrease are fully and clearly stated by Prof. George G. Tunnell in his able report on lake com¬ merce. 2 It is largely due to the destruction of the forests on the shores of the lakes and on the banks of the tributary streams. Lum¬ ber is now principally obtained at points so far in the interior that it is generally cheaper to saw logs at local mills and transport the product by rail than to carry or float them to the water and transship them. Moreover, there is a strong and increasing competition in Northern markets from Southern lumber. The exports of lumber from the lake region are now insignificant, and they must cease in the near future, as much more than our entire product will soon be needed for our own people. COAL. The total volume of eastward traffic on the lakes greatly exceeds that of the westward traffic. The lake movement of coal, which is entirely westward, is therefore of great importance, not only because it supplies the necessities of the territory west and north of Lakes Michigan and Superior, but also because it furnishes a return freight for the lake carriers. Professor Tunnell states that during 1896 coal constituted about three-fourths of the westbound traffic through the Detroit River and 86 per cent of the westbound traffic through the St. Marys Falls Canal. Most of this material is shipped from the ports of Lake Erie to Duluth and Superior, at the head of Lake Superior, and to Chicago and Milwaukee, at the head of Lake Michigan, the shipments to Lake Superior being much greater than those to Lake Michigan, as in the latter case the conditions are more favorable for railway competition. At the present time none of the coal transported on the lakes is sent to markets beyond sea, but if a deep waterway to the seacoast were constructed it would probably become an important factor in our export traffic. 1 Report of Committee on Canals of New York State, 1899, p. 15. -Document No. 377, House of Representatives, Fifty-fifth Congress, second session. 120 DEEP WATERWAYS. SUMMARY. To summarize the above statements, the freight traffic of the Great Lakes, already amounting to at least 40,000,000 tons per year, 1 may be expected to increase greatly and rapidly with increase of popula¬ tion and the extension and cheapening of facilities for transportation, but this traffic will tend more and more to domestic markets and less and less to foreign ones. CHARACTERISTICS OF TRANSPORTATION LINES. These conditions appear to fully justify the establishment of new facilities for transportation from the lakes to the sea either by the General Government or by State or private enterprise. At the pres¬ ent time by far the greater part of the traffic between lake and ocean is by railway, only about one twenty-fifth of the volume transported going by canal and river. If a new line for water transportation is to be established it must be done by the General or State govern¬ ments, not only on account of the great expenditure involved, but also because such a line is not so desirable for private ownership and operation as a railway upon which the carrier business can be monopo¬ lized by the owner, and therefore it probably would not be constructed by private enterprise. In order that the consequences involved in the proposed change of the greater part of the traffic from rail to water transportation maybe clearly understood, the principal charac¬ teristics of railways and waterways considered as instruments of commerce for the transportation of freight must now be pointed out. It is frequently asserted that water transportation is always much cheaper than transportation by rail; but this statement can not be accepted without qualification. If it is intended to mean that the cost of transport proper is generally less in the case of the waterway than in the case of the railway, the statement is doubtless true; but if the toll is included in the cost of transport for the waterway as well as for the railway, the cost of transportation will be often less for the railway than for the waterway, when the latter is an artificial channel of mod¬ erate dimensions. As a line of communication between the same terminals, the railway is, for obvious reasons, almost always shorter than the water line. Moreover, it carries passengers, and a considerable part of its freight is of large value in proportion to its bulk. The passengers and high- class freight are made to bear a large proportion of the mean cost of transportation. A distinguished authority 2 on this subject finds from 1 The registered tonnage of the lake traffic for 1898, as given in the report of the New York State committee on canals, is 62,023,000. A large percentage of this is the registered tonnage of passenger steamers. • C. Colson, tnghiieur des Ponts et Chaussees, Maitre des Requetes au Conseil d'Etat. Transports et Tarifs, Paris, 1890. DEEP WATERWAYS. 121 a study of experience on French railways and waterways that between two given points the mean net cost of transportation by rail is gen¬ erally lower than the cost of transportation by water; but it must be remembered that the canals of France are of small dimensions and not well adapted to economical traffic. In short, no general rule on this subject can be laid down. Each case must be separately investi¬ gated and the relative economical advantages of the rail and waterway must be determined in accordance with the existing special conditions. Even then it is not easy to make a satisfactory comparison, owing to characteristic differences in the methods of conducting transportation by the two lines. Generally the railroad carries passengers and a great variety of high-class as well as low-class freight, so that it is exceedingly difficult to determine the average cost of transportation of any assumed freight unit. One of the most important differences between the railway and the waterway arises from the fact that in the case of the former the pro¬ prietor of the line and depots for receiving and shipping freight and the carrier are one and the same party, while in the case of the latter these interests are generally in different hands. It results from this that railway service is much more regular and efficient than water service, because it is under a centralized management. The skill and efficiency with which the railway service is managed and improved and the lack of improvement and efficient management in canal transportation have often been pointed out, but it does not seem to have been observed that these differences are largely inherent in the different character of the organizations of the two services. The management of the railway is as much interested in the shipping, re¬ ceiving, and movement of the traffic as in the toll, while in the case of the waterway each interest is concerned with the others only so far as may appear to be for its own direct benefit. In the case of waterways of small dimensions delays are more lia¬ ble to occur from accidents and crowding than in the case of the rail¬ way. The railway, when compared with the small waterway, has gen¬ erally the great advantage of speed, which secures for it all the traffic in which time of transport is an element of importance. Finally, the railway is available for traffic during the whole year, while the waterway must be closed during the season of ice. M. Colson remarks that experience shows that generally these advantages of the railway cause it to be preferred for merchandise of moderate value when the rates do not exceed those of water trans¬ portation by more than 20 per cent. 1 This deduction, however, is doubtless based upon a study of the traffic upon the railways and small canals of France. The net cost of transportation upon the waterways herein consid¬ ered, for both domestic and foreign traffic, would of course be very 1 Tarifs et Transports, p. 310. 122 DEEP WATERWAYS. much smaller than on a railway or combined lake and railway line, even should the toll be included. Moreover, it is important that the facilities provided for increased traffic movement should be fully ade¬ quate to meet all possible future demands, and the waterways have a traffic capacity exceeding that which could be furnished by railways at the same cost. The total freight tonnage of the New York Central and Hudson River Railroad in 1S98 was 23,403,439 tons, which is much less than the maximum traffic capacity of the 21-foot waterway. An important advantage of the waterway over the railway results from the characteristic feature of its organization which has been already pointed out—that the various interests of line manager, freight shipper and receiver, and carrier are in different and inde¬ pendent hands. The maximum amount of benefit is derived from the traffic by the users of the line (or general public) when the toll and transport proper are made as small as possible. In the case of a railway, where the entire system is controlled by a single manage¬ ment, the natural effort is to obtain for the proprietor and carrier as much as possible of the value derived from the traffic—in other words, to make the traffic pay what it will bear. In the case of a large water¬ way open to the use of all carriers, the element of free competition regulates the rate of transport proper, and under the circumstances the charge for transportation must tend to approximate the net cost. But it is not merely from the reduction of rates that benefit is derived. One of the most injurious effects of the lack of free com¬ petition in railway traffic has been the variation of rates through a wide range, resulting from alternate competition and combination of transportation lines. It has been found difficult, if not impossible, to control these variations by law, but the influence of a large water¬ way, open to the use of all carriers, could not fail to prevent large fluctuations in railway charges upon bulky freight during the season of its operation. It has already been pointed out that this characteristic feature of waterways is a disadvantage so far as regards regularity of service and efficiency of management, and this is one reason why it may be considered desirable for the Government to own and manage the waterway and assume the toll. Under these circumstances the pub¬ lic will receive all the benefit derived from the traffic, after the car¬ rier has been paid his charges, and these charges will be kept from large fluctuation and near the net cost of transport by the action of free competition. It would, at first sight, seem unfair for the Govern¬ ment to assume the toll on one transportation line to enable it to com¬ pete to advantage with other lines constructed and operated by its own citizens, but it is claimed that the increased demand for a higher class of freight created by the business and prosperity which would inevitably follow the construction of a great waterway would more than compensate the railways for the loss of the low-class traffic. It DEEP WATERWAYS. 123 would not be to the public interest to have the high-class traffic diverted from the railways to the waterways, but high-class freight is package freight, not readily handled by mechanical devices, and therefore not likely to go by water. This characteristic feature of water transportation controls not only the movement of the freight, but also its shipment and delivery. In the ease of the railway, stations are established at which freight must be handled under the direction of the management. In the case of the waterway, every point upon its banks is a possible station. The result must be an active competition, which must control and cheapen the cost of handling and develop points of shipment and delivery best suited to economical receipt and distribution. It is claimed as a great advantage of waterways of sufficient dimen¬ sions for navigation by ships that they permit of the transport of the cargo through to domestic or foreign ports without transfer from one carrier to another, thus saving the time and cost of handling and loss by waste. This is an advantage of the ship canal as compared with the barge canal of moderate dimensions as well as with the railway. It is, however, considered by high authorities very doubtful whether a vessel can be so constructed as to navigate successfully and econom¬ ically the ocean, the lakes, and the canal. The ocean vessel must be stronger than the lake vessel and more costly in construction, opera¬ tion, and maintenance, and it must be fitted with expensive appli¬ ances which are not required in the lake traffic. In considering this question, it must be remembered that under existing conditions the lake vessel is compelled to be idle during about one-third of the year, while if it had free access to the sea and were constructed for foreign or coast navigation, it could be earning money all the year round. Mr. K irby estimates the cost of our type vessel No. 1 (see Appendix No. 5), when designed for lake and ocean business, at 1387,000, and when designed for lake business only, at $360,000. The daily cost of maintenance and operation, including 5 per cent on first cost, is in the first case $331, and in the second $404. Such a vessel, when designed for lake and ocean business, could carry a full cargo from Duluth or Chicago to New York, and, owing to the additional buoyancy of sea water, then take on all the coal required for the ocean voyage without overloading. The benefit to commerce which would result from giving access to shipping from the lakes to the sea, thus rescuing the lake fleet from enforced idleness during one-third of the year, would, of course, be enormous, if the problem of constructing a vessel economically adapted to both kinds of service can be satisfactorily solved. This is a benefit which is peculiar to the waterway and can not be derived from the extension of railway facilities. It is further stated that if adequate water communication with the sea were provided, a great industry in the construction of steel ships 124 DEEP WATERWAYS. would be immediately developed on the lakes. This industry is already an important one, no less than 1,258 vessels having been con¬ structed at the lake ports during the last ten years; but as there is no access to the sea for vessels of more than about 13 feet draft, the business is almost exclusively confined to the construction of ships for the Lake service. It is claimed that nowhere in the world are the conditions for the economical construction of steel vessels more favorable than at some of the lake ports. Cleveland, for example, is the center of a great iron and steel manufacture. It is farther from the coke-producing district than Pittsburg, but this disadvantage is counterbalanced by its advantage of receiving ores by direct and cheap water transporta¬ tion. The opening of a deep waterway to the sea would enable the shipyards of the lakes to compete with those of the seacoast in the construction of vessels for the ocean traffic. Finally, the argument has often been advanced that a deep water¬ way connecting the lakes with the sea would be of great military value in connection with the defense of the northern frontier of the country. Such a waterway would enable ships of war to pass between the sea and tlie lakes, and it would also permit the economical construction of such vessels at the lake shipyards. The preceding brief statement of commercial and transportation con¬ ditions and of the benefits which may be expected to result from the establishment of a dee}) waterway from the lakes to the sea is intended only as a basis for the comparison of the relative advantages and dis¬ advantages of the 21-foot and 30-foot waterways. Under the provi¬ sions of law it is not the duty of the Board to report upon the general question as to whether the requirements of commerce justify the con¬ struction of a deep waterway at the expense of the General Govern¬ ment, or to compare the advantages of such a waterway with those of one with moderate dimensions requiring transfers of freight at both its terminals. Nevertheless, before making tlie comparison required by law, it seems desirable to invite attention to two important points. The first point is that if any project is undertaken by the Govern¬ ment it should be fully adequate to the present and future purposes which it is intended to subserve. It has been remarked by a distin¬ guished authority that the life of any public work is practically coin¬ cident with that of the generation which began it. This is especially true in a country like our own, where population increases and com¬ merce develops with amazing rapidity. The reason why it is true is because, in the construction of such works, future necessities are almost invariably underestimated. For example, the first canal and locks at St. Marys Falls were completed in 1855 at a cost of about $1,000,000. To meet the necessities of the increasing traffic, a new and much larger lock and canal were commenced in 1870 and completed in 1881 at a cost of $2,171,000, including canal enlargement. This was DEEP WATERWAYS. 125 soon found to be insufficient for the requirements of the lake naviga¬ tion, and still another and larger lock was commenced in 1887 and completed in 1896 at a cost of about $3,700,000, including also canal enlargement. The volume of the lake traffic has so greatly increased that at the present time the construction of a new lock is under con¬ sideration. In 1870, no one could have been bold enough to suggest the construction of a lock of the size and cost of the Poe lock recently completed; and yet, if such a lock had been then constructed, the results would have been a large saving to the Government and a great benefit to the commercial interests of the Lakes. It is therefore of the highest importance that any waterway con¬ structed by the Government should be fully capable of meeting every possible commercial demand which may arise in the future. A lack of capacity for future commerce might necessitate its entire recon¬ struction at enormous cost, and require an adaptation of vessels and traffic to new conditions involving great loss to commercial interests. The second point is that any project undertaken by the Government should be of a national and not of a local character, benefiting many and varied commercial interests and exerting its influence over as great an extent of the country as possible. It is easily conceivable that a barge canal of moderate dimensions, requiring transfers at Buffalo and New York, might be of more direct benefit to the State of New York than a canal of sufficient dimensions for the uninter¬ rupted passage of ships; but much of this benefit would be at the expense of the producers and shippers of other parts of the country. Moreover, with such a canal the large interests of shipbuilding and winter traffic for the lake fleet would be unprovided for. It appears from the investigations of the Board that the most favor¬ able route for a 30-foot waterway from the Lakes to the sea is from Lake Erie to Lake Ontario via Lasalle and Lewiston, and from Lake Ontario to the Hudson River via Oswego and the Mohawk Valley, on the low level plan, and that the same route is practically as favor¬ able as any for the 21-foot waterway. This route is entirely in our own country and has a longer season of navigation than the more northerly line. The problem of its defense is, of course, much simpler than it would be were a part of it in a foreign country, and it is avail¬ able as a line of communication for ships of war. In the following comparison of the 21-foot and 30-foot waterways, this route will alone be considered. COST OF CONSTRUCTION. The estimated cost of the 21-foot waterway on the low-level plan is $206,358,000; the estimated cost of the 30-foot waterway is $317,284,500, to which should be added about $9,607,500 for the necessary deepen¬ ing of the harbors at Duluth and Chicago, making the total cost $326,892,000. DEEP WAT EE WAYS. 126 COST OF MAINTENANCE AND OPERATION. The annual cost of maintenance and operation is estimated at $2,343,478 for the 21-foot waterway, and $2,930,308 for the 30-foot waterway. COST OF TRANSPORT PROPER. The theoretical cost of moving the freight unit, exclusive of toll, from one terminal to the other on the lines considered, is given in the following table: Route, New York to— 21-foot waterway. 30-foot waterway. Domestic. Foreign. Domestic. Foreign. Total cost. Cost per ton mile. Total cost. Cost per ton-mile. Total cost. Cost per ton-mile. Total cost. Cost per ton-mile. Duluth. Chicago.. Mean ._ Cents. 45.2 42.3 Mills. 0.31 .31 Cents. 70.2 67.3 Mills. 0.48 .49 Cents. 45. 4 42.7 Mills. 0.31 .31 Cents. 40.9 38.2 Mills. 0.28 .28 .310 . 485 . .310 .280 It must be remembered that these values are purely theoret ical, and are not given as the probable freight rates; but they are believed to be proportional to the latter, and may, therefore, be taken as relative measures of the cost of transport proper for the waterways compared. The table shows that the cost of transport proper on the 21-toot waterway is about the same for domestic traffic as on the 30-foot water¬ way. For foreign traffic the 30-foot waterway shows a much lower cost of transport than the 21-foot waterway. TRAFFIC CAPACITY. The maximum annual traffic capacity of the 21-foot waterway (when the single-lift locks are duplicated) is estimated at 30,608,000 net tons, and that of the 30-foot waterway at 35,180,000 net tons, the traffic on the smaller waterway being greater than that on the larger one, owing to difference in time expended in lockage. It should, however, be remarked that with smaller locks properly proportioned for the most economical type carrier the traffic capacity of the larger waterway would be somewhat increased. SPEED. The average speed on the 21-foot waterway is 10.67 miles per hour. The average speed on the 30-foot waterway is 10 miles per hour. ADAPTABILITY TO TRAFFIC CONDITIONS. Our vessel No. 1, which is the type vessel adopted for the 21-foot waterway, has a draft of 19 feet and can enter all the important lake DEEP WATERWAYS. 127 harbors as well as navigate along the seacoast. It is, therefore, mnch better adapted to domestic traffic than vessel No. 2, the type vessel for the 30-foot waterway, since the latter has a draft of 27 feet and can not enter the lake harbors. The smaller vessel is not so well adapted to deep-sea navigation as the larger one. REGULARITY OF SERVICE. In the 30-foot waterway navigation would be freer and for smaller vessels a little more rapid than in the 21-foot waterway, and there would be less danger of delay from accidents and crowding. The time required for vessel No. 1 to make a single trip from Duluth to New York on. the 30-foot waterway is six days and three hours, while the same journey on the 21-foot waterway would require two hours longer. INFLUENCE ON RAILWAY RATES. As both waterways furnish low rates for large traffic volumes, there seems to be little choice between them in this respect. OUTLET FOR THE LAKE FLEET. Even should a 30-foot waterway be established between the lakes and the sea, it is probable that the number of vessels of large draft in the lake service would be comparatively small, since such vessels could not enter most of the lake harbors and would be adapted only to through and principally foreign traffic. The 21-foot waterway would, therefore, be practically as good as the 30-foot waterway as a means of access to the sea for the lake fleet. ROUTE FOR SHIPS OF WAR. In the very improbable event of a war with Great Britain every large ship of war possessed by this country would be required on the high sea. Such vessels would be unnecessary on the lakes, since the greatest depth of the Canadian waterways is only 14 feet. For purposes of naval defense the 21-foot waterway appears to offer ample facilities. SHIPBUILDING. The 30-foot waterway would enable the shipbuilders of the lakes to construct seagoing vessels of the largest size, both for commercial and naval purposes. With the 21-foot waterway this industry must be restricted to the construction of vessels of not too great dimensions to pass the locks. If the width of the locks were made greater than is necessary for the type carrier, ships of larger size could be floated from the lake shipyards to the seaboard when light. This would increase the cost of the canal and diminish its traffic capacity. 128 DEEP WATERWAYS. CONCLUSION. As the result of this investigation, it appears that the 21-foot, water¬ way promises a much greater return of value relatively to its cost than the 30-foot waterway. The main advantages of the 30-foot waterway are that it would furnish the lowest cost of transport proper to foreign markets and permit the construction of the largest seagoing vessels on the lakes. The Board desires to express its obligations and thanks to the Canadian government for permitting surveys to be made within Canadian territory, and to the Montreal harbor commission for facili¬ ties extended for examining the Canadian canals. Thanks are due to many Canadian engineers for their cordial cooperation and assistance. Among these, the Board is especially indebted to Messrs. Thomas C. Kiefer, John Kennedy, Ernst Marceau, Thomas Monro, T. S. Rubidge, and J. G. Macklin for courtesies and valuable information. The Board desires to express its sincere thanks to Brig. Gen. John M. Wilson, Chief of Engineers, United States Army, and to the officers of the Engineer Department in Washington, for their courteous assist¬ ance during the entire progress of the work. Its thanks are also due to Lieut. Col. G. -J. Lydecker, Corps of Engineers; Mr. Joseph Ripley, superintendent of the St. Marys Falls Canal; Professors E. A. Fuertes and G. S. Williams, of Cornell University; Mr. Frank E. Kirby and Mr. Edwin S. Cramp, marine engineers; the Detroit Bridge and Iron Works, and the United States Coast and Geological Surveys. The Board wishes to express its high appreciation of the ability, faithfulness, and efficiency of its secretary and of the assistant engi¬ neers in charge of the various divisions of the work, whose reports are appended hereto. Special mention should be made of Mr. James II. Brace, who has efficiently filled the important position of principal assistant engineer during the preparation of this report, since his name does not appear elsewhere as in charge of responsible work. Respectfully submitted. C. W. Raymond, Lieutenant-Colonel, Corps of Engineers. Alfred Noble. Geo. Y. Wisner. Hon. Eljhu Root, Secretary of War. ^EEEISHDXXES Appendix No. 1. LOCKS. One of the most strongly marked features of recent navigation, both on the ocean and the Great Lakes, has been the steady increase in size of ships. The most accessible record relating to the Great Lakes is that of the St. Marys Falls Canal. During the year ending June 30, 1882, there were 4,384 passages by registered vessels, with a total net registered tonnage for the year of 2,3711,216, or an average of 543 tons per passage. During the calendar year of 1891 the average was 862 tons per passage; in 1899 the average was 1,146 tons, an increase of more than 100 per cent in seventeen years. Although these figures show the rate of increase in registered ton¬ nage, they do not give an accurate idea of the character of that change. While it lias been on the whole a continuous increase, it has been greatest when large additions were made to the navigable depth of water in the principal lake harbors and in the channels con¬ necting the lakes. In 1870 freight through the canal was carried mainly in sailing vessels of 300 to 400 tons net register, carrying car¬ goes of 600 to 700 tons on 11 to 12 feet of water, which was then the limiting depth. During the next eleven years the deepening of the harbors and connecting channels to 16 feet at mean stage was in progress, but was not available until the opening of the Weitzel lock, in 1881. This period was marked by the introduction of freight steamships, each towing one to three sailing barges. The net regis¬ ter of these ships was in most cases less than 1,000 tons, but a few were built of about 1,500 tons. In anticipation of the opening of deeper waterways, the new ships were designed to draw 14 to 15 feet when fully loaded. With the opening of the Weitzel lock the building of small sailing ships was checked, and after four or five years ceased almost entirely. The old ships became comparatively unprofitable, and during sea¬ sons of low freight rates many were put out of commission. The building of a larger class of ships of from 800 to 1,700 tons net regis¬ ter was taken up, and they carried a constantly increasing proportion of freight. The Canadian lock was opened in 1895 and the Poe lock in 1896, with a depth of 20 to 21 feet on the sills. As in the case of the Weit- II. Doc. 149-9 129 130 DEEP WATERWAYS. zel lock, the increased draft had been to some extent anticipated by the building of ships which could not be fully loaded on a draft of 16 feet. Several exceeded 2,000 tons net register, and a few exceeded 2,500. These ships were designed to carry about twice the register, but up to the end of 1804 the maximum cargo was less than 3,800 tons, showing that the depth of water in the channels did not permit full loading. The season of 1805, when the Canadian lock was opened, happened to be one of low water. Several large ships of nearly 3,000 tons register were in commission and two which exceeded 3,300, but the maximum cargo was only 4,400 net tons. During the next two years there was little increase in size, but about thirty ships were built which slightly exceeded 3,000 tons net register. With improved con¬ ditions in the harbors and channels, the maximum cargo rose to 6,244 net tons. In 1808 three ships of more than 4,000 tons register were in service, and the maximum cargo was 7,840 net tons. The maximum cargo in 1899, 8,339 tons, was carried by the John Smeaton, which has a registered tonnage of 4,725. The economy of transportation in these large ships has been so marked that the building of ships of less net registered tonnage than 2,000 for the through freight business from Lake Superior to Lake Erie ports has practically ceased. The largest ships now in use on the lakes have a length of 500 feet over all and a beam of about 52 feet. It is hazardous to say that the limit of dimensions has been reached or neared, but when it is considered how rapidly the cost of a ship increases with its length and how difficult it is to secure structural strength without increase of draft, it seems reasonable to conclude that no further very marked increase will take place until the harbors and connecting channels are made deeper. Although the existing channels do not have quite 21 feet of water, the larger ships are designed to be loaded to 19 feet or more. For the rapid and safe movement of a ship in the 21-foot waterway there should be about 2 feet of water under its keel. The larger ships now in use, therefore, have reached the limit of draft that should be per¬ mitted in this waterway. If it were certain that ships no larger than the largest now in use would furnish the most economical transporta¬ tion, the locks should be no larger than required to receive them. There may be, however, considerable development in length and beam, and if such an increase should prove practicable and economical, it would be a serious error if the locks were made too small. If, on the other hand, the locks were a little larger than needed, the cost would not be increased very much nor the operation of the canal impeded appreciably. The dimensions adopted—600 feet long and 60 feet wide—are sufficient for a ship 550 feet long over all and 58 feet beam. Such a ship would have about 25 per cent greater carrying capacity than the largest now on the lakes. These dimensions provide for the passage of the larger ships singly. DEEP WATERWAYS. 131 The locks of the St. Marys Falls Canal were designed to pass fleets of four large ships at a single lockage whenever so many were in waiting, and in practice the average number of ships per lockage is about two. This canal offers the most conspicuous existing example of tlie han¬ dling of a large traffic through a canal with locks, and some reasons must be given for departing from such a precedent. The first reason is that a lock large enough to contain a fleet of four of the largest ships likely to navigate the waterway would be large beyond precedent. The new or Poe lock at the St. Marys Falls Canal is 800 feet long and 100 feet wide. Before it was completed, ships of more than half its length were in use and the lock would receive only two of them at once. Within three years after its opening, ships were in use of more than half its breadth as well as more than half its length, and only one of them could be taken into the lock at one lock¬ age. For the “type ship” in the 21-foot channel (see Appendix 5) the lock to pass four at once would have to be 1,020 feet long and about 10(3 feet wide. To pass ships of the larger size thought possible, 550 feet long and 58 feet beam, the lock should be 1,170 feet long and about 120 feet wide. The difficulties of operating such a lock would be very great. The second reason is that each vessel would be delayed while other vessels were being placed in the lock. This is not of great moment at a single lock, but with the large number of locks on the deep waterway the aggregate loss of time would be a serious tax on navi¬ gation. It is believed that the waterway should be so designed as to provide the quickest transit for each ship. The preceding reasons relate to the economy of the ship. A third reason for preferring the lock for a single ship is found in economy of water supply. This is of most importance on the Mohawk route, where it would be difficult to obtain a sufficient supply for the large locks. Ocean ships have had a less rapid development than ships on the Great Lakes, but similar in kind. The controlling conditions are more complex than on the lakes; the depth in harbors varies greatly; the nature of the cargo, the number of ports of call, and the port facilities for handling freight differ from the simpler conditions and more per¬ fect appliances on the lakes, and on many routes it is impracticable for large ships to be profitably employed. The greater part of the freight transported by sea is carried in ships not exceeding 500 feet in extreme length and 55 feet beam. For pas¬ senger service large and fast ships are preferred by the traveling pub¬ lic. This fact has led to the building of larger and faster ships for a combined passenger and freight service than usually required for freight alone; yet the largest ships of the long-established Peninsular and Oriental Steamship Company which traverse the Suez Canal and visit East Indian and Australian ports do not exceed the dimensions 132 DEEP WATERWAYS. above named. One of the new ships of this line, the Assaye , launched in October, 1890, is only 4-50 feet long and 54 feet beam. The new ships of the White Star Line for the Australian trade are much larger, being about 575 feet long over all by 54 feet beam. The most marked development in size of ocean ships has been on the North Atlantic route. This would seem particularly pertinent to the deep waterway, because the greater part of the materials pass¬ ing through the deep waterway for export would cross the ocean by the North Atlantic route. The longest ships on this route are devoted wholly or mainly to the transportation of passengers. They would not traverse the deep waterway. The next in length are ships carry¬ ing both passengers and freight and running from 14 to 1!) statute miles per hour. The largest of this class are the Pennsylvania and Graf I Valdersee, of the Hamburg-American Line, 550 to 565 feet long and 42 feet beam; the Cymric , of the White Star Line, 600 feet long and 64 feet beam, and the Ivernia and Saxonia, of the Cunard Line, 600 feet long and 644 feet beam. These ships when fully loaded draw upward of 30 feet in sea water, while the permissible draft in the 30-foot deep waterway would not much exceed 27 feet in fresh water, equivalent to a sea draft of about 26 feet There must be some ratio between length and draft for the best economical results. This ratio between length and draft is treated very ably in a paper by Prof. J. If. Biles recently presented to the (British) Institution of Naval Architects. This paper contains estimates of cost of two series of ships, running from 500 to 700 feet in length. In the first series the full-load draft is uniformly 27 feet 6 inches. In the second series the draft increases from 27 feet 6 inches for the ship 500 feet long to 38 feet 6 inches for the ship 700 feet long, the ratio of length to draft being about 18. A comparison is then made of the cost of carrying a ton of cargo 5,000 nautical miles at a speed of 12 knots per hour in these ships. The results are shown in the following table: Length of ship—feet.... 500 550 600 650 700 First series, clratt 27 feet 6 inches for all: Cost, in shillings, per ton of cargo.. 8.6 9.0 9.6 10.4 11.2 Second series, draft varying: Draft ... . .. 27' 6" 30' 3" 33' 0" 35' 9" 38' 6" Cost, in shillings, per ton of cargo. 8.6 8.0 7. (5 7.2 7.0 This table shows clearly the greater cost of transportation in large ships per ton of cargo on a limited draft, as well as the economy of transportation in large ships with draft increasing with length. In order to check these results by an independent estimate with more complete details of the several elements of cost, one of the most prominent naval architects in the United States, Mr. Frank E. Kirby, was requested by the Board to prepare estimates of the cost of build¬ ing, maintaining, and operating ships of various dimensions from 480 X DEEP WATERWAYS. 133 feet in length by 52 feet in breadth to 550 feet in length by GO feet in breadth, all with a load draft of 27 feet. The estimates are given and discussed in Appendix 5. They show that the cheapest transporta¬ tion through the 30-foot waterway from Duluth or Chicago to New York Harbor would be given by the ship 500 feet long. The beam of this ship was assumed to be 54 feet. It is interesting to note that this length is practically the same as that of the largest existing ship on the Great Lakes—the John W. Gates —and that its beam is only 2 feet greater, while its draft exceeds that of the Gates by 7 or 8 feet. This analysis therefore supports the idea that the limit of size of lake boats has been nearly or quite reached unless the harbors and channels are made deeper. The foregoing figures have been based on the assumption that the cargo would be of such density that the ships could be fully loaded. Certain classes of freight might require more space than a ship 500 by 54 feet affords. Professor Liles shows that the larger ship on constant draft gives greater proportional space. This might lead to the use of a small number of larger ships. Further improvements in terminal facilities and further economies in operating ships might tend in the same direction. Although the contingency is extremely remote, it is conceivable that it may at some time become extremely desirable to pass war ships through the waterway. To pass our largest existing battle ships would require a width of locks of 75 feet, and it maybe expected that larger ones will be built in the near future. A reasonable provision for future increase would be made if the width were fixed at 80 feet. Among the industries which it is expected the waterway will develop at lake ports is that of building ocean ships. With locks 80 feet wide this industry could include the building of battle ships. With these several considerations before it, the Board decided to estimate for the 30-foot waterway, locks 740 feet in length from quoin to quoin, giving an available length of about 700 feet and 80 feet in width. The additional width of lock increases the time required for filling and emptying the lock, and thus reduces to some extent the commercial usefulness of the waterway. There will be difference of opinion as to whether the possible military value and the possible service to shipbuilding interests justify the infliction of this loss on the interests of navigation. In view of the result of the investigation regarding the length of ship to give the cheapest transportation, the length of lock adopted may seem too great. In single locks or when two lifts are combined it is proposed to place intermediate gates in the locks, so that a shorter chamber can be used when desirable. In these cases tin 1 long lock is objectionable on account of first cost only. Where more than two lifts are combined the intermediate gates are omitted, and if the adopted length of lock is greater than necessary it affects unfavorably 134 DEEP WATERWAYS. both first cost and facility of operation; but it must not be forgotten that successive economies in cost of transportation have almost invariably favored a larger ship, and it is judicious to make a large provision for possible increase. CAPACITY OF A LOCK OR SYSTEM OF LOCKS FOR THE PASSAGE OF SHIPS. It is necessary to show what capacity for traffic the waterways pro¬ posed will have. It is obvious that the limit is fixed by the locks. The method followed for investigation is to calculate the maximum capacity of a lock or series of locks to pass ships, and then to apply to this theoretical result a coefficient derived from experience at the St. Marys Falls Canal, to obtain the practical capacity. The discussion of several problems of the deep waterway requires a determination or assumption as to the average size of ships that will traverse it. The question is taken up briefly elsewhere, with the result that the average net register of ships in the 21-foot waterway is taken at 2,500 tons and in the 30-foot waterway at 3,000 tons. The greater part of the freight carried through the deep waterway will be bound eastward. It has been estimated that the west-bound freight will be one-third as much. Through the St. Marys Falls Canal the west-bound freight in 1899 was about 22 per cent of the east- bound. For the present purpose it is assumed that ships east bound will carry full loads and when bound west will carry one-fourth as much. A ship fully loaded carries about twice its net registered ton¬ nage. The total freight carried will therefore be, on the assumption given, two and one-half times its net register for the round trip, or an average of one and one-fourth times its net register for each passage. In calculating the capacity of a lock it is more convenient to deal with the net registered tonnage. The delay at a lock is calculated from the following data: 1. A ship approaching a lock will reduce speed at the rate of 1 mile per hour while moving 400 feet. 2. When within 700 feet of the lock-gate quoin its speed will have been reduced to 2 miles per hour. 3. If the lock is open, the ship will continue at this rate until its stern is within 150 feet of the lock-gate quoin (200 feet if moving downstream). It will come to a full stop in the next 200 feet, back¬ ing the wheel if necessary. 4. If the lock is not ready, the ship, on arriving within 700 feet of the lock-gate quoin with a speed of 2 miles per hour, will come to a full stop within the next 200 feet, backing the wheel if necessary, and tie up. When the lock is opened, the ship will acquire a speed of 2 miles per hour in moving 400 feet and continue at this rate into the lock, stopping as before. DEEP WATERWAYS. 135 5. The time required for opening- or closing a pair of gates is taken at two minutes. 6. When the ship is brought to a stop in the lock, it will move forward until its bow is within 50 feet of the next quoin and tie up. This movement will be made while the gates are being closed. 7. The time for filling or emptying a single lift, for filling the upper lock of a series, and for emptying the lower lock of a series is the the¬ oretical time divided b} 7 the coefficient 0.75. For the filling or empty¬ ing of intermediate locks of a series the coefficient is taken at 0.66. The coefficient is smaller in this case on account of the greater length of the culverts. 8. A ship leaving a lock acquires a speed of 2 miles per hour in moving 400 feet, increases speed to 4 miles per hour in the next 600 feet, and then proceeds to increase speed at the rate of 1 mile per hour while moving 400 feet until full speed is attained. Lockage capacity of the 21-foot waterway .—The maximum capacity of a single lock would be developed if ships were constantly in waiting and were passed alternately up and down. The minimum time required for the passage of two ships through a lock of 20 feet lift would be as follows: From tie-up place into lock, first ship..minutes.. 9.3 Closing lower gates. . do - 2.0 Filling lock.. . . . . do.... 5.1 Opening upper gates____-.do_ 2.0 Leaving lock passing waiting ship. .. ...do- 8.1 Moving into lock from tie-up place, second ship . ....do _ 9.0 Closing upper gates.. do- 2.0 Emptying lock . . ..do _ 5.1 Opening lower gates, . do _ 2.0 Leaving lock and passing waiting ship..do _ 8.1 Time for two ships ..... do_ 52.7 Time per ship __ ___ .do _ 26.35 Number of ships in a navigation year of two hundred and thirty-seven days _ __ ____ 12,952 Annual net registered tonnage, at 2,500 tons per ship_ _ tons . 32,3s0, 000 Annual freight traffic, assumed to be one and one-fourth times the reg¬ istered tonnage.. ..1. tons.. 40,475,000 This would lie the maximum traffic capacity if the lock were in con¬ stant operation with no delays whatever. Notwithstanding the great¬ est perfection in operating machinery, the utmost skill in handling it, and the greatest care in the moving of ships at the locks, accidents and delays will occur, increasing considerably the average time of lockage. The condition that ships shall always be in waiting is inad¬ missible, because it would involve so much aggregate delay to ship¬ ping as to make the route unprofitable. By reason of varying weather, different rates of speed, and delays in receiving freight, the distribu¬ tion of ships along the route is far from uniform, and as the traffic 136 DEEP WATERWAYS. capacity of a lock is approached the delays to shipping augment in rapidly increasing ratio. Experience at the St. Marys Falls Canal shows this very clearly. The greatest traffic through this canal with only one lock in oper¬ ation occurred in 1894 through the Weitzel lock. The delays to ships were so great that it may be considered the practical traffic capacity of the lock was reached. The canal records enable a comparison to be made of the theoretical (or calculated) and the practical traffic capacity of the locks of the deep waterway. In applying the data it is necessary to make two corrections, which depend on the character of the shipping. These are: First. The ships which will traverse the deep waterway are assumed to average 2,500 tons net register (for the 21-foot waterway). This is more than twice the mean tonnage of the ships traversing the St. Marys Falls Canal in 1894, and the larger ships require more time in handling. Second. Two or more ships are usually locked through together at the St. Marys Falls Canal, while it is expected that ships will pass singly through the locks of the deep waterway. The time required for lockage should therefore be less for the deep waterway. A third correction depends on the time required for filling or emptying the lock. This will be less for the lock of 20-foot lift in the 21-foot-deep waterway than for the Weitzel lock. The following data relating to the traffic of 1894 through the Weitzel lock are taken from the official report of the engineer officer in charge: Net registered tonnage ____ The number of lockages__ The number of ships___ _ Average time of lockage____ Average time in lock for each ship... . 14491 The average number of ships per lockage was =2.25. 13.110,366 6,431 14. 491 40' 34" 34' 32" The net registered tonnage given above is taken to represent the practical traffic capacity of the Weitzel lock. It is to be shown what the corresponding capacity would have been if the ships had been locked singly, if the net registered tonnage of ships had averaged 2,500, and if the time for filling and emptying the lock had been the same as calculated for the lock of 20-foot lift in the 21-foot waterway. At the St. Marys Falls Canal the time for a lockage is taken as beginning when the bow of the ship enters the lock and ending when the stern leaves it. The entire cycle from the beginning of one lock¬ age to the beginning of the next is greater. Sufficient time must be added to the given time of lockage to permit the ship to move from t he outlet of the lock past the waiting ship (which is supposed to be lying as near the lock as practicable) and for the movement of the second ship from its waiting place to the entrance to the lock. The second ship can not move until the first ship passes it. The time DEEP WATERWAYS. 137 required for these two movements lias not been observed at the St. Marys Falls Canal. It has been calculated for the deep waterways to be 8 minutes. For the smaller ships of 1801 it would probably have been about two-thirds as much, or 5.3 minutes. Adding this to the reported time of lockage, there results when lockages are consecutive— Interval from the commencement of one lockage to the commence¬ ment of the next at the St. Marys Falls Canal in 1894,40.5 + 5.3=45.8 minutes. It is next to be determined how much this lockage interval would be reduced if ships were locked singly. Let n - number of ships per lockage. Let x = time of ship in lock when locked singly. Let u — additional time required to move another ship into or out of the lock. The additional time per lockage and additional time per ship will be as per following table: Num¬ ber of ships per lock¬ age. Delays to ships while others are entering lock. Delays to ships while others are leaving lock. Total delay to ships. agendo- Lockage age ue t j me *hip. increased. To first ship. To To second third ship. ; ship. To fourth ship. To To first second ship. ; ship. To third ship. To fourth ship. u = l .. (i 0 ' . (i 0 0 n —2 V 3 y 6 . 0 y 0 y 0 y 2 y y 1 2 // 2 y | 4 y 3 y fi// n =3 -. y o 2.V V 2 y 2// 12 y ix =4 . .. 0 From this table we can deduce the following form like: Time of ship in lock =,/•+(/<— 1) y. Time of lockage=,r+2 (n — 1) y. For 1894, =2.25, and tin* equations become .r+ (2.25— 1) y— 34.5. .r+2 (2.25—1) i/=40.5. Whence y = 4.8 minutes. ir=2S.5 minutes=time of ship in the Weitzel lock in 1894 when locked singly. File entire interval between the commencement of one lockage and the commencement of the next would have been 5.3 minutes more, or 28.5+5.3 = 33.8 minutes. With the larger ships assumed for the deep waterway, this interval would be greater on account of the additional time required to move the larger ship from the waiting place into the lock and from the lock past the waiting place, including the movement in the lock as well as in the approaches. The calculated time for these movements is 17.34 minutes; one-third of this is 5.8 minutes. Adding this to 33.8, there 138 DEEP WATERWAYS. results 39.6 minutes as the time required for locking the assumed deep¬ waterway ship through the Weitzel lock. If the deep-waterways lock of 20 feet lift were substituted for the Weitzel lock, the lockage time would be reduced by reason of quicker filling and emptying. This reduction would be about 3 minutes, making interval from commence¬ ment of one lockage of a single ship to commencement of next through the deep-waterways lock of 20 feet lift (21-foot waterway)=39.6 — 3.0= 36.6 minutes. This is larger than the calculated time given on page 135, as it ought to be, because it includes delays of all kinds in the locks, but it does not include delays of ships awaiting lockage. It is to be compared with the corresponding time actually required for fleet lockages in 1894, given on page 137 as 45.8 minutes. It may be presumed that if the lockage time were reduced from 45.8 to 36.6 minutes the number of lockages would be correspondingly increased and the traffic could be augmented in like ratio with no more delay to waiting ships. The actual number of lockages having been 6,431, the number through the deep-waterways lock with one ship of 2,500 tons net register per lockage would be— 6,43t X^f =____-..... 8,048 00.0 Net registered tonnage, at 2,500 tons per ship.... 20,120, 000 Freight tonnage, one and one-fourth times net registered tonnage_ 25,150,000 With this traffic the delays to waiting ships would be about the same as occurred at the St. Marys Falls Canal in 1894, and it may be con¬ sidered the practical maximum capacity of a single lock of 20 feet lift in the 21-foot waterway. It is about 0.621 of the calculated maximum capacity of this lock. This coefficient will be applied in the following cases to the calculated maximum capacity to obtain the practical capac¬ ity of the several lockage systems. This deduction of practical capacity of the lock of the deep water¬ way is based upon the best existing data. Although certain factors introduced are derived from assumptions as to movement of ships, it is not believed that any error contained in them can affect in an important degree the traffic capacity obtained from the discussion. If two successive lifts of 20 feet each were combined into one flight, the minimum time required for the passage of two ships in opposite directions would be 82 minutes, or 41 minutes per ship. This would make the calculated maximum capacity of the system 26,012,000 freight tons per navigation year of 237 days, or, applying the coeffi¬ cient, 0.621, a practical maximum capacity of 16,153,000 tons. If three or more locks are combined into one flight, the capacity of the system will be still less. If the system is doubled, ships will be passed down in one flight and up in the other. The minimum interval DEEP WATERWAYS. 139 between two ships will be as per following statement. Take the Lew¬ iston flight, with locks of 40 feet lift, as representing the worst case: Minutes. From tie-np place into lock (from below) ___ 9.3 Closing gates_______ 2.0 Filling lock______ 5.8 Opening second gates___ 2.0 Moving into second lock_____ 6.7 Closing second gates___ 2.0 Emptying first lock_ ____ 7.2 Opening upper gates_____-....- 2.0 Interval_ ___ 37.0 With an interval between ships of 37 minutes, the number of ships in the navigation year of 237 days would be 9,224; net registered ton¬ nage at 2,500 tons per ship, 23,060,000; freight tonnage one and one- fourth times registered tonnage, 28,825,000. For the duplicate line of locks the capacity would be the same, making the calculated maximum traffic = 28,825,000 X 2 = 57,650,000 tons. The practical maximum=0.621 of this=35,801,000 tons. Lockage capacity of the 30-foot waterway .—Following the same method as for the 21-foot waterway, the minimum time required for the passage of two ships in opposite directions through a single lock of 20 feet lift would be as follows: Minutes. From tie-up place into lock, first ship ..... 9. 6 Closing lower gates______ 2.0 Filling lock....... 8.4 Opening upper gates____-.... 2.0 Leaving lock and passing waiting ship ..____ 8.3 Moving into lock from tie-up place, second ship___ 9.3 Closing upper gates_____ 2.0 Emptying lock ------ 8.4 Opening lower gates.......... 2.0 Leaving lock and passing waiting ship-..-- 8.3 Time for two ships.-__ 60.3 Time per ship__..____. 30.15 Number of ships in a navigation year of 237 days. ___ 11,319 Annual net registered tonnage, at 3,000 tons per ship_tons.. 33,957,000 Annual freight tonnage, assumed to be one and one-fourth times the registered tonnage.do_ 42,446,000 This is the calculated maximum. Using the same coefficient as before, 0.621, the practical maximum would be 42,446,000 x 0.621 = 26,359,000 for a single lock of 20-foot lift in the 30-foot waterway. If two successive lifts of 20 feet each were combined into one flight, < the minimum time required for the passage of two ships in opposite directions would be 96 minutes, or 48 minutes per ship. This would make the calculated maximum capacity of the system 26,662,000 net 140 DEEP WATERWAYS. tons of freight in a navigation year of 237 days, or a practical maxi¬ mum of 0.621 of this = 16,557,000 tons. With three or more locks combined in one flight the capacity of the system will be less, as already noted in discussing the locks for the 21-foot waterway, and if greater capacity is desired the system must be doubled, one flight of locks being used for ships bound down, the other for ships bound up. The maximum capacity of a flight depends on the minimum interval required between ships. Taking the Lew¬ iston flight with a lift of 40 feet at each lock, the minimum interval between ships will be as follows: Minutes From tie-up place into lock Closing first gates _ Emptying first lock _ Opening second gates._ Moving into second lock. _ Closing second gates.. Filling first lock. Opening first gates.. 9.6 2.0 9.6 2.0 7 .1 2.0 11.9 2.0 Interval between ships.. 46.2 With an interval between ships of 46.2 minutes the number of ships in a navigation year of 237 days would be 7,387. Net registered ton¬ nage at 3,000 tons per ship, 22,161,000 tons. Freight tonnage, one and one-quarter times registered tonnage, 27,701,000 net tons. For the duplicate flight of locks the capacity would be the same, making the calculated maximum capacity— Tons. 27.701.000 X 2=.... .... 55,402.000 The practical capacity = 0.621 of this....... .... 34.405,000 This is less than the capacity of the same system of locks in the 21-foot waterway, although the assumed average tonnage per ship is 20 per cent greater. The result illustrates the cost to purely com¬ mercial navigation interests of adopting lock dimensions much greater than commercial ships are likely to require in order to make the water¬ way more useful for war purposes. The Board assumes, for the discussion of technical questions affect¬ ing the waterway, that the annual traffic will be 25,000,000 tons. It is believed that this can be passed through a single lift by a single lock; but the capacity of the lock will be nearly reached, and the delays to ships while awaiting lockage will be considerable. At the St. Marys Falls Canal in 1894 the delay to ships while awaiting lock¬ age was an average of 34 hours for each. The delays were due mainly to bad weather, causing ships to arrive at times in fleets. These delays increase rapidly when the waterway is overcrowded. In 1890, with a tonnage of 8,500,000 tons, the average delay was less than 14 hours. In 1896, when the Canadian lock was in successful operation and the new Poe lock afforded some relief, the delays were DEEP WATERWAYS. 141 reduced to less than one hour, although the tonnage had increased since 1894 from 13,100,000 to 17,200,000. In 1897 all three locks of tln- system were in successful operation, and the average delay was only one-fourth hour, with a tonnage of 17,600,000. The preceding estimates of capacity of locks have been based on the record of the St. Marys Falls Canal in 1894. If the treatment is cor¬ rect, it follows that when the annual freight tonnage in the 21-foot waterway reaches 25,150,000, the delays at that point with only a single lock in use would be about 34 hours per ship; in the 30-foot waterway this delay would occur when the annual freight tonnage reached 26,359,000. This unequal distribution of shipping would probably be greater at the St. Marys Falls Canal than at any other lake point on the route, but delays of a similar character will occur on the Mohawk route at Oswego to ships bound east and at Albany to ships bound west. On the St. Lawrence-Champlain route similar delays will occur at the Galops Rapids lock to ships bound east and at the Troy lock to ships bound west. The delays at the termini of the Niagara section would be much less on account of the greater capacity of the double lockage system provided there. The delay to ships at intermediate locks on each of the divisions of the waterway will be small, because the terminal locks will distribute them. If all ships moved at uniform rates, the delay would be inappre¬ ciable, but as some will move more rapidly than others, there will be small delays at every lock or flight of locks, and the aggregate will be of some importance. The doubling of locks, which increases the capacity and reduces the delays, has the further advantage that if one of the locks or flights is disabled traffic can be continued through the other, and a total stop¬ page of traffic averted. The double system has been adopted where two or more lifts are combined, because the anticipated traffic requires it. Careful consideration has been given to the question of doubling locks at single lifts (where a single lock will pass the antici¬ pated traffic), in order to lessen the danger of a stoppage of navigation by the disabling of a lock; but it is believed that the danger of such a contingency is not great enough to justify the large increase in cost. DIMENSIONS OF DUPLICATE LOCKS. Where the locks of the 21-foot waterway are to be doubled, the second lock is designed to be of the same width as the first, viz, 60 ■ feet. For the 30-foot waterway the width of 80 feet was fixed, as already stated, to permit the passage of war ships and to facilitate the building of large ships for ocean traffic at shipyards on the lakes. A lock 70 feet wide would pass the largest existing commercial ship, and a width of 60 feet would probably be sufficient for nearly all the ships which would use tlm waterway. The width of 60 feet was there¬ fore fixed for the second line of locks in the 30-foot waterway. 142 DEEP WATERWAYS. With this project executed, ships of the largest class could only he passed in the large locks. If such a ship were going in the direction of the traffic through those locks no difficulty would occur; if going in the opposite direction it would interfere with traffic to some extent. Where two lifts only are combined into one system the interference would be slight and unimportant, but where more than two lifts are combined, notably in the case of the six combined locks of the Lewis¬ ton flight, it would be necessary practically to fill (or to empty accord¬ ing to direction of ship’s movement) all the locks in the flight before the ship could enter the second lock, and the delay to the ship would be considerable. All ships waiting to pass the chain in regular order would also be delayed. If the proportion of ships exceeding 00 feet in breadth should be large, and if the traffic through the waterway were great, the delays would affect the usefulness of the waterway to a serious extent. The proposed width of 60 feet for the duplicate locks would then prove insufficient, and a width of 65 feet or more would be required. The cost of increasing the width 5 feet would be about $430,700 for the entire waterway. LIFTS OF LOCKS. In a few localities the topography favors larger lifts than have been adopted heretofore. In all these localities solid-rock foundations are available, and will carry safely any load that the largest lifts would impose on them. Other points, however, required examination. The first point was whether gates could be designed to safely stand the extreme heads which would result from the project investigated by the Board. The Lewiston flight, for example, with lifts of 40 feet, would subject the intermediate gates in the 30-foot waterway to heads in some cases as great as 77 feet. The study of this subject, given in detail in Appendix No. 2, showed that gates to sustain this head are entirely practicable. The next point was in regard to the relative usefulness for naviga¬ tion of low and high lifts. With low lifts the interval between ships would be less and the maximum capacity of the waterway would be greater. This may be illustrated by reference to the Lewiston flight. If the lifts were 20 feet each, the minimum interval between two ships in the 21-foot waterway would lie as follows: Minutes. For movement of ship from tie-up place into lock____ 9.3 Closing lower gates._.. 2.0 Filling lowerlock......... 4.1 Opening second gates . .. 2.0 Moving ship into second lock.... _.... 6.7 Closing second gates.... 2.0 Emptying lower lock..... 5.1 Opening lower gates..... 2.0 Interval...33.2 PEEP WATERWAYS. 143 With this interval between ships tlie number of ships in a naviga¬ tion year of 237 days would be 10,280 in each flight, and the net regis¬ tered tonnage, at 2,500 tons per ship, 25,700,000. Comparing this with the corresponding estimate for 40-foot lifts (see p. 139), the reduction in capacity resulting from doubling the lift appears to be about 10 per cent. If the waterway should become crowded, the delays to individual ships while awaiting lockage would be a little greater with high lifts than with low ones, because the traffic limit would be reached sooner; but unless the traffic exceeds present esti¬ mates the delay in either case would be inconsiderable. A more important feature is the delay that would result to every ship if the projected lifts were reduced. Again using the Lewiston flight for comparison, the estimated minimum time required to pass the 6 locks of 40 feet each in the 21-foot channel is 108 minutes. If the lifts were 20 feet each, 12 locks would be required, and the mini¬ mum time for passing them would be 18G minutes, giving a greater delay in the low-lift system of 78 minutes, to which every ship would be subjected at this single locality. This is the minimum possible, and will be exceeded, because the unforeseen delays in passing through 12 locks are greater than in passing through G. A third point affecting the lift of locks is the cost of the project. No general rule can be laid down and each case must be studied by itself. No estimate has been made of the cost of the Lewiston flight with 20-foot lifts, but comparative estimates with lifts of 30 and 40 feet show that the system with 30-foot lifts would cost 84,GOO,000 more than with 40-foot lifts, and a similar further increase may be expected if the lifts were reduced to 20 feet. From these considerations there appears no doubt that the high lifts adopted are in every way feasible, cheaper to construct, and bet¬ ter adapted to the use of navigation than those heretofore in com¬ mon use. GENERAL DESCRIPTION OF LOCKS. Under this head it may be well first to repeat the lateral dimensions already given. For the 21-foot waterway: All locks to be 600 feet long between quoins of lock gates and GO feet wide. For the 30-foot waterway: Single locks to be 740 feet long between quoins of lock gates and 80 feet wide. Where two or more lifts are combined into a flight, two lines of locks are provided. The locks in one line have the same dimensions as single locks; in the other line the length remains the same, but the width is reduced to GO feet. All single locks and combined locks, not exceeding two lifts in series, are provided with intermediate gates dividing the chamber into two unequal parts, so that the entire lock or a part thereof can be used. Iu both the 21 -foot and the 30-foot waterways guard gates are placed 144 DEEP WATERWAYS. at. the head and foot of each single lock and at the head and foot of each flight, so that any lock can he quickly prepared for pumping out or draining. This provision is made in all the locks at Sault Ste. Marie and has proved valuable. The lock walls are designed to resist overturning by gravity alone, except at the gate recesses of the middle wall separating the double flight on the Lasalle-Lewiston line. At these places if one of the locks is pumped out while the opposite one is in operation, a contingency very 1 ikely to arise, tlie thickness adopted is not quite sufficient to wholly prevent tension in the masonry. Three alternatives were con¬ sidered to meet this: First, to introduce enough steel members in the base of the wall to take the tension. The cost of this would be $163,200 for tlie entire'line. Second, to increase the thickness for a length of 120 feet at the gates. This would make the chambers of the locks several feet wider than the entrances to them, The increased width would not facilitate the passage of ships in any respect, but, on the contrary, would delay them by reason of the greater time required for filling and emptying locks. This alternative was therefore re¬ jected. Third, to increase the thickness of the middle wall for its whole length, the thickness at the gates recesses being the same as in the preceding alternative, making it a gravity section, but with a large surplus over theoretical requirements everywhere except at the gates. The cost of the additional thickness would be about $542,400 for the entire line. It would exceed the cost of the first alternative by $379,200. The first alternative was therefore adopted. All lock floors are in the form of inverts. At the head and foot of each lock the thickness of the invert is sufficient to withstand the maximum difference in head at the gates. In the interior of the cham¬ ber the thickness is reduced, the amount of reduction depending on the character of the rock foundation. The strength of the inverts as shown on the plans is doubtless excessive where rock is sound, but without further information by means of borings with the diamond drill or actual examination of the excavated lock pits it is not prudent to make a close estimate in this respect. An additional reason for this liberal provision is found in the fact that lock excavation and construction is work of an extremely hazardous nature with large contingencies, and if economies may be effected in some localities in the extent and cost of inverts they are likely to be balanced by unex¬ pected difficulties and expenses in other localities. CULVERTS FOR FILLING AND EMPTYING LOCKS. For tilling and emptying the locks a culvert 87 square feet in cross section is placed in each wall, extending its whole length. Each cul¬ vert communicates with the chamber through a large number of small branches, each 9 square feet in cross section. The combined area of the cross sections of the branches is about double the cross section of the culverts, this excess being provided in order to reduce the force DEEP WATERWAYS. 145 of the currents into the lock. The spacing between the branches becomes less toward the foot of the lock, because the amount of water issuing from each in filling diminishes from the head toward the foot, and it is desirable to make the amount of discharge from the culverts as nearly uniform as practicable per unit of length. The culverts are of the same dimensions for locks of all sizes. It may be asked why larger culverts were not used for the larger locks. The speed of filling, when the culverts are properly designed, is not limited by the behavior of ships in the locks, but by the effect of the current caused in the immediate vicinity. This effect is practically independent of the area of the locks. The culverts adopted are be¬ lieved to be as large as it is prudent to make them. If the principle of dimensioning the culverts according to the currents induced in the waterway were carried out rigidly it would lead to a reduction of size of culvert with increase in lift of lock; but this would greatly increase the time required for filling or emptying high-lift locks. The difficulty can be met by opening the culverts only partially until the first and strongest rush of water is past. The culverts at the inlets and for some distance inward are to be lined with granite; the remainder, and greater part, of the length of the main culverts is to be lined with hard brick, equal in quality fo the best paving brick. This is in conformity with the practice at the Manchester Canal. The branches from the culverts to the lock cham¬ bers are to be lined with cast iron. These branches are of smaller dimensions than usual in foreign practice. It will be noted that in height and direction they are designed to discharge under the hull of a ship lying in the lock, not against it. In the locks of greatest lift it may possibly be necessary to line the culverts with cast iron. The upper lock of the Lewiston flight will be filled, and the lower emptied with a maximum head of 40 feet, while in the intermediate locks the maximum head will be 80 feet. The current due to a head of 80 feet will have twice the velocity and four times the destructive effect of one due to 20-foot head. A continuous culvert lining of 1^ inches of cast iron, with 50 per cent added for flanges and stiffening ribs, would weigh about 3,000,000 pounds per lock, and would add about $90,000 to its cost. For the six locks of the Lewiston flight and the two combined locks of 40 feet each near the foot of the Lasalle-Lewiston route the additional cost would be about $1,440,000. The inlets and outlets of the culverts are designed to impede the flow of water as little as practicable. The water for filling a single lock, or the upper lock of a flight, is received immediately above the gate and discharged along the chamber, the average movement of water from the pool above the gates to the chamber being about half the length of the lock. The water travels about the same distance in the culvert in emptying a single lock or the lower lock of a flight. In these cases a coefficient of 0.75 is applied to the theoretical II. Doc. 149-10 146 DEEP WATERWAYS. velocity in calculating time required for filling or emptying a lock. This is considerably larger than that deduced from observations on the locks of the St. Marys Falls Canal, where the design of the cul¬ verts is such that the loss of head is very great. It is somewhat less than that observed by General Abbot at the Manchester Canal, where the culverts are carefully designed to avoid loss of head. For the fill¬ ing and emptying of intermediate locks in a flight the average travel of water in the culverts is a full lock length; the coefficient of velocity should therefore be smaller and is taken at 0.66. The valves or gates for opening and closing the culverts are to be placed in the main culverts near the lock gates. They are designed to be of the form known as Stoney sluices, counterbalanced, and moving on live rollers to reduce friction. CHARACTER OF MASONRY. All the masonry of the locks except the culvert linings already men¬ tioned, the hollow quoins, and the upper portions of the miter sill walls is to be of concrete. A thickness of about 2 feet in the face of the lock walls and all the concrete in the miter sills, head wall, and invert are designed to be of concrete containing mortar of one packed volume of Portland cement to two packed volumes of sand, with as much stone as practicable. A cheaper concrete, containing 1: 3 mortar with Portland cement, will suffice for the body of the main walls, and a still cheaper mixture, made with natural cement, is intended for a small portion of the middle wall between duplicate locks, where weight only is needed. LOCK GATES. The lock gates are designed to be of steel. Excepting the interme¬ diate gates of the 40-foot lifts on the Niagara Canal lines, there is no air chamber, the skin being on the upstream side only and the gate hangings having ample strength to sustain the entire weight. In the excepted cases a small air chamber is provided near the bottom of the gate. The design of the gates and the t.lieoiy of strains in them have received, it is believed, more careful, extended, and practical study than heretofore given to the subject. (See Appendix No. 2.) Miter gates, single-leaf swinging gates, and single-leaf rolling or sliding gates were designed and compared, resulting in the selection of the standard miter gates. The design adopted is of simple outline, with cheap shop details and of ample strength. OPERATING MACHINERY. It lias not seemed necessary to take up the subject of operating machinery in detail. It appears probable that electrical machinery will be found most suitable, but the development in this line is so rapid that any design made now would almost certainly be out of date in less than five years. It has therefore seemed more judicious to merely include in the estimate a lump sum sufficient to provide hydraulic or pneumatic machinery of ordinary character. DEEP WATERWAYS 147 ESTIMATE OF COST. The details of the estimate of cost of a standard lock of 20 feet lift and of the upper flight at Lewiston, for both depths of channel, are given in the folloAving tables. Excavation and back filling are not included, but are provided for in the general estimates for the waterway. Table No. 1. —Cost of standard single lock of 20 feet lift vnth intermediate gates. [Thirty-foot channel.] Quantity. Cost per unit. Total. Portland-cement concrete. Granite miter sills. Granite hollow quoins and culvert linings... Brick... Structural steel. Steel castings.... Steel forgings. Bronze bushings.. Iron anchor bolts. Iron port castings. Cast iron pipe... Timber, oak.. .do — .do_ .do_ .pounds.. .do_ .do .do_ .do- ...1,000 feet B. M_. 134,007 585 933 1,417 2,209,000 83,200 12,100 3,900 98,560 342,000 17,415 60.204 $6.00 45.00 55.00 12.00 .05 .06 .10 .40 .03 .025 .014 50.00 $804,042 26,325 51,315 17, (Hit 110,450 4.992 1,210 1,560 2,957 8,550 244 3,010 Total cost _ _ _ _ . _ _ _ 1,031,659 Table No. 2. —Cost of a standard single lock of 20 feet lift. [Twenty-one-foot channel.] Quantity. Cost per unit. Total. Portland-cement concrete.cubic yards.. Granite miter sills.do_ Granite hollow quoins and culvert linings.do- Structural steel...pounds.. Steel castings. do- Steel forgings ....do- Bronze bushings.do — Iron anchor bolts.do— Iron port castings.....do_ Cast-iron pipe...do- Timber, oak....1,000 feetB. M . Total cost.......^ _. 83.200 347 614 1,215 946,000 63.200 9,200 2,800 63,040 273,600 11.880 34. 776 $6.00 45.00 55.00 12.00 .05 .06 .10 .40 .03 .025 .014 50.00 $499,200 15,615 33,770 14,580 47,300 3,792 920 1,120 1,891 6,840 166 1,739 626,933 Table No. 3. —Cost of Lewiston flight of 6 locks, lifts 40 feet each. [Thirty-foot channel.] Quantity. Cost per unit. Total. Portland cement concrete. 1,654,819 $6.00 $9,928,914 Natural cement concrete . .do_ 145,567 3.00 4:36,701 Granite miter stills..... ..do_ 2,789 45.00 125,505 Granite hollow quoins and culvert linings. .do_ 8,707 55.00 478,885 Brick. 15,985 12.00 191,820 Structural steel... 12,488,100 .05 624,405 Steel castings.. .do_ 3,648,660 .06 218,92(1 Steel forgings... .do- 45,000 .10 4,500 Bronze bushings.. .do- 15,000 .40 6,000 Steel in middle wall... 4.080,000 .03 122,400 Iron anchor bolts. .do_ 397.600 .03 11.928 Iron port castings. .do- 4,185,000 .025 104,625 Cast iron pipe. .do_ 2,243.895 .014 31,415 Cast iron in stairs. .do- 172,800 .03 5,184 Timber, oak. .‘B. M.. 108.144 50.00 5,407 . 12,296,609 1 1,000 feet. 148 DEEP WATERWAYS. Table No. 4.— Cost of Lewiston flight of 6 locks, lifts 40 feet each. [Twenty-one foot channel.] Quantity. Cost per unit. Total. Portland cement concrete.cubic yards.. Natural cement concrete.do_ Granite miter sills...... ....do_ Granite hollow quoins and culvert linings..do_ Brick. do_ Structural steel.pounds.. Steel castings. do — Steel forgings. do- Bronze bushings. do- Iron anchor bolts ..do_ Iron port castings. do_ Cast iron pipe ..do.... Cast iron in stairs. .do_ Timber, oak.iB.M.. Total cost ____-.. 1,079,580 93,721 2,348 8.193 12, 209 8,686,000 2,754. 600 45, 000 14,600 364,480 3,364,200 1,902,060 172, 800 153,992 $6.00 3.00 45.00 55.00 12.00 .05 .06 .10 .40 .03 .025 .014 .03 50.00 $6,477,480 281,163 105,6611 450,615 147.228 434,300 165,276 4,500 5,840 10,034 84,105 26,629 5,184 7,700 8,206,614 1 1,000 feet. Respectfully submitted. Alfred Noble. The Board of Engineers on Deep Waterways. Appendix No. 2. LOCK GATES. Detroit, Mich., June 1 , 1900 . Gentlemen: I submit the following report upon the work done in making designs and estimates of the gates to be used in the locks of the various deep waterway routes investigated by the Board: As a preparation for the design of the gates, the plans of many existing large locks were carefully studied and the literature upon the subject, both in English and in foreign languages, was made the subject of an extended research. A list of the authorities consulted upon the subject of mitering gates is given at the end of this report, with a brief note in regard to each. It is believed that the list is quite complete, covering nearly all that has been written upon the subject. An investigation was made to determine whether some one of the numerous forms of single-leaf gate or the mitering gate is the more desirable tjqoe for the work in hand. The result was the adoption of the mitering gate as a basis of the designs and estimates. A study was then made of the relative merits of the horizontal and vertical systems of framing for mitering gates, resulting in the adop¬ tion of the horizontal system. Steel mitering gates of the horizontally framed type having been decided upon, a lengthy investigation was made to determine the rise of sill and the form, whether girder or arch, of horizontal frame which DEEP WATERWAYS. 149 will give the best results when economy of construction and facility of operation are considered. The question of the relative economy of the arch and girder form of horizontal frame has been a mooted one for a long time. The result of the investigation was the adoption of a rise of sill of one-fifth the width of the lock for all cases, and the adoption of a horizontal frame straight on the downstream side and curved on the upstream side, with depths varying for different widths of lock. This is the so-called girder type. It was found that it presents advantages of greater stiffness and requires a shallower recess in the side wall than the arch form, and that, considering the state of the labor and steel markets at the present time and during the last decade, it is actually cheaper. The detailed design of the gates for the GO and 80 foot locks, 21 and 30 feet depth of water on sill and lifts varying from 10 to 50 feet, was then proceeded with.- The accompanying general drawings, plates 70, 71, 72, 73, and 74, show typical designs for a number of cases. While only a limited number of drawings are shown, it is believed that they give all the important features of the design. For the quoin and miter posts, wooden-bearing pieces were used on gates having the lighter pressures, but it was found necessary to use metal in some of the gates for locks of very high lifts. A study was made of the variation in the position of the center of pressure at the miter and its influence in determining the most eco¬ nomical form of horizontal frame. The results of this study are given later. The use of air chambers and rollers to reduce the reaction upon the anchorage and pivot was considered and rejected except in the case of gates weighing more than 500,000 pounds per leaf, in which case the lower part of the gate is inclosed, but rollers have not been adopted even for these extreme conditions. The method of proportioning the girders is given in some detail in a subsequent section of this report. Although tlic gates are of tlie horizontally-framed type, there must necessarily be a few verticals. What the action of these verticals is, and what effect they have upon the distribution of loads between the horizontals, is a very complicated problem, and considerable time has been given to its solution by several different methods. For the purpose of an estimate, a large number of gates were designed in detail and their weights carefully computed. The law of variation of weight of gates for different widths and lifts of lock and depth of water on sill was discovered and some general equations giving the weight of gates for any width of opening, depth of water on sill, and lift of lock within limits were derived. The foregoing is a brief statement of the work done and the results obtained. 150 DEEP WATERWAYS. Following will be found, under their proper heads, detailed discus¬ sions of the points mentioned above. CHOICE OF TYPE. Besides the ordinary double-leaf mitering gate, which has been used so long and so exclusively that it has become the standard form, there are numerous forms of single-leaf gate. The differences between these various forms depend entirely upon the method by which they are moved into and out of place. The sliding gate moves endwise into a recess in the side wall of the lock. The lifting gate moves vertically upward, and the plunging gate vertically downward. The single-leaf swinging gate revolves about a vertical axis at one of its ends, and the quadrant gate about a vertical axis in the center line of the lock at a distance from the gate of the half width of the lock. The tumble gate revolves about a hori¬ zontal axis at its lower edge. The bascule gate revolves about a hori¬ zontal axis at one of its lower corners, operating like the bridge of the same name. All the above forms of gate have been proposed, and most of them have been used. All forms of single-leaf gates possess in common the advantage of allowing a somewhat clearer and more definite analysis to be made of their stresses than does the mitering gate, and also the advantage of allowing simpler operating machinery. For the work in hand the lifting gate could not be used on account of the masted vessels which will use the locks. The plunging gate is practically untried and presents serious dif¬ ficulties in the problem of keeping clean the pit into which it descends, and in most cases the masonry would cost more than for mitering gates. A roller bearing similar to that used in the Stoney sluice gate is usually introduced, which gives the gate the advantage of being moved rapidly and operated under much greater head of water than the mitering gate. The sliding gate has been used in numerous cases in Europe and presents many advantages, chief of which is its ease of movement, as it is pulled endwise, thus offering less resistance to the water than it would if moved in any other direction. In a tidal lock it is especially adaptable, since it will withstand a pressure from either side. The gate may move on trucks or rollers on its bottom, or it may be sus¬ pended from above and run upon a fixed or a swing bridge. The single-leaf swinging gate possesses the merit of reliability and of easily analyzed stresses, but is not economical, is slow of move¬ ment, requires much more power to operate it, and when used for a lower gate it reduces greatly the effective length of the lock. The tumble gate has the advantage of easily analyzed stresses, simnle masonry, and for low lifts it is quickly moved and is eco- DEEP WATERWAYS. 151 nomical of masonry, but it requires a pit in the bottom of the lock which is not easily cleaned. It is more particularly adapted for upper gates in the case of high-lift locks. When used as a lower gate it decreases too much the effective length of the lock. It has been used abroad and upon the Erie Canal in this country. The advantages claimed for the quadrant gate are that it is quick of operation, and gives a maximum effective length of lock with minimum cost of masonry. On the other hand, it is as yet untried, and, considered with respect to its stresses, it is unscientific, since its reactions on the side walls are in a direction to increase the bending movement upon the horizontal girders, and, unlike all other forms of single-leaf gate, its reaction has a component normal to the side wall. The bascule gate has been hardly more than suggested. It has no apparent advantages over the other forms of gate. Designs and careful estimates were made of a single-leaf swinging gate, a single-leaf sliding gate, and a mitering gate, all for the same location, and it was found that they did not differ materially in weight. The form of gate which would actually cost least will not be the same in different cases, but will vary with the width and depth of opening and lift of lock and with local conditions, but the difference of cost between any two of these forms of gate will not be great. Speed of operation, reliability, and immunity from accident are more important considerations than first cost. The mitering gate lias stood the most rigid test, the test of time, and is practically the standard form of gate for locks. Its use shows no sign of abatement, and, as all the forms of single-leaf gates are more or less untried for large locks, it was deemed best to make the miter¬ ing gate the basis of estimate for the present work. MATERIAL. The choice of material for the gates lay between wood and steel. Wood has been used almost exclusively for small gates, and for large gates it has had its full share of favor. In this country, where timber is abundant, it has been used and has given excellent satisfaction, but like the wooden bridge, its use is on the decline, and the more recent large gates have been built of steel. In England, however, although timber for gates has to be imported, it is still used quite as much as steel, even for the largest gates. A case in hand is the Manchester Ship Canal, which has all its gates built of green heart timber. The advantages of timber for gates are its lightness when sub¬ merged, its ease of repair in case of accident, and in many localities its low first cost. On the other hand, the life of a timber gate is com¬ paratively short. It begins usually to show weakness after from ten to fifteen years, involving repairs, and requiring to be renewed after from fifteen to twenty years. 152 DEEP WATERWAYS. Steel for gates is more in favor than wood in this country, and is preferred about equally in England, while on the Continent of Europe it is used exclusively. When designed to admit of inspection and painting, steel gates are much more enduring. Some of the early examples show little rust¬ ing after thirty years. The chance of accident is not believed to be great enough to war¬ rant the use of wood on account of its ease of repair. In many of the proposed locks for the deep waterways the lift is so great that it would be almost impossible to provide the necessary strength in a wooden gate. For these reasons, and because steel is, structurally, a much better material than wood, steel gates have been designed for all the locks. 1 RISE OF SILL. The principal functions of the two leaves of the mitering gate when closed is to form an arch, which takes the water pressure and transfers it to the side walls. What the rise of this arch shall be is a question of some importance. The weight of the gate depends somewhat upon the rise of sill, but, as will be seen from the tables given under the head of “ shape of horizontal frames,” only to a slight extent. There are more impor¬ tant practical considerations which decide the question. The less the rise of sill the longer will be the time required to open and close the gate, and the greater will be the thrust upon the masonry and the effect of any change in the length of the gate due to change of temperature. On the other hand, the greater the rise of sill the longer the gate and consequently the shorter the effective length of the lock, and if the gate be economically designed the deeper will be the gate and gate recesses. The choice of rise of sill, then, becomes a compromise between the above advantages and disadvantages. Inasmuch as a rise of one- fifth the width of lock seems to best satisfy the practical considera¬ tions and also to require a somewhat lighter gate than any other, it has been used for all the gates designed. AIR CHAMBERS. If the gate is not otherwise supported the pivot must, carry the total weight of the gate less whatever flotation it may have, and, besides, both pivot and upper hinge sustain horizontal reactions of consider¬ able magnitude. Two methods have been followed to relieve the pivot upper hinge. The first is to place under the gate a roller, which runs upon a track in the bottom of the lock and carries part of the weight of the gate. This device has been used with timber gates quite extensively, espe- 'Franzius. Der Wasserbau, p. 113 (Handbucli der Baukunde Abtk III). DEEP WATERWAYS. 153 cially in England, but the custom is being discontinued. In order that the rollers shall work perfectly the axis of rotation of the gate must be perpendicular to the plane of the roller’s track. Any inex¬ actness in this plane or settling of the track may make the roller useless. Wearing of the roller bearing may do the same. The second method of reducing the reactions at the upper-hinge pivot is to place a water-tight sheathing on both sides of the gate, forming an air chamber, which supports the gate by buoyancy. This method is quite generally used. It has, however, many serious dis¬ advantages. As gates with sheathing on both sides have usually been built, their interiors have been very difficult to inspect, clean, and paint, on account of the too confined space for efficient work and especially the difficulty of ventilation while painting. The result lias been that many inclosed gates have been left without painting of their interiors during their entire life, which was materially shortened thereby. The gates sheathed on both sides are more expensive because the shop work and assembling of parts is more difficult; there are more joints to calk and the calking of joints must be done with greater care, and because the extra sheathing, especially when horizontal frames straight on the downstream side are used, adds weight and little or no strength. The pivot and upper hinge must be made strong enough to sustain the reaction of the gate when it is swung in air at times when the lock is pumped out, therefore the pivot can not lie made much smaller when double sheathing is used than when it is not. It would seem that the gate with a single sheathing is preferable in every way, provided the pivot is not overstrained and its size is not impracticable. All the gates designed, with the exception of a single case, have, as the drawings show, sheathing on one side only, leaving the interior open and allowing of inspection and painting at all times. This results in a very simple construction, which will cost little more than ordinary bridge work. The single case for which it was thought necessary to provide addi¬ tional buoyancy is that of the gates between the locks of the flights in the canal around Niagara Falls. Plate 71 shows a typical gate for this case. These gates withstand at times a head of water of 75 feet, and are unusually heavy in consequence. In these gates the sheath¬ ing on the downstream side is below the lower pool only, leaving all the upper part of the gate open. In the chamber or hold thus formed a system of manholes is pro¬ vided, which it is believed will partially overcome the principal disad¬ vantage, except that of greater first cost, of the double-skin gate. Four manholes are provided in all the horizontal frames in the air chamber. The manholes in the top and bottom of the chamber have 154 DEEP WATERWAYS. covers, the others are open. By removing the covers four shafts will be formed, in which it is expected a draft will be established, thus ventilating the interior and from which all parts of the chambers may be reached for cleaning and painting. QUOIN POST. The function of the quoin post is fourfold. It must act as a column in carrying the weight of the gate; it must act as a girder transversely to the gate and also in the plane of the lines of thrust of the hori¬ zontal frames and distribute the pressure along the quoin; and withal it must furnish a satisfactory closure between the gate and wall. The post as designed consists of a very thick web plate with four heavy flange angles. This forms the vertical girder acting trans¬ versely to the gate. To the flanges of this girder and lapping well upon the horizontal frames are riveted a heavy 36-inch plate on the upstream side and a 42-incli plate on the downstream side, which dis¬ tribute the thrust of the horizontal frames. It is deemed advisable to make the hollow quoin of stone and the bearing pieces of the quoin posts of wood, except in the heaviest gates, where metal is used. In most of the gates designed the thrust in the quoin is so great that the required bearing area demanded by the masonry is very large. A large bearing can be provided much more cheaply and satisfactorily by the use of wood than by steel for the bearing part of the post. Wood is easily shaped. It adjusts itself to any unevenness or errors of workmanship and is more easily repaired or replaced in case of accident or wear. As the drawings show, the wood is made cylindrical on the back to a radius of 15.5 inches, giving a projected width of bearing of 23 inches. The axis of rotation has 1 inch eccentricity from the center of the cylinder. Thus the wood is relieved from bearing as soon as the gate begins to be opened. The eccentricity being small, there is very little chance for any foreign substance to lodge between the wood and masonry. Some designers have maintained that the bearing surface should be flat instead of curved on account of the tendency of the curved bear¬ ing piece to split the quoin stones by its action as a wedge. This tend¬ ency may be entirely overcome by cutting the quoin stones so that the reactions at their backs will introduce a bending movement which shall nearly or quite balance that due to the thrust of the quoin post. Such a shape of quoin stones is shown on plate 68. Fortunately this is the most natural shape to give them. From the consideration of the quoin post itself the cylindrical form has every advantage. Much greater width of bearing can be provided by its use than with the flat form, since the width of a flat bearing is limited by the consideration that the center of rotation must lie on the upstream side of a perpendicular to the bearing plane erected at DEEP WATERWAYS. 155 its upstream edge. The principal advantage, however, of the cylin¬ drical bearing piece is the fact that the line of thrust must always pass through the center of curvature, while with a flat bearing piece the position of the line of thrust will always be in doubt, the only sure thing about it being that it has considerable variation, reducing the allowable average intensity of pressure upon the surface and causing an uncertainty and variation of the stresses in the horizontal frames. In cases where the pressure per vertical inch of quoin does not. exceed 9,000 pounds, it is proposed to use wood in the quoin post. As the width of bearing is 23 inches, the greatest intensity of pressure amounts to about 400 pounds per square inch between the wood and stone. In some of the gates designed the thrust at the quoin passes this limit, as it was not thought practicable to make the timber bearing wider than 23 inches. A metal quoin and metal bearing pieces on the quoin post are used for these heavy pressures. This is shown in detail in figs. 7, 9, and 10, plate 74. The post itself is the same as before, except the steel-casting bearing pieces. This metal quoin is intended for use on all gates for the 80-foot locks, which withstand a head of water of more than 30 feet, and on gates for 60-foot locks, which withstand a head of more than 40 feet. All the castings are of steel. No eccentricity is used, as in the case of wooden bearing pieces, the quoin and quoin post being in contact at all times. This eliminates all danger of foreign substances lodging between the bearing surfaces. It may appear at first sight that this arrangement will cause a large frictional resistance in the quoin. These resistances will not be much greater without than with eccentricity in the quoin, since the horizontal thrust, due to the weight of the gate, must be taken care of either in the quoin itself or concentrated upon a bearing at the bottom of the gate. The lever arm of the frictional resistance of this bearing for the very heavy gates could not be made much shorter than the radius of the quoin, and, as friction is independent of area, little would be gained and much lost by providing an eccentric bearing. The only gain would be the slight reduction of the horizontal thrust of the gate while turning, on account of the slightly lower posit ion of its center of resistance and the fact that the smaller bearing would keep cleaner and have a somewhat lower coefficient of friction. However, even assuming a very high coefficient of friction, less than one-half horsepower will be required to do the work of overcoming friction in the heaviest gate. MITER POST. The functions of the miter post are practically the same as those of the quoin post, the only difference being that it bears against 156 DEEP WATERWAYS. another similar post instead of against a hollow quoin and does not carry the weight of the gate. As designed, it is structurally the same, the only difference being in the shape of the bearing pieces, which are of the same material in the quoin and miter post in all gates. Details of the posts having timber bearing pieces are shown on pi. 70, and those with steel bearing blocks are shown on plate 74. The drawings explain themselves without further description. Where timber is shown it is proposed to use white oak. The practice in the past has been to use wood exclusively for the meeting faces of the gate. In a few of the more modern gates, how¬ ever, steel is used. Wood has been preferred to steel because it makes a more water-tight joint on account of its more yielding nature. This is certainly true in cases where the thrust of the gate is not great. It is probable, however, that if the intensity of pressure between the steel bearing pieces be high and the parts carefully made a very good joint will result. In the present work the use of steel bearing blocks in some cases was imperative, since the thrusts are so great that a bearing 3 feet wide would be required if timber were used. Considerable time has been given to the study of the position of the center of pressure at the meeting faces, since upon it depend quite largely the stresses in the horizontal frames. The meeting faces may be made plane or curved. The former is the more usual shape and is entirely satisfactory when the thrusts are moderate enough to allow of a comparatively narrow bearing face being used; but when very wide bearing is required, plane meeting faces give trouble by “nipping” or meeting at their extreme edges. If the faces of the bearing pieces are planes, the most favorable con¬ dition is a uniform pressure over their entire surface. This ideal condition can not be realized in practice, as it can only occur for one definite length of the gate leaves, and then only if the angle at which the meeting pieces are fitted is absolutely correct. Since the length of the gate varies with the temperature and the stress to which the structure is subjected, and, furthermore, absolute exactness either in the shape of the gate or the fitting of the miter pieces is impossible, a strictly uniform pressure will never occur. When gates are being closed and the two leaves approach each other, the bearing pieces will first meet on their extreme upstream or down¬ stream edges, the faces making a very small angle with each other. As the water pressure comes upon the gate the material compresses, and from being in contact at a line along the extreme edge they come in contact along a vertical strip extending from the edge of the meet¬ ing faces toward their middle, and in most cases bringing their entire surface into contact. The position of the center of thrust depends upon the variation of intensity of pressure across this face. Just what is the law of variation of intensity of pressure across the face of the timber bearing can not be told, as the wood will in most JULIUS BIEN S CO PHOTO LITH- H Doc 149 56 2 DEEP WATERWAYS. 157 cases be pressed beyond its elastic limit. As close an approximation as can be made is to assume that distortion is proportional to stress. In fig. 1 and fig. 3 Let a— one-lialf the angle between the meeting faces, i=thickness of the timber bearing, C x =compression of timber at its upper edge, C 2 =compression of timber at its lower edge, p^= intensity of pressure on timber at its upper edge, p 2 =intensity of pressure on timber at its lower edge, A=one-half width of bearing face, E 2 =modulus of elasticity of timber, P= thrust per vertical inch of miter, Then, in fig. 1, which represents the case of contact over the entire face, p — P\t p — Pit or ^2 — C'l —p (Pi Pi) but C 2 —C,=2 A tan a from which: 2 A E., tan a Pi~Pi= - \ - again: p=a (iJj+ib) Eliminating p,= P 2A A E, tan a , --and p x P A E., tan a ¥A r ( 1 ) Center of resistance from middle=e=A — ~ ~ ( w Til 3 {Pi+lh) In fig. 2, which represents the case of partial contact: But By elimination, C »=d tan a E 2 C 2 E 2 d Pi ~~ ^ ~ == “ tan a ~r pi — 2 P d d =Va P t E 2 tan a Pi 2 =v» P E 2 tan a ( 2 ) (3) e=A— iJ g p t 3 v Eo tan c ( 1 ) And, 158 DEEP WATERWAYS. Equations 1 and 2 apply when 9 ,= / 2 P t < Y E 2 tan a and equations 3 and 4 apply when 2 A 2 P t > y E 2 tan a The same results would be obtained in case of first contact at up¬ stream edge. The unknown factors in these equations are a and E 2 . The angle a may be said to be the result of error in workmanship or the result of a change in the length of the gate. To determine a we turn to fig. 3. Let qoi represent a line of length L drawn from the hollow quoins to one edge of the bearing piece in the miter post. Let ft = angle 0 \ q. v. Let 02 represent the new position of the edge of the bearing piece after change of length of gate. Let qr = qo 2 . Let d L— change in length of gate. Let a — angular change in plane of bearing face due to change in length of gate. Let d]L= change of length of gate due to change of temperature. Let d 2 L = change in length of gate due to compression of steel. Let d 3 L = change in length of gate due to compression of wood. Let K = the angle between the meeting faces due to error of work¬ manship, when the gates are first adjusted. Let S = intensity of compression in the steel horizontal frames. Let T = total thickness of timber in quoin and miter post. Let Ei = modulus of elasticity of steel. Let P = thrust per vertical inch of miter post. Then in circular measure: a= dt diL —d 2 L —AX L tan ft + d t L L tan ft SL Ei L tan ft PT - 2AE 2 ±K L tan /i ±lv = ± d L L_S_ PT L tan fi Ej tan ft 2AE 2 L tan ft ±K (') In this formula the first term represents that part of the angle caused by change of temperature, the second that caused by compres¬ sion of steel, the third that caused by compression of the wood, and the fourth that caused by error of workmanship. DEEP WATERWAYS. 159 If the gate be supposed to be subjected to a range of temperature of from 32° to 112° F., then the greatest change from mean temper¬ ature will be 40° F., and will equal .0000065 x40L=.00026L. 8=9,000 pounds per square inch for all frames, Ej=29000000 A=114 inches L=53G inches in gate for 80-foot lock, T = 22 inches. As for Iv, let it be supposed that, owing to error of workmanship, the bearing blocks are set so that when the gates are closed they bear on one side, and the edges of the other side are separated by one- eighth inch. This would be a very noticeable error. Then tan K=zb -.~ q =4;.0026 for gate for 80-foot lock. it) ✓N -J By substituting the constants in equation (7) the value of oc is obtained in terms of E 2 and P. As a is very small, we may put tan a=a and substitute in equations (1) and (2) or (3) and (4), which will give the intensity of pressure in the wood (jq and ji,) and the eccentricity of pressure (e) in terms of E 2 and P. P varies from nearly 0, at the top of the gate, to a maximum not greater than 10,000 pounds per vertical inch of miter post at the bot¬ tom of the gate or below the lower pool. E 2 is quite largely a function of the condition of the timber as regards moisture. It ranges from 20,000 to 200,000 for white oak. As low values as 20,000 were obtained from measurements of the com¬ pression of the timber in the miter posts of the Poe Lock at Sault Sainte Marie. These measurements gave no values higher than 40,000. The .timber in these gates had been in use about three years and was well saturated. As high values as 200,000 have been obtained from laboratory experiments upon well-seasoned wood. Substituting these two extreme values of E 2 in connection with various values of P in the proper equations, either 1 and 2 or 3 and 4 as determined by equations (5) and (6), the following values e and jq are obtained: E, = 20,000. p e Pi P e P». 1,000 6.9 inches. 130 pounds per square inch. 6,000 2.0 inches. 377 pounds per square inch. 2,000 4.9 inches. 188 pounds per square inch. 7,000 1.8 inches. 424 pounds per square inch. 3,000 3.6 inches. 237 pounds per square inch. 8,000 1.6 inches. 469 pounds per square inch. 4,000 2.8 inches. 283 pounds per square inch. 9,000 1.5 inches. 517 pounds per square inch. 5,000 2.3 inches. 330 pounds per square inch. 10,000 1.4 inches. 564 pounds per square inch. 160 DEEP WATERWAYS. In the same way for E.,= 200,000: p e Pi 1,000 3,000 6,000 10,000 10.3 inches. 9.2 iuches. 8.0 inches. 6.9 inches. 400 pounds per square inch. 702 pounds per square inch. 1.000 pounds per square inch. 1,300 pounds per square inch. The first table in which E 2 = 20,000 probably represents the most favorable condition, so far as e is concerned, that can be expected. Under these conditions the maximum distance of the center of pressure from the center of the bearing piece is, as the table shows, 7 inches at the top of the gate, gradually decreasing to less than 2 inches below the lower pool. It must be remembered that, so far as the steel horizontal frames are concerned, great variation of the center of pressure in the upper part of the gate is quite as objectionable as in the lower part, since, although the intensity of pressure on the wood is not so great, the horizontal frames, being spaced farther apart, have as high a stress as those below. The excessive values of e in the upper part of the gate may be elimi¬ nated by making the bearing face vary in width from a minimum of, say, 0 inches at the top of the gate to the full width at the surface of the lower pool. Fig. 4 shows the plan and face elevation of such a bearing piece; a, b , c, e, /, h represents a top section and a, b , cl, g a bottom section. Under less favorable conditions when the wood is dry, as seen in the second table, in which E = 200,000, the distance of the center of pressure from the center of the bearing face may vary from 10.3 inches at the top of a 24-incli bearing piece to 7 inches at the bottom. If the designer is convinced that the bearing blocks will be thor¬ oughly saturated at all times, and if he will proportion them so that the pressure per vertical inch of miter, divided by the width of bear¬ ing face in inches, shall be about 400, thus giving to them the shape indicated above, he may use bearing blocks with plane meeting faces with little fear that the center of pressure will be more than, say, a tenth the width of face from its center. If he- is not sure of this, he must make whatever assumptions seem to him reasonable. No formula can give exact values of e, since the condition of the timber changes from time to time, becoming softer with age and saturation, and the foregoing analysis is not given for the purpose of obtaining exact values, but to assist in making more intelligent assumptions. The analysis is of value in showing what the shape of the bearing pieces should be and indicating what effect upon the strains in the gate the use of wood in the quoin and miter post has. JULIUS BIEN & CO PHOTO LITH H Doc 149 56 2 - - . ’ t JULIUS B1EN 4 CO. PHOTO LlTH H Doc 149 56 2 DEEP WATERWAYS. 161 The deduction from the above analysis is that with a plane bearing face 24 inches wide there is a wide range (from 0 to ± 8 or 9 inches) of eccentricity possible, depending upon the condition of the wood, and that the uncertainty is so great that it is very desirable to elim¬ inate it if possible by the use of another shape of bearing block. As the following analysis shows, by giving the bearing blocks a cyl¬ indrical form of large radius, there will be much greater certainty that the range of the center of pressure will be confined to narrower limits. As before, the position of the center of pressure depends upon change in the length of the gate and accuracy of workmanship. In Fig. 5 Let dL = change in length of gate. Let q be the back of hollow quoin. Let yd = angle Oi qv. Let r = radius of the bearing pieces. Let Ci be the center of the curve of the face of the bearing piece before change of length of gate. Let C 3 be the center of the curve of the face of the bearing piece after change of length of gate. Let w = distance between hollow quoins. Let. Oj represent the point at which the gates will touch before change of length. This is supposed to be the middle of the face of the bearing block. Let 0 3 represent the point at which the gates will touch after change of length. I Let e = distance from center of bearing block to center of pressure. LetL = length of gate = qOi. Suppose that a change of length dL takes place without revolution of the gate, and Oi passes to a position 0 2 , and Ci passes to a position C 2 . Then CiC 2 will be parallel to q0 4 . Suppose, then, that revolution of the gate takes place until the gates meet at 0 3 , Then tan u, = i w tan yd — dL sin yd y w— r— dL cos yd i iv — r tan fd —dL sin yd) s -\- {\w — r — dL cos ft) 2 e—r tan («j—ff 2 )4r tan K. ( 8 ) J - (9) LI. Doc. 149-11 162 DEEP WATERWAYS. In which: t = one-half extreme range of temperature in degrees Fahrenheit. E t = modulus of elasticity of steel. E 8 =modulus of elasticity of wood in quoin and miter post. T = total thickness of timber in quoin and miter post. P thrust per vertical inch of miter post. A half width of wood in quoin and miter post. L = length of gate. s = intensity of compression in horizontal frames. The first term in the value of dL represents the change due to change of temperature; the second, the change due to the compression of the steel, and the third, the change due to the compression of the wood. It is evident that dL will be greatest when P is greatest and E 2 is least. If the range of temperature be from 32° to 112° F., then t =40. L =536 in gate for 80-foot lock. Ej =29,000,000. P =from 0 to 10,000. E 2 =from 20,000 to 200,000. = 498 inches in 80-foot lock. ft =arc tan .4. K =an angle produced by an error of workmanship in making one side of the bearing piece thicker than the other. Supposing that one side of the bearing block be made one-sixteenth inch thicker than the other, K will be, arc tan Q . =0.0026. 16 X 24 Putt ing r = 300 inches and substituting the above value in the proper equations, using E 2 = 20,000 and P = 10,000, and using all minus signs, we get for a maximum e= — 1.8 inches. This is measured on the upstream side of the center of the bearing block. The value of e for the downstream side, obtained by making P small and E 2 large and making all signs plus, is 4- 0.7 inch. By comparing this maximum with 9 or 10 inches, found for bearing blocks with plane faces, it would seem that without doubt the curved bearing is the better one to use. As the drawings show, all the bearing blocks have been made 8 inches thick and cylindrical to a radius of 300 inches. The practice of those in charge of locks bears out the wisdom of this. It is their custom, when the gates give trouble by nipping, to slightly round the faces of the bearing blocks, which expedient has been found to remove the trouble. DEEP WATERWAYS. 163 To determine the intensity of pressure upon the timber, let fig. 6 represent a bearing piece compressed to any line, as f g. Let s be any point on the compressed surface. Let x = its distance from c 0 3 . Let ()., be the point of first contact. Let P = pressure per vertical inch of miter post. Let p = intensity of pressure at s. Let r = radius of face-bearing piece. Let b = the breadth of the surface in contact. Let t = thickness of bearing piece (if r is large, t may be taken as uniform). Let d = amount of compression at s. Let E 2 = modulus of elasticity for wood. Then d = Jr 2 — x 2 — Vr a — jb 2 ; also d = j(- P'a From which p is a maximum at point x = O Then p (max) ( 11 ) By using the maximum value of 10,000 pounds for P and making r = 300 inches and t — 8 inches it was found from equations 10 and 11 that the maximum value of p for E 2 = 20,000 is 830 pounds per square inch, and for E 2 = 200,000 it is 1,500 pounds per square inch. STEEL BEARING PIECES. Very much more favorable conditions as regards eccentricity of pressure prevail when steel bearing blocks are used in the quoin and miter posts. Very little uncertainty exists in regard to the position of the center of pressure, and it is quite certain that its variation will be small. The steel meeting faces are curved to a radius of 300 inches. The last term of equation (9) vanishes, leaving d L = ± 0.0000065 t L — s ' L E. 164 DEEP WATERWAYS. All the factors of this equation are definite and well known, and dL is small under the worst conditions of change of temperature which are likely to exist. Assuming the same error of workmanship and substituting as before in equations (8) and (9) we get e x — 1.2 inches and h or Te, = 02.5 b li i e x and Te 2 =02.5 b h t e 2 In which Te, and Te 2 are thrusts at any section of any frame. By inspection it is seen that the upper flange will be most strained near its middle and when the line of thrust is in its upper position, and that the lower flange will be subject to reversal of stress, maximum tension occurring at its middle and maximum compression near its ends. In all the gates designed the maximum tension in the lower flanges is very small. ’ ' . I i Fig. 9. F 1 - 1 —~ -1 i i i i i i i 1 ! 1 1 1 I 1 1 1 1 1 » l 1 !. i Fig JO. r JULIUS BICN i CO PHOTO IITH. H Doc 149 56 2 DEEP WATERWAYS 171 By reference to plates 70 to 74 it is seen that the lower flanges are heavily reenforced near their ends, and a few trials show that the crit¬ ical section for the lower flange is Y Y, taken near the end of the reen¬ forcing plates. In fig. 9, Let T be the thrust at any section. Let x be the distance from the center of gravity of the section to the center of thrust. Let C , and Cbe the distance from the edges to the center of gravity of the section. Let/, = stress in upper edge, and Let f 2 = stress in lower edge. Let A be the area of the section. Let I be the moment of inertia of the section. Then, T y-C T / 1 = [ + 7 .(13) and /, t _ .x-cyr A 1 (14) By assuming various flange sections and making a series of trials by the use of the above formula, flange sections may be found which will give satisfactory values for/, and/.,. This method was found very tedious, as the flanges must satisfy the conditions at each of the two different sections, and especially in the investigations of the relative economy of different shapes. In these cases it was necessary to so proportion the frames that the maximum stress should be practically the same in the various cases compared; otherwise the comparison of weights would be of little value. A more direct method was sought, with the result that the follow¬ ing formula? were developed: Let X X, fig. 10, be the critical section for the upper flange. Let Y Y, fig. 11, be the critical section for the lower flange. Let T, be the thrust acting at section X X. Let a be its greatest distance from the lower edge of the web. Let d x be the depth of the frame at X X. Let T 2 be the thrust acting at Y Y. Let b be its greatest distance from the upper edge of the web. Let d> be the depth of the frame at Y Y. Let Fj be the area of the upper flange. Let F 2 be the area of the lower flange. Let/ be the stress in the upper flange at section X X. Let/ be the stress in the lower flange at section X X. Let/, be the stress in the upper flange at section Y Y. Let/, be the stress in the lower flange at section Y Y. DEEP WATERWAYS. *7 O < 2 Then in section X X, if it be assumed that the stress varies uni¬ formly from f. at the lower edge to f x at the upper edge, the resistance of the web will be ( ]i t (/i 4-/3) 9 and tin* distance of its center of resist¬ ance from its lower edge will be f/ i ( 2 /i + fs) 3 (A 4-/3) If we assume the center of gravity of the flanges to be at the edge of the web, the resistance of flanges may be taken to be Fj f\ and F 2 / 8 , and by taking moments above the lower edge we have: Also for equilibrium: T 1 =/ 1 F 1 +/ 3 F 2 + ( h 1 [fi + ./;;) (15) (16) Making these two equations simultaneous, any two of its factors may be found, the rest being known. In like manner, equations for Sections Y Y, fig. 11, may be formed. They are: 5t 2 =cZ 2 / 2 F2+ ^M+^) T,=/ 2 F 2 +F 1 / 4 4- t ifz+fi) (17) (18) The conditions of economical proportioning are that F, and F a shall be such that f x at Section X X shall have the value of the maxi¬ mum allowable intensity of pressure when T el is acting, and/ 2 at Sec¬ tion Y Y shall have the maximum allowable intensity of compression when T,., is acting. Eliminating f 3 from equations 15 and 16, we have: 8aT t fjdi 6aT t d t 2 of , v\+ dit Eliminating/ 4 from equations 17 and 18, we have: sbh_m_ T _o f¥ d, 2 1 - 2 66T 2 BfiF, ^ dMT dd (19) ( 20 ) If the flanges do not change in section between X X and Y Y, then equations (19) and (20) are simultaneous in F, and F 2 . The web is made sufficient to resist all shearing strains, which con¬ sideration may determine /. although for practical reasons in the DEEP WATERWAYS. 173 present work the webs were made much heavier than this requirement would demand. For any particular case a, b, d u d 2 , r l\, and T a are known from the stress diagram. If their values and the maximum allowable inten¬ sity of compression be substituted for f\ and/ 2 , equations 11) and 20 become quite simple and may readily be solved for F, and F 2 . This method is direct, and it was found to require very much less time than the tentative process. In many frames of the shape used the lower flange was determined by the minimum allowable section. By substituting F 2 in equation 19, Fj may be found at once. After Fj and F 2 are found by equations (19) and (20), they may be readily checked by equations (13) and (14). This was done in all cases. The lower or downstream flange is relatively light, and consists of two angles reenforced at their ends. The section required for the upstream flange is in most cases large. It consists of two heavy angles, a certain amount of the sheathing, and a cover plate, which acts also as a splice plate for the sheathing. What part the sheathing plates play in resisting flange strain is in doubt. Some designers neglect its action on account of the doubt and others would consider the entire sheathing as part of the flanges. That part of the sheathing plate which is connected to the flange angles must certainly act with the rest of the flange in resisting strain, since any change of length of the angles must cause a like change in that part of the plate which is attached to the angles. The uncertainty lies in the inability to determine just how far away from the flange this action extends. In the present work the flange section has been determined for the low-unit stress of 10,000 pounds per square inch, and when two 6-inch angles are used a strip of sheathing 16 inches wide has been counted as part of the flange section, and when two 4-inch angles are used a strip of sheathing 12 inches wide was counted. The probability is that the sheathing has much greater effect than has been assumed, reducing the maximum stress to, say, seven or eight thousand pounds per square inch. On the other hand, even should the sheathing not act at all the stress in the flanges would not be excessive. VERTICAL FRAMING. The vertical system of framing consists of the quoin and miter posts already described and two systems of vertical frames built in between the horizontals. They are shown in tig. 4, plate 70. These frames, when combined with heavy vertical plates, which extend from top to bottom on each side of the gate, form girders of considerable strength. They are not, however, introduced for the purpose of carrying any part of the load, but to stiffen the whole structure. 174 DEEP WATERWAYS. In the middle of the gates for the 80-foot locks, a system of light frames is introduced to stiffen the flanges and webs of the horizontal frames. EFFECT OF VERTICAL STIFFNESS. The ideal condition of distribution of loads among the frames of the gate would be for each horizontal frame to resist the water pressure which falls upon it. This condition can not exist, however, so long as the gate has any vertical st iffness and bears against the sill. It is impracticable to build a gate with no vertical stiffness, and in fact impossible if curved skin plates are used, for the curved sheath¬ ing forms a strong vertical girder especially as it is stiffened very frequently. This sheathing, when combined with the quoin and miter posts and the other vertical framing necessary, makes the gate a vertical girder, which has the effect of changing the loading of the horizontal frames very decidedly from that due to direct water pressure. This may be seen from the following illustration: Let fig. 12 be a transverse section of a gate taken at the middle of one leaf, acted upon bj' a head of water II + 11 on one side and D on the other. If O be the origin of coordinates, the axis of X being hor¬ izontal, the abscissas of the line abc will represent the intensities of water pressure at any point on of. If each of the horizontal frames were proportioned to resist with the same stress simply the water pressure that falls upon it, and there were no contact at the sill, then the whole section would deflect to a new position A'O' parallel to AO. If, however, it is held at the sill and has vertical strength, it will take the position A' O. The horizontal frames near the bottom of the gate will be relieved of a part of their load, some of which will be carried to the sill and the remainder to the upper horizontal frames, which will now be over¬ loaded. To meet this overloading and also for practical reasons, the horizontal frames in the upper part of the gate must be made stronger than would be required to resist simply the load of water pressure which comes upon them. To determine the distribution of the loads upon the horizontal frames is important. The problem, however, is somewhat complicated since it involves the elasticity of every part of the gate and is therefore not soluble by the equations of statics, and for the same reason the solu¬ tion can not precede the design, but must follow and be a check upon it. The determination of the distribution of the loads among the hori¬ zontal frames answers the two questions that are of prime importance in connection with the subject. First. Are the horizontal frames properly proportioned? (The dan¬ ger exists that the upper frames will be overloaded when the gate bears against the sill, and that the lower frames will be overloaded if the gate does not touch the sill.) /■ // jL A. A. JULIUS BIEN « CO PHOTO. UTH H Doc 149 56 2 ' DEEP WATERWAYS. 175 Second. Will the deflection of the gate be such as to overstrain the vertical system? The earliest record that we have of an attempt to solve this problem is the Annales des Fonts et Chaussees for 1850, in which M. Chevalier gives an account of a series of experiments upon wooden models. From the results of these experiments he drew the conclusion that the horizontal frames should be made of equal strength and be equally spaced. This rule has been quite generally followed by French engi¬ neers to the present time, and it seems to be a very good one for most of the gates that they have had to deal with, namely, tidal gates, and gates for locks, the lift of which is small. In 18(17 M. Lavoinne published in Annales des Ponts et Chaussees a paper on the subject, in which he gave a very complicated analysis of the effect of vertical stiffness upon the distribution of loads among the horizontal frames. In 1886 M. Galliot published also in Annales des Fonts et Chaussees the results of his investigations of the same subject. lie worked upon the same lines as Lavoinne and obtained formulae which give practi¬ cally the same results. Both Lavoinne and Galliot made their analyses upon the supposition that the horizontal frames are of equal strength and equally spaced and that the gate is of the same material through¬ out. In their analyses they were obliged to make numerous other assumptions and approximations in order to simplify the work. The condition of equal strength and equal spacing of horizontal frames is far from existing in the gates which have been designed for the locks of the deep waterways. Neither are the gates of the same material throughout. The fact that timber has been used in the quoin and miter posts is a very important factor in the problem. These facts make the formulae of Lavoinne and Galliot entirely inapplicable to the case in hand. Captain Hodges, in his book “Notes on Mitering Lock Gates,” has developed an excellent practical rule for use in proportioning the horizontal frames. It is, however, purely empirical. Other than the cases just mentioned, no record has been found of any solutions of this problem. As none of the solutions mentioned above are entirely satisfactory, and the importance of the problem is such that it can not well be neglected, considerable attention has been given to the developing of a method of treating the subject. Numerous measurements were made of the deflection of the steel gates of the Poe lock at Sault Ste. Marie as a check on the results of the theoretical analysis. The method of treating the problem which has been developed is believed to be theoretically correct, and its results agree very closely with the results of the measurements taken on the Poe locks gates. The framing of the gate consists of a series of horizontal frames DEEP WATERWAYS. 17(5 and practically a single vertical girder covering the entire gate. This vertical girder consists of Ihe quoin and miter posts, two vertical frames, and the curved skin plates. All of these taken together form a girder of considerable stiffness. This vertical girder may be considered to have a moment of inertia equal to the moment of inertia of all these parts considered as a single piece, so long as no buckling takes place in the skin plates. A certain amount of buckling will no doubt occur, but as the skin plates are stiffened very frequently they will develop a very large percentage of their full strength. In the following analysis Ihe assumption is made that no bending occurs in the horizontal frames. As a matter of fact, bending will nearly always occur, even when the horizontal frames are circular arches, and it will always occur in the frame shown in the adopted designs. The deflection of the gate due to bending of the horizontal frames is, however, insignificant compared to that caused by short¬ ening of the frames under compression. As an example, consider the horizontal frame next to the bottom of the upper gate for the lock at Oriskany, N. Y., shown on plate 72. The deflection of the miter post, due to shortening of the gate, is l.G inches, according to equation (21). The positive bending moment under normal conditions takes place along a length of about 15 feet in the middle of the horizontal frame. Outside this range the line of thrust lies either on or below the center of gravity of the sections. Computing the maximum deflections of this same horizontal frame, caused by bending, we get at the middle only 0.09 inch, which is small in comparison with 1.6 inches, due to shortening of the gate. The strength of the vertical girder is distributed by the skin plates nearly uniformly over the entire width of the leaf and may be taken quite so. Under the assumption that no bending occurs in the hori¬ zontal frame, the deflection of any frame will vary uniformly from 0 at the quoin post to a maximum at the miter post. The resistance, after deflection of a vertical beam of uniform strength and a width equal to the width of the gate, will also vary in the same way. The center of resistance will therefore be at a point two-thirds of the distance from the quoin post to the miter post. The magnitude of this resistance will be the force required to deflect the whole beam one-half the deflection at the miter post. Fig. i:> represents an isometric projection of the frames of a miter¬ ing gate. qq represents the hollow quoins. trim represents the center line of the meeting faces. F 0 Fj, etc., represent the horizontal frames. — A.- X.- • ( 25 ) or- A 5 - Iv The condition of the sill contact determines the value of A 5 . If there is no sill contact the first value of A R is to be used, or— A 5 = X. R ^5 ^ X- X 5 . If the gate bears against the sill, then the deflection is no longer a function of the strength of the lower horizontal frame, but is a cer¬ tain amount (K), which may usually be taken as zero. When the gate bears against the sill, X. will be the sill reaction. We may express X 0 and X 5 in terms of X t , X.,, X 8 , and X 4 . By moments about the bottom of the gate, X y = — (I X, 4- 4 X., 4- 2 X 3 + X 4 ) . . and about top of gate, 1 X H = — 5 (X, 4- 2 X, -f 3 X 3 4- 4 X 4 ) . . Equations ('25) then become: A o = A„ P 0 4- Yg X„ (-1 Xj 4- 3 X, + 2 X 3 4- x 4 ) A i = A t P 4 g A, X, (27) or- A 2 - X, P;J g A., X 2 A 3 = A 3 P 3 -g A 3 X 3 4 A 4 = A 4 P 4 — g a 4 x 4 A r. = A 5 P 5 4~ Y 5 X 5 (X 4 4- 2 X. 4 -f 3 X 8 -f 1 X 4 ) A. = K. (28) Let fig. 14 represent a transverse section of the gate taken on the center line of the vertical girder, in which g g represents the center line of the vertical girder before deflection and the curved line g' g' represents the same line after deflection takes place. DEEP WATERWAYS. 179 Let A 0 Aj, etc., v y v 2 , etc., be as indicated in the figure. Then 4 1 — A | £ A 0 g A 5 3 2 v 2 —A a g A„ g A 5 2 3 ^3—^3 5^0 A 5 1 . 4 V 4— A 4 fj A " g ^ 5 Substituting the values of A given by equations (28) into the above equations, we have for gate without contact at sill: /4 4 64 t 4 4 \ /48 4 8 . \ /32 . —^i—( 3 A i+ygA 0 + 75 A 5 JX, + ( j 5 A„ + ^ 5 A 5 jX.,+ (^ 75 A„ + 75^5^)^3+(^75^0+-^A 3 ) X 4 -A 1 P l +gA 0 P„+gA 5 P B /48 . , 8 A X v , A4 A 36 4 16 . X v . /24 , v * “ V75 Ao+ 75 A V Xi+ \3 A 2 + 75 A °+75 A V^ 2 v75 A ° ' 3 “A.W 18 75~'° 3 '32 32 7gA 0 +7gA 5 j X 4 —A 2 P 2 -f-g A 0 P 0 +g A 5 P 5 12 . \ /24 . 24 . \ / 4 . 16 V * V75 Ao "*"75 A 5 ) Xi + V75 A(,+ 75 Afi ) X ‘ i+ ( ‘S A;,+ 75 A " + 75 A s )x 3i + ( T^An + Tg A 5 ) X 4 —A 3 P 3 +g"A c ,P„+g'A 5 P 5 (29) — v A 5 48 16 . 16 \ /T2 32 \ f 8 48 . \ 7gA 0 +^AgJX 1 +(^7gA 0 +^A 5 )x 2 +(^7gA ( ,+7gA 5 ) X 3 +( g A 4 +7gA 0 +^A s ) X 4 —A 4 P 4 -|-g A 0 P 0 +gA 5 P b And for contact at sill: '4 . .64 , . 48 ■Vi — — ( ,r A, + ~ A () ^X,4-~ A 0 X 2 +— A 0 X 3 +,__ A 0 X 4 — A t P t + 4 -a,,p„+‘k 48 /4 36 \ 24 12 l 'i 75 -XoXt"i”( — A 2 +^-A 0 j X 2 +—A 0 X 3 +— A 0 X 4 A 2 P 2 -(- 32 y /i / II) \ ^ =75 A(>Xi+^g A 0 X a 4-( g A 3+ 75 A 0 ) X 3 + — A 0 X 4 — A 3 P 3 + 5 A » P o+| K —- v 4 = ^ A oX t +^r AoX 2 + — A 0 X 3 4-( y A 4 + __ A 0 ) X 4 —A 4 P 4 + 3 VO /o 75 —r 24 5 A » P »+S K 16 ( 30 ) 75 75 75 L 5 5i A 0 P () +- K 180 DEEP WATERWAYS. The equations given above express the deflections i\ r 2 , etc., in terms of the loads and the properties of the horizontal frames. We will,now derive independently the value of the same deflections expressed in terms of the loads and the properties of the vertical girder. We will flrst demonstrate the following proposition: If a simple beam be acted upon by a series of transverse forces P, P 2 , etc., then in which M=the bending moment at any point. E =the modulus of elasticity of the material. I =tlie moment of inertia of the beam. P„ = any transverse force. v n = the movement of the point of application of the force P n with respect to a straight line joining any two points of the beam (usually points of support). v n must be measured in the direc¬ tion of the force P n . L = the length of the beam. In fig. 15: Let ^ p-z2hPil J 5pti represent the center line of a straight beam under the action of the forces Pj P 2 . P 8 . Let p x p i be considered fixed points. The position of the center line of the beam without load is a straight line drawn through p x p 4 . Let the movement of the points p x p 2 , etc., under the action of tne forces Pj P 2 , etc., be v x v 2 , etc. In passing from one condition of loading to another work is done. 'I’he total work of all the external forces acting upon the beam is— By the doctrine of the conservation of energy, this must be equaled by the internal work done in distorting the beam. This internal work is done in lengthening and compressing the material, and stored as potential energy in the material to be given back if the forces are removed and the beam springs batik to its initial position. In fig. 16: Let cc be an infinitesimal portion of length <1 L of the neutral axis of the beam under stress. Let n n' and m w' be normal sections. Let i be the angle n' r r'. Let y be iht* distance from the neutral axis to any point q on the sec¬ tion n n'. Fig*. 16 . JULIUS BIEN l CO PHOTO LITH H Doc 149 56 2 ■ . DEEP WATERWAYS. 181 Let f be the stress at point q. Let y x = c n. Let y- 2 = c ri. Before the beam was stressed n ri was parallel to m m', or in a position r r'. The average force acting upon unit area at q is — /, and the distance through which it acts is y i. Then neglecting the distortion caused by shear, which is practically zero, the work done on an area a at q is: 7 f a V i a w = : - - • The work upon the whole prism n ri vri m is: W-V Vi The total internal work in the whole beam is: L *—. y 2 ^ W V V Z »Mi Zj.„ ' 2 The total external work done upon the beam is: w = Pi V X + P 2 V t • • • Pn Vn . Equating external with internal work and clearing of fractions, we have: .v 1 — '■ Pi^i + P 2 ^2 • • • Pn Pn = J I fay i. o ~V\ Differentiating this equation with respect to P n , we have: L ^ — . y % ‘-S S df_ d P„ a y i. ~V\ But, Then, and , My f = I df _ d M y d P n dP n I v j — / d L _ M i/ d L J E El 182 DEEP WATERWAYS. 7 n Substituting the values of and y i we have (X x n k L — ^ ft V Z_J cZ M y a My d L _ (Z M M i7 —/A dP n I El (Z Pn E I 2 y% M 2 ' i i 2 a y 2 d L but We therefore have: ft Y, ay 2 =I -ft L fti = aKST* 1 QED - Turning now to the vertical girder of the gate, the forces acting on it are X Q X t X 2 , etc. Let Mj be the bending moment at any point in the top panel of the vertical girder. Let Mo M 3 , etc., be the moments in the second and third panels etc., Let y be the distance of any point in any panel from the horizontal ■ frame next below it. M 5 = y X, = - y (-Y + | X, + | X s + | X 4 ) d M„_ y d X x 5 d X~ 5 J d ^*y d X, d M, o 4 d X 4 y d M 5 M y dxfEI=25Er( X . + 2 X *+ 3 X *+ 4X *) X d Xj EI f ^“150 El ( 2 X i+ 4X 2+6X,+8X 4 ) y 2 Hflf=2#El( 2X .+ 4 'Y+ liX »+ 8X *) I" 50 KI 0 X, + 8 x 2 +12 X s + 16 X 4 ) DEEP WATERWAYS. 183 In the same way, Cci d M 5 M.. a 3 , J 0 d X 3 EI ( - y_ 150 El I a tZ M 5 M, a 3 . X 4 ) M 4 = (« + »/) X, - ~{a+y)X. i -^{a+y)X. i -^(4:a-y) X 4 d M 4 1, dx~ ?>( a +y) d M 4 2 Tx.r~ 5 {a + y) d M 3 , ar-5 (ZM 4 1 , dx=- 5 ^ a -y ] d M 4 M 4 _(a + -if) /y , •> y _L-1 V \_l 1 a 2 3 ay — y 2 v fZ Xj E I 25 E I + w As!+ ‘ 25 E I X < •a :!4 L ?n = OT < 14 - X ' +2S x *+« x »+» x <> In like manner, *a d M 4 M 4 7 — a / 0 £ y , Kg y I jjj y | />o y \ d X., E l ,l,J - BobTI A,+ '” A ' i+ 4 As+ x <> a 6 n (12 Xj+84 X a +126 X 3 +93 X 4 ) 150 E (Z ^ L4 /7 V = WET < 31 X i + 62 x 2 +93 x 3 +74 X 4 ) For the third panel, M,= ' 2 - a +V (X.-+2 X 2 )-f (3 a—y) X, - \ (3 a-y) X, O t) o ft' 5 150 E I ■37 X 4 ) or 150 E I 184 DEEP WATERWAYS. •a iy o jJA „ = 0 X s +226 X s +160 X 4 ) Equating these values of r, r 2 , etc., with those in group of equations (29) and (30), we have, after clearing of fractions, for no contact at sill: (200 A t +128 A 0 +8 A,+^^) X, + (96 A 0 +16 A S +^A°) X a +(64 A 0 + 24A 5 + ^- 3 )X 3 +(32 A 0 +32A 5 +^ 3 )X4-150A 1 P 1 + 120A 0 P o + 30 A 5 P b =0 O OK n 3 360oA (96A„+10A S +^) X, + (200 A a + 72 A„ + 32 A,+^) X, 225 a 3 + (48A ( ,+48A 5 + 3 ^2 3 ) X s +(24 A 0 +64 A„+^A°) X. - 160 A 2 P 2 +90 Aq P (i +60 A 5 P 5 =0. 200a\ (64 A,,+24 A 5 +^'" J ) X, + (48 A„+48 A s +^L‘) X a +(200A S + 32 A„+72 A,+?^) X,+ (l« A.,+98 A,- 225a 3 E I ) X 4 —150 A 3 P 3 (31) + 60 A n Pq+90 A 5 P 5 =0. (32 A (l +32 A-+ y^y ) X 4 +(24 A 0 +64 A 5 + y^y-) A 2 +(16 A„+ 925a 3 96 Ag+^y 1 ) X 3 + (200 A 4 +8 A 0 +128 A.-,+ ^^y ) X 4 —150 A 4 P 4 + 30 A 0 P 0 +120 A 5 P 5 =0. DEEP WATERWAYS. 186 For contact at sill— . . , , 0 . . IGOa 3 , . /nfl * , 225a\ v . . . 200a 3 \ (200 Aj + 128 A 0 +-gj) X, + (96 A 0 +-^-j-) X 2 +(64 A 0 +^-^) X 3 +(32 A 0 + 1 ^) X 4 —150 A, Pj + 120 A 0 P 0 +30 K=0. its 1 (96A„+^£) X I + (200A.,+ 72A„+^ 3 ) X 2 +(4S A„+^ 8 ) X, (24 A„ + 7^_) X.-150 A., P.,+!)0 A„ P„+«0 K=0. (64A,+-'^) X, + (48 A 0 +'^- 3 ) X.,+ (200 A : ,+32 A„+^) O O o/y 3 X 8 +(16 A 0 + jjr j -) X 4 —150 A3 P3+6O A 0 P 0 + 00 K=0. hi l (32A I1 + I ^" S ) X, + (24 A„+^| S ) X ! +(l(iA„+^2?)X,+ (200 A 4 +8 A.+ fa 3 ) X,-150 A, P,+ 30 A„ P„+120 K=0. (32) The only unknown quantities in these equations are X 4 X., X 3 X 4 . The numerical values of A 0 A t A 2 . A- may be substituted and the simultaneous equations solved for X, X 2 , etc. These equations apply to a gate with six horizontal frames. A set of equations may be derived in a similar manner for a gate with any number of horizontal frames. There will always be two fewer equations than there are horizontal frames. In some of the higher gates there will be as many as 30 horizontal frames. It would be impracticable to attempt the solution of 28 simul¬ taneous equations, and, in fact, it is not necessary to do so. The horizontal frames may be grouped into a few groups, each of which may be then considered a horizontal frame. It is evident that the error in so doing is due to considering the center line of the vertical girder to be straight between the half panel points. As the curve of the vertical girder is extremely flat, the error in such an assumption must be very slight. We may then group the horizontal frames into six parts equally spaced, and determine the constants, which may then be substituted into equations (31) or (32), as the case may bp, and from them the desired loading found. To make this clearer, let us make the application to one of the gates designed. Fig. 17 represents the section of the lower gate for the 80-foot lock No. 7 of the Tonawanda-Olcott route. The first diagram shows the arrangement and sectional area of the horizontal frames; tin* second shows the arrangement and sectional area of the frames after they are grouped. 092 = 02 - ----x...,, S>Ot? =, + 2 ---- £ 03 - ^ £\ £9 -- SQ. //v. ///. 2 SQ. /N. A SI/ s/ /7S.O 2 / 3.5 2653 235.3 / 5 < 3.5 Fig. 17. JULIUS BIEN & CO PHOTO UTH H Doc 149 56 2 ' DEEP WATERWAYS. 187 L=552 see plate 73. 17=528 inches. r=200 inches. f=22 inches. E' = 29000000. E'=20000. A„= A, 552 3 7 528 — 4x200 2 V A' 29000000 .00003467 , .00209484 - A' + A" .00003467 .00209484 l 111.25 + 1722.6 ~ .00003467 .00209484 178 3445.2 _ .00003467 . .00209484 ~ 213.5 3445.2 ~ .00003467 . .00209484 ~ 285.3 3445.2 ~ .00003467 .00209484 _ 285.3 3445.2 ~ .00003467 .00209484 158.5 1722.6 oo u u A” 22,000 .000,001,525 .000,000,802 .000,000,770 .000,000,729 .000,000,729 .000,001,433 P 0 = 47,000 pounds. 1*!= 452,000 pounds. P 2 = 926,000 pounds. P 3 =1,100,000 pounds. P 4 = 1,100,000 pounds. P.- = 550,000 pounds. If the gate had no vertical stiffness then X 0 X t , etc., would be zero and the deflection of the miter post would, from equation (24), be 2 A 0 P 0 , 2 A, P„ etc. 2 A 0 P 0 = .1432 inches. 2 Aj Pj= .726 inches. 2 A., P 2 =1.426 inches. 2 Ag A 3 = 1.604 inches. 2 A 4 P 4 =1.604 inches. 2 A 5 P 5 = 1.590 inches. These values are plotted in the line a b c , fig. 1, plate 80. 1 = 190000. a = 156.6 inches. Substituting the known quantities in equations (31) we have: .4784 Xj + ,3272 X 2 +.2714 X 3 +.1743 X 4 -22070=0 .3272 X, + .5803 X,+ .3790 X 3 +.2676 X 4 -52870=0 .2714 Xj + ,3790 X,+ .5484 X 3 +.3198 X 4 -44110=0 .1748 Xj-f.2676 X 2 +.3198 X 3 +.4531 X 4 -22450=0 188 u KKP WATERWAYS. From which, X 4 = — 34,200 pounds. X 2 = -(-80,100 pounds. X., = + 50,750 pounds. X 4 =—25,090 pounds. From equations (26) and (27) X 0 =—41270 pounds. X 5 =—38730 pounds. Substituting now in equations (25): A 0 =.156 inches. A , = .399 inches. A a =.022 inches. A 3 =.754 inches. a 4 = .826 inches. A s =.862 inches. The deflection of the mitre post will be: 2 A 0 = .312 inches. 2A,= .798 inches. 2 A 2 =1.244 inches. 2a 3 =1.508 inches. 2 A 4 = 1.052 inches. 2 A 5 =1.724 inches. These deflections are plotted in fig. 1, plate 80, and the curve of the miter post gh is drawn. Substituting the known quantities in equation (32) and making K=0, we have: .4070 Xj + ,3043 X 2 +.2370 X 3 +.1290- 45920=0 .3043 X, + .5145 X a +.3102 X 3 + .1700—100570=0 .2370 Xj-f .3102 X,+ .4454 X 3 +. 1824-115710=0 .1290 X 4 + .1760 X 2 +.1824 X 3 +.2690-117850=0 Solving, Xj = —112,700 pounds. X 2 = 50,700 pounds. X 3 = 134,700 pounds. X 4 = 303,000 pounds. And from equations (20) and (27) X 0 = — 70,300 pounds. X 5 =—371,400 pounds. Substituting these quantities in equation (25) and doubling the value of A we have as the deflection at the miter. 2A 0 = .429 inches. 2A j= .904 inches. 2 A 2 =1.312 inches. 2 A 3 =1.340 inches. 2A 4 = .898 inches. 2 A 5 =2 K=0. DEEP WATERWAYS. 189 These values are plotted in fig. 1, plate 80, and the curve of the miter post f o drawn. We are now able to determine the loading on any horizontal frame. Let the loading on a frame after deflection be equivalent to a cer¬ tain water pressure P'. Then P' =P (33) In which P„ is the water pressure upon the horizontal frame, and X„ is the reaction of the horizontal frame upon the vertical girder. The deflection of the miter post at any horizontal frame may be found from the curves fig. 1, plate 80. By substituting this deflection and the proper constants in equa¬ tion (23) the value of X„ may be found. This substituted in equa¬ tion (33) gives P the load upon the frame. Using this new loading, the stresses in the frame may be computed as in the beginning. If, as usually will be the case, the stress of the horizontal frame rather than their loading is desired, they may be arrived at by a much shorter process. The line a b c, tig. 1, plate 80, represents the deflection which the miter post would have if there were no vertical stiffness and each horizontal frame withstood its own water pressure. All the frames except the six upper ones were designed to have a maximum stress of + 10,000 pounds per square inch under the action of water pressure alone; and the six upper frames are all made like the seventh. Then the abscissas of the broken line e d b c represent the deflec¬ tions which the miter post would have if all the frames were loaded in such a way that the maximum stress would be 10,000 pounds per square inch. It will be seen that the curves of theoretical deflections pass beyond the line e d b c in two places. When there is perfect contact at the sill the horizontal frame at a point 46 feet above the sill may be stressed to more than 10,000 pounds per square inch, and likewise with frames near the bottom when there is no contact at the sill. In neither case is the deflection more than 8 per cent greater than that corresponding to a maximum stress of 10,000 pounds per square inch. Considering stress proportional to deflection, the maximum stress in these two cases will be only 10,800 pounds per square inch, which is certainly not serious. Knowing the loading on the vertical system, the maximum bending 190 DEEP WATERWAYS. moment may be readily found. The maximum stress/of the vertical beam is found by the formula: _ 2 y x M 7i - i in which y x and y 2 are the distances from the neutral axis to the extreme edges of the vertical beam. M is the maximum bending moment which evidently occurs when there is contact at the sill. The stress is double that given by the usual formula, since the deflection at the miter post is twice the mean deflection. Determining M, y x , and y. 2 and substituting, we have: f x — —17,700 pounds per square inch. f 2 = +16,000 pounds per square inch. This may seem rather high, but there is no danger of rupture, for, unlike the ordinary beam, if the deflection increases beyond that which causes this stress the beam is immediately relieved of its loads. Fig. 2, plate 80, illustrates the case of the upper gate of an 80-foot lock for 21-foot, channel, which is the lowest gate that has been designed, a b represents the deflected position of the miter post if the gate has no vertical stiffness. As it is a straight line, it is also the position of the miter post if there is no contact at the sill, no matter what the vertical system may be. o c represents the position of the miter post if the gate has vertical stiffness and there is contact and no deflection at sill. b d represents the position of miter post when all horizontals have a maximum stress of 10,000 pounds per square inch. As this line falls outside of all others, we may be reasonably certain that in no horizontal is the stress greater than 10,000 pounds per square inch. Fig. 6, plate 80, shows the behavior of an upper gate of 80-foot, lock for 60-foot channel, with 10-foot flood water. This gate receives the highest pressures of any for which timber bearings in the quoin and miter posts are proposed. The line a b represents the deflected position of the miter post if there were no vertical stiffness. e d represents the deflected position of the gate with vertical stiff¬ ness and no contact at the sill. o e represents the deflected position of the miter post with vertical stiffness and contact at the sill. bfd represents the deflected position of the miter post when the maximum stress is 10,000 pounds per square inch. This line lies well outside all others, as in the previous case. DEEP WATERWAYS. 191 At first thought it might be expected that the line bf d would be vertical for the whole height of the gate, inasmuch as it represents a condition of uniform stress in the steel. The section of the timber in the posts, however, is not, as in the case of the steel, proportional to the pressure upon it; hence the inclina¬ tion of the line bf Fig. 4, plate SO, is fora gate between the 80-foot locks of the Lewiston flight. These locks have 31 feet deptli of water on sill and a lift of 40 feet, with a flood at times of 6 feet. The gates may therefore be required to withstand a head of 77 feet of water. a b c is the line of deflection of miter post if there is no vertical stiffness. d c is the line of deflection of miter post under a maximum stress of 10,000 pounds per square inch. o e is the line of the deflected miter post with contact at sill. This line passes outside of d c, indicating that the maximum stress in the horizontals at that point is more than 10,000 pounds per square inch. The maximum excess is only 8 per cent, making the maximum stress only 10,800 pounds per square inch. The comparison of figs. 1 and 4 shows the influence of the wood upon the deflection, as both are drawn to the same scale, and h c fig. 1 and b c fig. 4 would have the same abscissas if it were not for the presence of the wood. These four cases represent the extremes of all the gates designed. The above method is purely theoretical, and as all theoretical an¬ alyses are liable to be in error, it was thought very desirable to obtain a check upon the results by actual measurement of the behavior of large steel gates. With this end in view, a series of measurements were made upon the gates of the Poe lock at Sault Ste. Marie, Mich. These measurements were made by Mr. Henry Goldmark, assisted by Messrs. Joseph Ripley and B. Rohnert, United States assistant engineers, and Messrs. C. M. Ayres, and II. C. MacNaughton, in June, 1808, and April, 1899. The following, quoted from Mr. Goldmark’s report upon the meas¬ urements, described the gates and method of measurements: The lock has a clear width of 100 feet, with a depth of water on sill of 21 feet. The total height of the lower gates is 44 feet, and the average lift of lock is 19 feet. The mitering angle is 21° (rise of about one fifth). The leaves are of soft steel and curved in plan, forming a continu us arch when closed, and are sheathed on both sides. The distance between sheathing plates is 80 inches at the quoin and miter posts and 36 inches at the middle of the leaf. Their thickness varies from three-eighths inch to one-half inch. The frame consists of six horizontal arches spaced uniformly 30 inches between centers and of seven vertical frames, besides the quom and miter posts Four of the vertical frames extend only half¬ way up from the bottom ot the gate. DEEP WATERWAYS. 192 Nom 1 of the vertical frames are cent nuous, except through the riveted con¬ nections. The pressure at the quoin is transmitted through a continuous fiat steel plate 10 inches wide, hearing directly against the cut stone. At the miter posts the contact is through oak timber with 12-inch faces. To determine the deformation of these gates, the fol owing quantities were measured: (a) The change in the versed sines of the two gates considered as a continuous arch between the hollow quoins. This is equivalent to the movement of the miter posts parallel to the axis of the lock. (b) The corresponding movements of points at the middle of each leaf. (c) The change in the versed sine of each leaf. (d) The change in the length of the chord connecting the quo n and miter post. Ail measurements were made for each horizontal frame down to the lower level. The changes in the chord and versed sines of the separate leaves are small and difficult of determination, hence the measurements (a) and ( b) give more satis¬ factory results than those marked (c) and ( d ). The method used in determining the movement of the miter posts and the points at the middles of the leaves may be readily understood by reference to fig. 18, which shows ground plans of the gates and lock walls, in which A and B are points in the gates whose movements are to be measured. They were marked by pr ck punches. I,. I„, I 3 , I 4 , are points on the lock walls at which transits were set up. I M, I N, are base lines parallel to the axis of the lock on the top of the lock walls. Mj, M:, are points at each base line in which it is intersected by a vertical plane passing through the center of the transit and the point A. N,, N„, are the corresponding in ersections for points B. The successive steps in the measurements were the following: The gates were first carefully closed so as to get a symmetrical mitering, and the locks filled to the level of the upper pool. The transits were then set up at I,, T, etc., the points of intersection M. M,, determined (both by direct foresight and after double reversing the instrument), and the lengths I M measured with steel ta es laid flat on the lock walls. These measurements were taken consecutively for all points A on the different, horizontal frames. The water in the lock was then al.owed to run out until it stood about 1 foot higher than in the lower pool. A further lowering it was found was liable to cause the gates to open. The new positions of a 1 points A were then determined by finding the points of intersection M with the same base line and measuring the new distance I M, etc. The comple e series of measurements was, of course, made in every case with¬ out opening the gates or changing the miter, so that the defections obtained are bebeved to be true elastic deflections due to stress in the materbil due to water pressure acting against the gates. Two transits were u-ed. < ne at each wall, merely in order to save time, as the work of each instrument was entirely independent. The measurement for the points marked B were made in exactly the same way as explained for the points A. The actual horizontal movements of A and B, parallel tothesxis of the lock, were obtained by simple proportion from the geometric figure ana the measured distance I M and I N. The above describes the measurements made in June, 1898. At tliis time observations on the gate could be made only above the lower pool. ST.MA RYS FALLS C ANAL. Poe Lo cK — L owe r G ale. General Plan showing Method of measuring Deflections. H Doc 149 56 2 JULIUS BIEN & CO PHOTO. UTH ' - ' DEEP WATERWAYS. 193 In the measurements made in April, 1899, the lower guard gate was closed and the space between it and the lower lock gate was pumped out. The lock was then filled to various stages, and at each stage of water a complete set of observations was made, using the same methods as described above. The measurements made at this time included those designated above as (ft), and a measurement of the compression of the wood, which was accomplished by lowering a man from the top of the gate in such a way that he was able to measure the distance between the prick-puncli marks AA at each horizontal frame for each stage of water. In both 1898 and 1899 very satisfactory results were obtained, except those designated above as (c) and (d). The changes to be measured in these cases were so slight that they could not be accurately deter¬ mined by the means used. The results of the measurements are very uniform. Two cases which are typical of them all will be given. Fifteen and one-half feet depth of water on the upper side of the gate and no water on the lower is called water stage A; 87 feet depth of water on tlie upper side of the gate and no water on the lower is called water stage B; 8 feet depth on the upper side of the gate and no water on the lower is called water stage C. The abscissm of the broken line in fig. 5, plate 80, represent the measured change of deflection in passing from stage A to stage B. The abscissa of the broken line in fig. 6, plate 80, represent the measured change of deflection in passing from stage 0 to stage B. The theoretical deflections were computed in the same manner as in the example worked out on tin* previous pages by substituting the proper values in equations (32). In this case, however, there is a de¬ flection at the sill, as is evident from tigs. 5 and 6. Unfortunately, the lowest arch observed was the one next to the bottom. Extending the probable curve of deflection down to the bottom, we get for the deflec¬ tion at the sill 0.24 inch. Solving the equations and substituting in equations (25), the theoret¬ ical deflections are found which plot into the curve shown in fig. 5, plate 80. The curve shown in fig. 6 is obtained in the same way. The modulus of elasticity of the wood was determined from the computed intensities of pressure at the miter and the measured com¬ pression. It was found that the timber below the level of the lower pool, which was immersed at all times, had a very uniform value of 20,000 as a modulus of elasticity, while that above the lower pool had a modulus of about 30,000. The agreement of the theoretical deflections with those actually measured is as close as can be expected and is very strong evidence that the theoretical results deduced for the gates designed are fair approximations to the truth. H. Doc. 149-13 194 DEEP WATERWAYS. There are three factors in the computations that are necessarily somewhat indefinite. They are the moment of inertia of the vertical girder, the deflection at the sill, and, if timber bearing blocks are used, the modulus of elasticity of this timber. To be on the safe side, the first should be taken large and the sec¬ ond and third small. It is impossible to say how much the effective moment of inertia will be reduced by buckling of the skin plates and defect of the riv¬ eted joints. If it be taken at the value it would have if all verticals of the gate formed one solid piece, then the results will be on the safe side, as the greater the vertical stiffness the greater will be its effect. The value of K, or the deflection at the sill, depends upon the man¬ ner of placing the sill timbers. It is probably on the safe side to take K=0, as the sill timber will compress considerably under the pressure which it will be required to withstand. If timber is used, it will be well saturated at all times, and its modulus of elasticity or compression will be small. In the computa¬ tion, the lowest value obtained from the measurements on the locks at Sault Ste. Marie was used. The simultaneous equations are easily solved by the aid of the slide rule. Mr. Henry Goldmark has worked out this problem, using the method of “least work.” In his analysis he lias made the same gen¬ eral assumptions as have been made in the foregoing analysis, and, as should be expected, his method gives the same results. SHEATHING. The resistance of plates, supported at their edges, to forces normal to them is not well understood. No satisfactory theoretical analysis of their stresses has been made. The designer must rely upon empirical formulae, derived from the results of experiment, to guide him in the proportioning of sheathing plates. The most recent well-conducted series of experiments upon large test plates are those of Prof. C. Bach, of Stuttgart. From the results of these tests formulae were derived which are probably the most reliable of anj T in use. For flat rectangular steel plates, acted upon by fluid pressure, Bach’s formula is: i = cb (34) in which t = tlie thickness of the plate. c = an empirical factor, depending upon the manner in which the edges of the plate are held. DEEP WATERWAYS. 195 For a plate fixed at the edges c = 0.61, a = length of plate between supports. 1) — breadth of plate between supports. p = intensity of fluid pressure. f = maximum allowable tensile stress in the metal. This formula was used to determine the thickness of the sheathing plates for all gates. The results were checked by various other for¬ mulae in general use. No plates less than three-eighths inch thick were used and none more than one-half inch thick were required. The sheathing plates are proportioned by the flat plate formula, although they are in reality curved and thereby derive considerable additional strength from arch section. PIVOT. Details of the pivots which are proposed for use are shown in figs. 15, 16, 17, 18, and 19, plate 70; figs. 10, 11, 12, and 13, plate 71, and figs. 9 and 10, plate 74. The pivots are all of the same general design, although a difference is made in the shape of the castings in upper and lower gates on account of the difference in the sill contact. The bearing is bronze upon polished steel. This combination has a low coefficient of friction, especially when working in water. The bearing parts move very slowly one upon another, tin 1 maximum rate of speed being only 0.02 foot per second. A cast-steel base embedded in the concrete floor of the lock holds the pivot proper, which is a steel forging hemispherical on top. Bolted to the gate is another casting which holds a bronze bushing or hollow hemispherical cup which fits the pivot. The pivot is propor¬ tioned to take both the weight of the gate and the horizontal thrust. UPPER HINGE. The details of f he upper hinge are shown in figs. 11 and 12, plate 70. The force acting at the upper hinge is: h=4w h In which, W=weight of one leaf of gate. «=distance from center of gravity of gate to vertical through pivot center. h —distance between upper hinge and pivot. In this formula, as the height of the gate increases the weight increases in nearly the same ratio. The pull upon the upper hinge does not differ greatly between the lightest and heaviest gates designed. As an example, II for the lower gate for a lock of 40-foot lift is only 15 per cent more than for the lower gate for a lock of 5-foot lift . For DEEP WATERWAYS. 196 this reason little change is made in the anchor bars, and none at all in the castings for the different gates. The anchorage consists of eyebars extending back in the masonry to beams embedded in the concrete. Sufficient masonry is embraced to preclude any danger of movement. The angle between the anchor bars is slightly greater than that through which the gate swings, and they are so placed that both are always in tension, thus avoiding reversal of stress and the consequent danger of loosening and play of the parts. Provision is made for adjustment by means of wedge-shaped keys. The castings are of steel and proportioned for a very low unit stress. The anchor bars are also proportioned very liberally, to provide for any reduction of section which is likely to occur from rusting. SILL CONTACT. For the lower and intermediate gates the sill contact is made, as shown on plate 71, by bolting to the flange of the lower horizontal frame a straight timber bearing piece which closes against a straight timber sill. When the gates are closed a head of water equal to the difference of level of the two sides of the gate acts upon the bottom of the gate, tending to lift it. This lifting effort upon the lower gates is but slightly greater than the weight of the gates themselves, so that the friction of the quoin posts in the hollow quoins will overcome all tendency to upward movement. The coefficient of friction required to do this is only about 2 per cent. On account of this upward pressure upon the bottom of the gate, numerous web stiffeners are placed upon the bottom frame, and its lower flange is strengthened by a plate extending to the flange of the frame above. If the upper and guard gates and gates between locks of a flight should expose so great a bottom area as this to maximum water pres¬ sure, the upward lift would be far in excess of the weight of the gate. To meet this difficulty the bottom frame of these gates is made very much narrower than the others and no timber bearing piece is used. Instead, the heavy steel flange angle is finished and bears against the sill. This arrangement is shown on plate 70. The lower frame is attached to the one above at intervals by cast- steel brackets. A curved sill is required for these gates. FOOTBRIDGE. For the convenience of the workmen and others, a footbridge is provided upon the top of the gate. It has a railing which is remov¬ able and built in short sections, so that one man can handle a section. DEEP WATERWAYS. 197 ESTIMATES. In connection with the estimates, it is appropriate to review briefly the conditions which prevail in tlie structures. The gates are of steel, with horizontal framing. The spaces between the frames are variable, being 3 feet 3 inches near the top and less than 2 feet near the bottom of the highest gates. The frames are straight on the downstream side and curved on the other; their breadth is 4 feet in gates for 60-foot, locks and 4^ feet in gates for 80-foot locks. They are proportioned under the assumption that the center of pressure at the miter may have a range of position 8 inches each side of the center of the bearing face. The maxi¬ mum stress in the frames is 10,000 pounds per square inch. The maximum stress in the sheathing is 15,000 pounds per square inch. The minimum thickness of metal is three-eighths inch. The minimum angle used is 34x34x§ inches. Diameter of rivets f and f inch. Detailed designs and careful estimates of the weight of gates for about one hundred and twenty-five different cases were made. These comprised single and double skin, upper and lower gates for 60, 65, and 80-foot locks for a 21-foot and 30-foot channel, and lifts varying from 0 to 50 feet. The estimated weights of the rolled steel in the lower gates are plot¬ ted on plate 79. Abscissas are the lifts of the locks and ordinates are the weights of structural steel in one leaf of the gates. To each diagram shown, a parabolic curve is fitted as shown on plate 79. The ordinates of the computed points, with two exceptions, do not differ more than 2 tier cent, and in these cases not more than 34 per cent, from the ordinates to the parabolic curves; therefore, if the weights of the gates for a lock of any lift between 0 and 50 feet be read from the curve, they will be within the practical limits of accuracy for structural steel work. The general equation of these parabolic curves is: W=AH 2 +BH+C . . . (35) in which W is the net weight in pounds of the structural steel in one leaf of the lower gate of a lock, the lift in feet of which is II. A, B, and C are constants dependent upon the width of the lock and depth of water on sill. The equations of weight of lower gates for 80 and 60 foot locks are— For 80-foot lock, 30 feet depth of water on sill, 'W=S5II 2 -f-3000H+ 130000. For 80-foot lock, 21 feet depth of water on sill, W = 83IF + 301011 + 103000. For 60-foot lock, 30 feet depth of water on sill, W=38.85ir~-f-3009II+ 106500.- For 60-foot lock, 21 feet depth of water on sill, W=38.76H 2 -|-27131I+ 80500. 198 DEEP WATERWAYS. The weight of a lower gate for any particular width and lift of lock, but of varying “depth on sill,” may be written: W=Kfl-CD, in which K represents the weight of that part of the gate above the surface of the lower pool plus that part below the top of the sill, and C is the weight per vertical foot of that part of the gate between the surface of the lower pool and the sill, and 1) is the depth in feet of the water on sill. It is evident from the nature of the design that this form of equation is correct, because the water pressure per vertical foot upon the gate below the lower pool is constant, and therefore the horizontal frames, their spacing, the quoin and miter posts, the verticals and sheathing are all uniform from the surface of the lower pool to the sill. Making A, B, and C linear functions of I) and determining con¬ stants, we have— For 80-foot lock W = (78.33+.222D)H 2 + (1033.5+65.55D)H+(40000+ 3000D) . . . (36) For 60-foot lock W = (85.55 + .01D)H 2 +(2022.6+32.88D)H + (19860+ 2888D) . . . (37) These are the equations of the two series of curves shown on plate 79. These formulae and curves apply to upper and guard gates as well as to lower gates. II is to be taken as the maximum head of water acting upon the gate. For lower gates this is the maximum lift of the lock. In the designs all upper gates were proportioned the same as guard gates. By the use of either the equations or the curves the weight of struc¬ tural steel in one leaf of a gate within limits previously mentioned may be found. To this must be added the material in pivot and upper hinge details, given on plate 79. These are the equations which are especially intended for use in making the estimates of the gates for the locks of the various deep waterway routes investigated by the Board. Twenty-one and 30-foot channels only are proposed for the deep waterway, yet the depth of water on sill varies widely for different cases on account of varying flood water and other local conditions; otherwise the diagrams, fig. I on plate 79,-might be used to get the weight of the gates. The curves on plate 79 were used to make the estimates for all the gates for the proposed deep waterway, except those between locks of a flight. These estimates are included in the general estimates, which may be found in the reports of the various assistant engineers. The gates between the locks of a flight are made the subject of spe¬ cial detailed estimate, since they do not come within the limits of II =d and II = 50. In the flight of 40-foot lift locks H is as great as 77 feet. DEEP WATERWAYS. 199 The principal object of all the work clone upon the subject of lock gates was attained when equations (30) and (37) were developed and estimates for the proposed work had been obtained from them. As a matter of general interest, however, the work was carried somewhat further. Numerous diagrams were plotted, weights of gates being ordinates and widths of locks being abscissas. The lift of lock and depth of water on sill remained the same for each diagram. It was found that the points plotted in these diagrams fitted parabolic curves, or, in other words, for any given lift of lock and depth of water on sill: W=E6 2 +F6+G .... (3S) in which 6 is the width of the lock and E, F, and G are certain con¬ stants depending upon the lift of lock and depth of water on sill. This might have been expected, since certain parts of the gate, such as the flanges of the horizontal frames, increase as the square of the span and certain other parts of the gate, such as the sheathing, increase directly as the width, and certain other parts, such as the posts, do not vary at all. The three equations representing the weight of single skin gates for GO, G5, and 80 foot locks, with 30-foot depth of water on sill, are: 80-foot lock, 30-foot depth on sill W=85 L 2 +3600 L + 130000 (55-foot lock, 30-foot depth on sill W=47.12 L 2 +3250 L+109700 GO-foot lock, 30-foot depth on sill W = 38.85 L 2 -|-3009 L-j-106500 The only form of equation that will satisfy the conditions of both equations (35) and (38) is: W=(d6 2 4-e 6+/) JV+{gb 2 +h b + k) 11 +(j 6 2 + m b+n) . . (39) in which any constant may vanish. We now have: G400 d+80 e+f=85 4225 (/+G5 e+/=47.12 3600 (/+G0 e+f— 38.85 from which we have d=. 043G e= — 3.796/= 109.65. In like manner, g, It, 1c, j, m, and u may be found and substituted in the general equa¬ tion (26). We then have the general equation for the weight of one leaf of a single skin gate for varying widths and lifts, but for a con¬ stant. depth of water on sill of 30 feet: W= (.04366 2 — 3.7965 + 109.65) H 2 + (— 1.2436 2 + 203.56 - 4725) II + 35.6G66 2 — 38186 + 207200. Proceeding in the same way with the gates for 21 feet depth on sill, we have for the weight of one leaf: W= (.03886 2 - 3.226 + 92.28) H 2 + ( — .7036 2 + 113.36 — 9756 + 85000. 1553) II + 156 2 200 DEEP WATERWAYS. .Vs before, we know from t lie nature of the design that the weight of a gate is a rectilinear function of I), the depth of water on sill. We may, then, combine the above equations into a general equation giving the weight of a single skin gate for any depth of water on sill, width and lift of lock within limits. We then have: W = [(.0005331) + .0270)0* + (— .0641) — 1.876)0 + 1.931) 4- 51.75] H* + [(— .061) + .557)0* 4- (10.0221) — 97.16)6 — 352.441) + 5848.3] II + [(2.2961)— 33.215)0* 4* ( — 315.891)+ 5658.6)0 +135771)-200100]. In which W = the weight in pounds of the structural steel in one leaf of a single skin mitering gate with horizontal framing, straight on the downstream side. 0 = the breadth of the lock in feet. II = the maximum head in feet of water acting upon the gate. I) = the depth on sill, or depth in feet above the top of the sill of the water on the downstream side of the gate when head II acts. In the same manner the equation giving the weight of double skin gates was developed: W = [(.000531) + .0276)6* + (- .0641) - 1.876)6 + (1.931) + 51.75)] H 2 + [(— .061) + .3833)6* + (10.033D — 68.71)6 + ( — 353D + 5107)] II + [(2.0441) - 29.6)6*+( — 274.671) + 5121.47)6 +(123331) —179600)]. The caution may be again repeated that these equations are empir¬ ical and should not be used outside of the limits of the data from which they were derived; that is, they should not be used for values of L greater than 50 feet, nor for 6 greater than say 85 feet, nor less than say 55 feet. From the nature of the design, I) may be almost any¬ thing. BIBLIOGRAPHY OF ARTICLES ON MITERING LOCK GATES. (A) General Hydraulic Text-Books and other Works Containing Chapters on Lock Gates. M. Becker. I)er Wasserbau. Stuttgart, 1861. An excellent text-book with a good general discussion on lock gates. Plans are given of the timber gates for the 100-foot lock in the Canada docks, Liverpool. Built 1853-1858. G. Hagen. Handbuch der Wasserbau-Kunst (2terTheil, 3terBand). Sehitfahrts canaele, Berlin, 1874. A standard German work on hydraulic engineering. Contains an historical account of the development of locks and a description of the details of small gates as built in different countries. DEEP WATERWAYS. 201 Of larger gates, the following are described and in part illustrated by plans, viz: Cherbourg Harbor. Width of lock equals 55 feet. Timber gates built near the beginning of this century. Coburg dock, Liverpool. Width of lock equals 68 feet. Timber gates. Canal lock, near Mulhouse (Alsace), 17 feet wide. These were the first wrouglit- iron gates built on the Continent (1845). Bremerhafen docks, 1848. Width of lock equals 76 feet. The first large wrought-iron gates built in Germany. Geestemunde docks. Width of lock, 74.68 feet. Wrought-iron gates, built in 1861. Willem III lock (Holland). Wrought-iron gates, built in 1861-1865. Width of lock equals 58 feet. Charenton lock, France. Small wrought-iron gates. Built in 1865. Width of lock equals 25.5 feet. L. Franzius. Der Wasserbau. Berlin, 1890. Contains a short but excellent chapter on lock gates. L. Brennecke (chapter on locks) in Der Wasserbau. (Ill Band d. Handbucli der Ingenieur wissenschaften) 2te Abtheilung 2te Halfte. Schleusen. Schiffahrts Ivanaele. Leipzig, W. Engelmann, 1895. Latest German treatise on hydraulic engineering. The chapter on locks is written by the designer of the lock gates on the North Sea—Baltic Canal. It contains a very full discussion of the entire subject, including the theory of stresses and many detailed plans. The following large gates are described and illustrated: Canada docks (Liverpool): Timber gates, built 1853-1858. Width of lock equals 100 feet. Transatlantique docks (Havre): Timber gates, built in 1862. Width of lock equals 100 feet. Antwerp docks: Timber gates. Width of lock equals 80 feet. Willemsvord docks (Holland): Wrought-iron gates. Amsterdam Ship Canal: Wrought-iron gates. Width of lock equals 59 feet. Geestemunde docks: Wrought-iron gates. Width of lock equals 74.68 feet. North Sea, Baltic Canal: Steel gates. Width of lock equals 82 feet. A. Debauve. Navigation Fluviale et Maritime. (Manuel de l’lngen- ieur des Pouts et Chaussees 19 eme Fascicule). Paris (Dunod), 1878, 795 pages and atlas. Standard French treatise. Has a detailed discussion on lock gates, including the theory of stresses as based on Chevallier's experiments and Lavoinnes’s mathe¬ matical investigations. Description and plans of the following large gates: Havre (Transatlantique docks): Timber gates. Width of lock equals 100 feet. Dunkirk dock: Timber gates. Width of lock equals 68.9 feet. Dieppe Duquesne dock: Timber gates. Built in 1869. Fecamp docks (compound gates of timber and iron): Width of lock equals 54.1 feet. Boulogne docks (iron gates): Built in 1866. Width of lock equals 68.9 feet. Amsterdam Ship Canal (iron gates): Built in 1874. Width of lock equals 59 feet. P. Guillemain. Navigation interieure. Rivieres et Canaux. (En¬ cyclopedia des Travaux Publics) Paris. (Baudry et Cie) 1885. Tome II, p. 32 to 97. Chapter on lock gates. 202 DEEP WATERWAYS. The general subject is discussed and many small wooden gates are illustrated and described. The most valuable feature of this book is the account given of the author's experiments with models, showing the effect on the distribution of pressures of spacing the horizontal girders in a gate in various ways. A mathematical theory is also developed by which the loading on each horizon¬ tal can be approximately obtained. F. Laroche. (Ports Maritimes.) (Encyclopedic des Travaux Pub¬ lics.) Paris. (Baildry et Cie) 1893. Tome I, p. 185. Chapter on lock gates. This treatise gives the general theory rather than special designs. In the appendixes, detailed calculations of the stresses in two gates are given, one of them based directly on Chevallier's experiments with models and the other on Lavoinne's formulae for the distribution of stresses. C. Colson. Notes on docks and dock construction. London (Long¬ mans, Green & Co.), 1894. This book gives the views of an experienced engineer on various topics rather than a systematic exposition. In the chapter on lock gates, various mooted ques¬ tions in gate design are taken up from a practical standpoint. Plans of the following gates are given: Dunkirk docks (wrought iron): Width of lock equals 69 feet, Amsterdam Ship Canal (wrought iron): Width of lock equals 59 feet. Avomnouth docks: Timber gates. Width of lock equals 70 feet. Whitehaven docks: Timber gates. Width of lock equals 50 feet. Barry docks: Wrought iron. Width of lock equals 80 feet. San Fernando docks. Buenos Ayres: Wrought iron. Width of lock equals 64 feet. West India docks: Wrought iron. Alexandria docks, Hull: Timber gates. In the appendix the stresses are treated graphically. Charles B. Stuart. The Naval dry docks of the United States. New York, 1852. Gives plans and description of 60-foot dock gates in Brooklyn Navy-Yard, prob¬ ably the first large gates built entirely of wrought iron. Chief of Engineers, United States Army. Annual Report for 1895. Contains reports on the steel gates for the 100-foot Poe lock at Sault Ste. Marie, Mich., viz: Page 3024. Report on inspection by David Molitor, United States assistant engineer. Page 3028. Descriptive memoir of the lock gates and methods of calculating the stresses, by Capt. H. F. Hodges. Page 3041. Distortion of steel gates, by David Molitor, United States assistant engineer. Chief of Engineers, United States Army. Annual Report for 1897. Page 2975. Report on the machinery, by F. M. Dunlap, United States assistant engineer. J. Fuelscher. Der Ban des Kaiser Wilhelm Kanals. Berlin, 1898. (North Sea-Baltic Canal.) The large lock gates at Holtenau and Brunsbuettel (82 feet width of lock), as well as the smaller structures, are described in detail and illustrated. DEEP WATERWAYS. 203 Encyclopedia Britannica, ninth edition. (Article, harbors.) Drawings of the timber gates for the Great Grimsby dock (width of lock equals TO feet) and of the wrought-iron gates in the Victoria dock, London (width of lock equals 80 feet), are given. (B) Special Monographs on Lock Gates. Sylvain Perisse. et en Angleterre. Etude sur les portes d’ecluse a la mer en France Paris, 1872. pp. 09. A reprint in book form from Memoires des ingenieurs civils, 1872. An excellent theoretical and practical essay on lock gates in general, with detailed description and plans of most of the following gates: Timber gates. —Dunkirk, 1856, width of lock equals 68.88 feet; St. Nazaire, 1856-1859, width of lock equals 81 feet; Havre, 1863, width of lock equals 100 feet; Dieppe, 1871, width of lock equals 54.12 feet; Fecamp, 1865, width of lock equals 54.12 feet; Boulogne, 1866-67, width of lock equals 68.88 feet; Great Grimsby, 1848, width of lock equals 70 feet; Canada docks, Liverpool, 1857, width of lock equals 100 feet. Wrought-iron gates .—Brooklyn Navy-Yard, 1847, width of lock equals 60 feet; Victoria docks, London, 1857. width of lock equals 80 feet; Jarrow docks (Tyne), 1858, width of lock equals 80 feet; Limeliouse docks, width of lock equals 60 feet; Surrey docks, width of lock equals 50 feet; Boulogne docks, 1867, width of lock equals 68.9 feet: Havre docks, 1871, width of lock equals 52.5 feet. This essay covers the ground very well up to 1872. The series of articles on dock gates published in The Engineer in 1873 are practically a translation of this paper. Harry F. Hodges. First lieutenant, Corps of Engineers, United States Army. Notes on Mitering Lock Gates. Washington, Govern¬ ment Printing Office, 1892, 132 pages, 7 plates. (Professional Papers of the Corps of Engineers, United States Army, No. 26.) This is one of the best papers on lock gates. It treats very fully of the stresses in ail the members and develops a practical method for correct designing. Plans of the following gates are given: Boulogne dock (iron gates), 1866. Width of lock equals 68.88 feet. Havre (Transatlantique dock), iron gates, 1866. Width of lock equals 100 feet. Tyne docks (iron gates), 1857. Width of lock equals 80 feet. Barry docks (wrought-iron gates), 1899. Width of lock equals 80 feet. Avonmouth docks (timber gates), about 1877. Width of lock equals 70 feet. St. Marys Falls Canal (timber gates), 1881. Width of lock equals 60 feet. Charenton dock (iron gates), 1865. Width of lock equals 25.5 feet. Theodor Landsberg. Die eisernen Stemnithore dor Sehiffschlensen. Leipzig, W. Engelmann, 1894-. (Fortschritte der Ingenienrwissen- schaften, 2te Gruppe.) This monograph on metallic mitering lock gates is of considerab’e value, but apparently not the work of a practical designer. Many of the calculations given are of little value. Descriptions are given of the gates for the 100-foot lock at Havre and for the second Harbor entrance at Wilhelmshafen. G. Weitzel. Brevet major-general, United States Army. Construe- 204 DEEP WATERWAYS. tion of iron lock gates for the harbors of the Weser Kiver, Germany. Washington, D. C., 1874. Translation of German notes and plans of: Bremerhafen dock gates, built in 1848. Width of lock equals 76 feet. Bremerhafen dock gates, proposed in 1872. Width of lock equals 60 feet. Geestemunde dock gates, built in 1861. Width of lock equals 74.68 feet. Edouard Widmer and Henry Desprez. Port du Havre. Memoire sur les nouvelles portes en tole de l’ecluse des transatlantiques. Paris (Dunod), 1887. Reprinted from Annales des Ponts et Chaussees, 1887. It gives a full account of the design, calculation, tests of material, etc., of the wrought-iron gates for the 100-foot lock at Havre. II. Desprez. Port du Havre. Notice sur le Bassin Bellot. Paris (Dunod), 1889, 97 pages. Reprinted from Annales des Ponts et Chaussees, 1889. Full plans and description of the wrought-iron gates lor a lock 100 feet wide. (C) Periodicals and Transactions of Engineering Societies. Institution of Civil Engineers (London). Transactions (4to). Yol. 1. 1836. Barlow. An early article on lock gates. Of much interest, but containing some mistakes. Minutes of Proceedings (8 vo.). Yol. 6, 1847, page 47. Sebastopol locks. Built by Rennie. Plans and descrip¬ tion of gates for lock 64 feet wide. Cast-iron frame and wrought-iron plating. Yol. 15. 1855. Sunderland docks. Plans, but no description, of gates with cast- iron frame and timber sheathing. Yol. 18, 1858. Kingsbury. Wrought-iron gates for lock 80 feet wide for Vic¬ toria docks, London. Excellent paper. Insists on greater economy of arched gates, reasoning inconclusive. Yol. 31, 1870. Browne. Strength of lock gates. Mathematical discussion of stresses in a wrought-iron horizontal gate girder of the most general form. Very complex, but gives results no better than simple graphical method. Yol. 34, 1871-72. H. Vernon Harcourt. Wrought-iron gates for South dock, West India docks. Width of lock equals 55 feet. Vol. 55. 1878-79, page 54. In discussion on Whitehaven docks. Browne eluci¬ dates his opinions further. Yol. 58, 1878-79. Blandy. Dock gates. Good theoretical analysis (also by graphics) of stresses in wrought-iron and timber gates. Vol. 59. 1879-80. Discussion on Blandy‘s paper. Yol. 62, 1879-80. Hayter expresses his opinion on the value of eccentricity at quoin post. Vol. 70, 1881-82. Bo’ness Harbor. Plan and brief description of timber gates for a lock 50 feet wide. Yol. 92, 1887-88, page 153. Alexandra dock, Hull. Plans and description of timber gates for a lock 85 feet wide. Vol. 97, 1888-89, page 336. W. J. Hall. Steel gates for Limerick floating dock. Width of opening equals 70 feet. Yol. 101, 1889-90, page 139. Barry docks. Brief description—no plans—of wrought-iron gates for lock 80 feet wide. Machinery described in detail. Yol. 107. 1891-92. Moncrieff. Good practical paper on lock gates, by a man who has built many modern gates. DEEP W ATEEWAYS. 205 The Engineer (London). December 27, 1861. The Grangemouth dock, by James Milne. Details of tim¬ ber gates for lock 25 feet wide. January 22, 1869. Avonmouth dock, Bristol. Timber gates for lock 85 feet wide. Large inset sheet of details. September 10, I860. Millwall lock gates. Brief description and general plans of wrought-iron gates. Each leaf is 42 feet long by 81 feet high. February 7. 14, 28, March 21, April 11, May 2, June 6, 20, July 18, October 10, 1878. “On the construction of dock gates.” An excellent series of articles.based largely on M. Perisse’s paper, published by the French Society of Civil Engineers, in 1872. The drawings in Perisse's article are also reproduced. November 21, 1884, and January 30, 1885. C. H. Romanes. “The strains in circular lock gates.” The theory here given is quite inadequate. April 1, 1892. E. Duncan Stoney. “ Distribution of beams in lock gates." A graphic method for proportioning the horizontal girders so that the leading shall be the same on each gilder. (Of no special value.) Civil Engineer and Architects’ Journal. (London.) 1861, page 87. Grangemouth docks. Plans of gate for lock 25 feet wide, same as described in Engineer for December 27, 1861. Engineering News. (New York.) March 28, 1895. Lock at Sault Ste. Marie, Ontario (Canadian lock). Plans of operating machinery moved by electricity. October 21, 1897. Cascade lock (Columbia River). Photograph and descrip¬ tion. No drawings. July 9, 1869. Kings Lynn dock. Plans and description of timber gates for a lock 50 feet wide. February 25, 1870. Keefer. Plans and description of solid timber gates with¬ out posts, used on the Welland and other Canadian canals. October 16,1874. Middlesborough docks. Wrought-iron gates. 24 feet high by 35 feet long. October 8,1875. Alexandra docks, Newport. Plans and description of wrought- iron gates for a lock 65 feet wide. March 21, 1879. Ayr dock. Plans and description of timber gates for a lock 60 feet wide. April 8 and May 6, 1881. Sunderland dock. Plans and description of timber gates for lock 65 feet wide. December 1. 1882. Grangemouth dock. Plans and description of timber gates for a lock 55 feet wide. January 26. 1894. Manchester Ship Canal. Timber gates for locks 30, 45, 50, 65, and 80 feet wide. Descriptions and photographs, but no plans. October 12, 26, November 9. 1894. West India docks. Plans and description of wrought-iron gates for a lock 60 feet wide. June 21, 28, July 5, 12, 1895. North Sea-Baltic Canal. Plans and description of steel gates for the large locks 82 feet wide, and of fan gates for a lock 39.4 feet wide. January 31 and February 14, 1896. Lady Windsor dock, Barry. Plans and description of wrought-iron gates for lock 65 feet wide. September 2,1898. Glasgow graving dock. Plans and description of steel gates for a lock 83 feet wide. Annales des Ponts et Chaussees. 1832. II Semestre, page 261. Acollas. Canal de Berry. (Small cast-iron gates.) 1849. II Semestre, page 177. Early wroughr-iron gates at Mulliouse, Alsace. 1850. I Semestre, pages 309-356. V. Chevallier, Recherches experiinentales DEEP WATERWAYS. 206 des portes d'ecluses. (Classic experiments on models showing relative deflections of horizontal and vertical girders in lock gates.) 1852. I Semestre, page 253. Feburier. (Tests made on the deflections of a timber “horizontal” in the St. Malo lock gate.) 1861. I Semestre. page 113. La Ferine. (On a method of unshipping a large lock gate.) 1865. 1 Semestre, page 139. Malezieux. Description and plans of small wrought-iron gate at Charenton. 1866. I Semestre, page 126. Lermoyoz. Sur le merite comparatif des portes d'ecluses en hois et des portes en metal. (Gives reasons for preferring wooden canal lock gates to metallic ones.) 1867. I Semestre, pages 321-430. Lavoinne. Memoire sur la flexion des entre- toises et du bordage dans les portes d'ecluses. (Classic work, awarded a prize of 300 francs; gives mathematical analysis of strains in horizontal and vertical girders in gates.) 1868. II Semestre, page 339. Cambuzet. Note comparative sur les portes en metal et en bois qui existent aux ecluses du canal du Nivernais. (Describes acci¬ dent to cast-iron gates. Prefers timber.) 1869. II Semestre, pages 81-102. Carlier. Portes d'ecluses du Port de Fecamp. (Combined timber and iron gates for a lock 53 feet wide.) 1881. I Semestre, page 540. Boutan. L'appareil hydraulique des portes d'ecluse a Bordeaux. (Calculations and plans of hydraulic machinery.) 1887. II Semestre, pages 411-463. E. Widmer and H. Desprez, Port du Havre. Nouvelles portes en tole d'ecluse des transatlantiques. (Full plans and descrip¬ tion of the wrought-iron gates for 100-foot lock.) Also published in book form. 1887. II Semestre, pages 704-756. Galliot. Etude sur les portes d’ecluses en tole. (Excellent mathematical investigation of (1) strength of flat iron plates; (2) relative stresses in the different horizontal and vertical girders in a gate. This is same problem as treated by Lavoinne in 1867.) 1888. I Semestre, page 1018. Laroche. Methode elementaire pour calculer la resistance des portes d’ecluse. (Elementary investigation of stresses. Not very valuable.) 1889. I Semestre, pages 5-97. Desprez. Le Bassin Bellot. (Excellent plans and description of iron gates for a lock 98.4 feet wide.) Also published in book form. 1892. I Semestre, pages 633-801. Maurice Widmer. Canal du Havre a Tan- carville. (Excellent plans and calculations for wrought-iron gates for lock 52^ feet wide.) 1892. II Semestre. page 783. A. Fontaine. Les Ecluses a grande chute (5.2 m.j du Canal da Centre. (Interesting small gates with buckled steel covering.) 1893. II Semestre. page 44. Maurice Renard. Nouvelles ecluses du Canal St. Denis. (Single skin gates for lock 27 feet wide, with 32| feet lift, small but interesting.) 1895. I Semestre, pages 459-602. C. de Franchimont. Note sur la construction du 3 erne bassin a flot da Rochefort. (Excellent account of gates and machinery. Reasons for preferring “ horizontal’ - system of girders.) MEMOIRS DE LA SOCIETE DES INGENIEURS CIVILS. 1872. Pages 319-415. Sylvain Perisse. Etude sur les portes d'ecluse a lar Mer en France et en Angleterre. (Excellent essay on dock gates. Reprinted as a book. Paris, 1873.) PORTEFEUILLE ECONOMIQUE DES MACHINES. 1883. July, page 102. Appareils de Manoeuvre des portes et des ventelles des ecluses double du Canal du Nord sur Paris. (Plans and description of operating machinery and valves for small locks,) DEEP WATERWAYS. 207 TRAVAUX DE VACANCE (ECOLE CENTRALE). Port du Havre. Porte d'ecluse gome Bassin. (Two sheets of illustrations: no text.) ZEITSCHRIFT DES HANNOVERSCHEN ARCHITEKTEN UND INGENIEURVEREINS. 1852-53, page 339; 1853-54, page 241. Rulilmann. (Short mathematical study on most economical sill angle. No special value.) 1855, page 475. H. Huebbe. Details of timber gates for Great Grimsby Lock, 70 i'eet wide. 1861, page 93. Plener. Great Western Dock gates at Plymouth (England). Plans and description of wrought-iron gates for lock 80 feet wide. 1865, page 226. Welkner. Wrought-iron gates for Geestemunde Docks. Width of lock equals 76 feet. Valuable experiments on strength of flat plates. 1865, page 491. Interesting historical note on wrought-iron gates. 1866, page 309. Hess. Bemerkungen uber die neuen belgischen und franzo- sisclien Konstruktionen der Kanal Schleusenthore. 1838, page 419. Professor Barkhausen. “Uber einige neuere Englischen See- schleusen.’’ Excellent notes and sketches on modern English locks. 1889, page 743. Van Horn. Plans of gates for Transatlantique Dock. Havre. (Taken from Annales des Ponts et Chaussees, 1887.) 1891, page 349. E. Rechtern and H. Arnold. Der Bau der zweiten Hafenein- fahrt zu Wilhelmshafen. (Good plans and descriptions of wrought-iron gates for lock 80 feet wide. Reason for preferring girder to arch shapes.) DEUTSCHE BAUZEITUNGr. 1891, September 12 and 19. L. Brennecke. Die Entwickelung der Schleusen¬ thore der Neuzeit. (Brief but excellent statement of the author's views on lock gates.) TIJDSCHRIFT KONINKLIJK INSTITUUT VAN INGENIEURS (HOLLAND). 1863-64, page 14. J. Strootman. Over wijde zeesliuzen en sluitdeuren van plaatijzer. (Complete study on calculation and design of iron lock gates.) 1866-67, page 116. Strootman. Docks Willemoord. (Wrought-iron gates about 36 feet square.) 1870-71. page 187. J. F. W. Conrad. (Full description of Willem III lock, includ¬ ing plans of wrought-iron gates for a lock 59 feet wide.) 1885-86, page 429. J. Strootman. Ijzeren deuren voor sluizen op binnenlandsche scheepvaartskanalen. (Good theoretical study of small iron gates.) G-rateful acknowledgment is here made to Mr. Joseph Ripley, United States assistant engineer in charge of St. Marys Falls Canal, for valu¬ able information freely given and assistance rendered in making meas¬ urements of the deflection of the lock gates at Sault Ste. Marie. The work covered by this report was begun in December of 1897 by Mr. Henry Goldmark, assistant engineer, and remained under his charge until June, 1899. After his retirement on account of ill health, the undersigned, who had acted as principal assistant from May, 1898, completed the investigations and prepared the present report. Very respectfully submitted. S. II. Woodard. The Board of Engineers on Deep Waterways. 208 DEEP WATERWAYS. Appendix No. .'3. BREAKWATERS AT CANAL ENTRANCES, OSWEGO AND OLCOTT, N. Y. BREAKWATERS AT CANAL ENTRANCES. The plans of the Board of Engineers on Deep Waterways provide for canals entering Lake Ontario at Oswego and Oleott, N. Y. I have the honor to submit for the consideration of the Board the following projects for the construction of breakwaters at these entrances, with estimates of cost. The breakwaters first considered are adapted to the requirements of a waterway having a depth of 30 feet. OSWEGO. The existing harbor at this locality is situated at the mouth of the Oswego River, near the eastern end of Lake Ontario. The prevailing violent winds are from the west and northwest. The greatest water exposure is to the westward, which is therefore the direction of the heaviest seas. The movement of gravel and other material derived from the attrition of the shores is from west to east. The material of the bottom is sand and gravel underlaid by argillaceous sandstone of the Utica formation. The construction of breakwaters to form an anchorage and cover the entrance to the Oswego River w r as commenced by the Government in 1827, and works of extension or repair have been in progress down to the present day. The existing breakwaters are of the usual crib formation so long employed in the construction of such works for the improvement of the lake harbors. Their locations are shown on plate 18. The proposed canal, as designed by the Board, enters the lake at Sheldons Point, which is a short distance west of the existing break¬ water harbor. The first lock is located on the shore of the lake. The channel connecting it with deep water in the lake will be about 1,700 feet long and 600 feet wide, and is to be bordered by timber cribs for mooring and guiding vessels, as shown on figures 1 to 6, inclusive. PLAN OF THE BREAKWATERS. The breakwaters required at this locality have for their objects to shelter the entrance to the canal from heavy seas, to protect the excavated channel from shoaling from the movement of materials worn from the shores, and to form a harbor of refuge for vessels in time of storms. The heaviest seas come from the westward. The main breakwater is therefore located west of the canal and extends in a northerly direction, following the arc of a circle, with 4,700 feet radius, in order to increase the anchorage area. This direction has been selected so as to be about normal to the direction of the most DEEP WATERWAYS. 209 violent wave mot ion, so as to avoid the formation of an accumulated wave. The long breakwater in the existing harbor has a northeast direction, and makes so small an angle with the direction of greatest wave motion that an accumulated wave is formed at its eastern end during the prevalence of westerly storms, making entrance to the harbor difficult and dangerous. The main breakwater commences at the shore, and therefore serves as a jetty to stop the eastward movement of material which might form shoals in the excavated channel. Since northerly winds sometimes create heavy seas in this locality, a second breakwater has been designed to extend from a point on the face of the old breakwater, in a northeast direction, toward the eastern extremity of the main breakwater. The interval between the outer ends of these breakwaters, which has a width of 1)50 feet, forms the entrance to the harbor. It will be observed that these breakwaters are located, witli refer¬ ence to each other, so as to leave between their ends a clear channel GOO feet wide, measured perpendicular to its axis, which curves with a radius of 5,000 feet from the entrance to a point between the heads of the piers. The tangent to the axis of the channel at the entrance has a direction about north. The shelter could be increased by ex¬ tending the western breakwater farther to the eastward; but this would increase the curvature of the channel at the entrance, and is therefore considered undesirable. The harbor thus formed will have an area of nearly 127 acres, meas¬ uring to the low-water line. The existing harbor, which has an area of about 180 acres, will always be available for the anchorage of ves¬ sels of small draft, and it can be made conveniently accessible from the new harbor by the removal of the breakwater at its western end. Large vessels entering the harbor will, of course, pass directly into the canal. It is believed that the mooring piers and the anchorage area in their vicinity will afford ample accommodation for those descending the canal and detained by storms. The aggregate length of the proposed breakwaters is 0,635 feet. Their locations are shown on figures 1 to (5, inclusive. CHARACTER OF STRUCTURES. Heretofore the piers and breakwaters constructed in the Great Lakes have been formed of pile work or timber cribs filled with stone. When the exposure is great, crib breakwaters have been found preferable to those formed of piles. These structures are, to a certain extent, of a temporary character, and require frequent and troublesome repairs. Below the water level they may be considered permanent, if not displaced by storms, but above that level they require renewal every ten or fifteen years. The cribs require care¬ fully prepared foundations, and when all precautions have been taken H. Doc. 149-14 210 DEEP WATERWAYS. it is not always possible to make them retain their places upon the bottom. Breakwaters thus constructed are, of course, much cheaper than structures formed wholly of concrete or stone masonry, but it is a question worthy of investigation whether they are as advantageous or as economical, considering the cost of maintenance, as random stone breakwaters constructed in accordance with modern methods, like those which have been established at certain localities on the coast of the sea. The random-stone breakwaters adopted for the formation of harbors on our seacoast consist of a substructure formed by depositing stones along the site of the work to about the level of mean low water, and of a superstructure constructed of heavy stones laid carefully in posi¬ tion. The characteristic advantages of such breakwaters, as com¬ pared with those of other types, are the facility and simplicity of their construction and repair. Their peculiar disadvantage is that they require a large volume of material in the substructure, most of which is not needed to resist wave action and serves only to support the comparatively small resisting parts of the work. In the locality under consideration random-stone breakwaters will, of course, be much easier to construct than crib breakwaters. They require no leveling under water, they are composed of only one kind of material, and the plant and method of construction are of the simplest charac¬ ter. The experience of the breakwaters at the entrance to Delaware Bay appears to show that works of this character, when properly con¬ structed, have sufficient stability to resist the action of very violent storms. The injuries inflicted upon crib breakwaters by storms and ice are of a very troublesome character. The cribs slide on their foundations or tip over, the timber and iron parts are twisted and broken, and the filling is washed out. These works require constant attention and frequent expensive renewal or repairs. The injuries inflicted by storms on the breakwaters in Delaware Bay have, on the contrary, been of the most trifling character and easily repaired at very small cost. It only remains to compare the cost of construction of these two kinds of breakwaters. At the time the method of crib construction was adopted for works in the Great Lakes lumber was cheap and plentiful and small stone suitable for filling could be readily obtained. On the other hand, economical plant and methods for quarrying, transporting, and depos¬ iting very large stones did not exist. The construction of random- stone breakwaters was at that time difficult and costly, the compar¬ atively small size of the stone which could then be economically quarried and deposited gave to the works little stability, and it was considered absolutely necessary that the substructure should have a very large cross section with gentle slopes exposed to the action of the waves. Under these circumstances the adoption of the crib DEEP WATERWAYS. 211 method of construction was doubtless most economical and advan¬ tageous. At the present time the conditions are entirely different. The cost of lumber has greatly increased, while the cost of quarrying, transporting, and depositing large stone lias greatly diminished. Moreover, experience in Delaware Bay and elsewhere seems to show conclusively that the cross section formerly adopted for random-stone breakwaters can be greatly reduced without sacrificing stability, if the work is constructed of large stone properly deposited. Assuming that a crib should have a width not less than its height, it can easily be shown that, at present prices and with the depths at Oswego, a crib breakwater with the crib work carried nearly to the bottom would be much more expensive than a random-stone break¬ water of the Delaware Bay type. There is, however, another method of construction in which cribs are employed which deserves careful consideration, and which is adopted in two of the breakwaters now in process of construction in Buffalo Harbor. These breakwaters consist, of a random-stone substructure rising to a level of 22 feet below the water surface and a crib superstructure the top of which is 12 feet above lake level. The method is evidently modeled upon modern European practice, which carries the superstructure down to a level at which wave action is supposed to have no appreciable effect upon the substructure. The use of cribs instead of concrete in blocks or in mass is, of course, much more economical and can be safelj* adopted in the fresh water of the lakes. In forming an approximate estimate of the difference in cost of breakwaters on the crib and random-stone methods at Oswego the following assumptions have been made: 1. The substructure for the crib breakwater will be formed of large rubblestone deposited upon the bottom. Its upper surface will be 22 feet below lake level and will have a width of 50 feet. The slopes will have an inclination of 1 on 1.3. Each cubic yard of substruc¬ ture volume is assumed to require 1.34 net tons of stone. At Buffalo a trench is excavated and filled with gravel to form a foundation for the work, as the hard bottom is covered with soft material of considerable depth. This is unnecessary at Oswego, where there is a hard bottom, and accordingly the cost of this part of the work is omitted from the estimate. 2. The superstructure will be the standard timber crib adopted at Buffalo. The cost of the timber and iron in 1 linear foot of this crib is estimated at $77.37, the cost of hemlock being assumed at $23 per M, white pine at $32 per M, and iron at 4 cents per pound. To this $1 is added to cover the extra cost of leveling the upper surface of the substructure under water. The assumed cost per linear foot of the superstructure without filling is therefore $78.37. The crib will contain 45.833 net tons of stone filling per linear foot. DEEP WATERWAYS. 9 1 9 W - — 3. The dimensions, method of construction, and estimated cost of the random-stone breakwaters are given in another part of this paper. The cost of the stone in place is estimated at $1.20 per net ton. Under tlie above assumptions the difference in cost per linear foot between the random-stone and crib breakwaters for all depths greater than 22 feet is given by the formula A = 2.0361) + 0.0211)' — 94.88, in which I) is the depth and A is positive when the cost of the random- stone breakwater exceeds that of the crib breakwater. At a depth of 34.4 feet the breakwaters of the two methods are ecpial in cost. This is almost exactly the average deptli on the breakwater sites beyond the depth of 22 feet, so that the cost would be practi¬ cally the same for the two methods. At depths less than 22 feet the random-stone method will be somewhat the cheaper. The cost of maintenance is, as before remarked, much less for random-stone than for crib breakwaters. For the reasons above stated the random-stone method of construc¬ tion described below has been adopted for the breakwaters proposed herein. It should be remarked, however, that in cases where it is desirable to utilize a breakwater for the purposes of mooring or unloading vessels, as at Buffalo, tin? crib superstructure should be adopted. METHOD OF CONSTRUCTION AND CROSS SECTION. The method of construction and the cross section adopted for the works under consideration are based upon the results of experience in the construction of random-stone breakwaters at the entrance to Delaware Bay. The old breakwater harbor in that locality was com¬ menced in 1828, and was completed in accordance with the original design in 1809. It was formed by two detached breakwaters separated by an interval of 1,390 feet. A new breakwater closing this interval was constructed between 1884 and 1898. Another extensive break¬ water, to form a national harbor of refuge, was commenced in 1897 and is still in process of construction. The methods followed in these works were first employed in the con¬ struction of the breakwater to close the interval in the old harbor. The history of this work is given in detail in the final report submitted to the (hief of Engineers on June 19,1899, and published in the Annual Report of the Chief of Engineers for 1899, page 1346. The experience and reasoning upon which the method of construction is based will be found in this report, and it is therefore unnecessary to repeat them here. Substructure .—This part of the work extends from the bottom to the level of low water. In construction it is best to raise it to a level of about 1 foot above low water to allow for settlement and to facil- DEEP WATERWAYS. 213 itate the construction of the superstructure. All the stone must be deposited within the limits of the low-water width, so that the mass may be permitted to form its own slopes under the action of the waves. The stones deposited to form the slopes should have a minimum weight of about 3 tons. For the interior smaller stones may be employed. The most economical and rapid method of depositing the material, up to a level of about 10 feet below the water level, is from bottom¬ dumping barges. Above that level steam derrick barges are more advantageous. The substructure should be allowed to settle at least a year before constructing the superstructure upon it. It should be carefully lev¬ eled as the superstructure is built. After the completion of the superstructure the slopes of the substructure should be examined and all holes should be filled with stones placed in position. The width of the substructure at the water level is determined by the dimensions adopted for the superstructure, and is in this case 27 feet. Experience shows that for the purpose of estimate the interior slope may be assumed as 1 on 1.3, and the exterior slope as 1 on 2. From the experience at the Delaware breakwaters, it is assumed that each cubic yard will require 1.34 net tons of stone. This esti¬ mate includes an allowance for settlement and misplacement. Superstructure .—The inner and outer walls of the superstructure should be formed of very heavy stones laid endwise to the sea, and the space between them should be compactly tilled with rubble. At the Delaware breakwaters the height of the superstructure above high water is about 9.5 feet, and the top width 20 feet. From experience at these breakwaters, however, it is considered perfectly safe to adopt for the Oswego breakwaters a height of 8 feet and a top width of 15 feet. The slope at the Delaware breakwaters is 1 on 0.7, both on the harbor and sea side of the works. This is very steep for the exterior slope, and although it has been found satisfactory at the old Delaware breakwater, a gentler slope might be found desirable where the exposure is great. A slope of 1 on 0.75 is adopted for the purposes of estimate. The width of the superstructure at the base will be 27 feet. The walls should be laid up in three courses. Each stone should be laid with its shortest dimension vertical, and in the outer wall should weigh not less than 9,000 pounds. For the inner w^all the stones may be of considerably less weight. The length of each stone should, however, be greater than its width, and it should be laid in the wall with its shortest dimension vertical and its longest dimension perpen¬ dicular to the axis of the breakwater. Assuming 1.4 net tons per cubic yard of superstructure volume, each linear foot of superstructure will require 8.71 tons of stone for its construction. 214 DEEP WATERWAYS. The number of tons of stone contained in one linear foot of the breakwater is given by the formula T=1.34D + 0.08187D 2 +8.711 in which D represents the depth of the water. MATERIAL. A good quality of limestone, weighing about 168 pounds t-o the cubic foot, can be obtained at Chaumont and Three-Mile Bay, N. V. This stone is believed to have sufficient durability for breakwater con¬ struction. It can be deposited in the work at a cost not exceeding $1.20 per net ton. This price is assumed in the estimate as the aver¬ age cost per ton in place for both substructure and superstructure. LIGHTHOUSE AND FOG SIGNAL. The harbor will require a light-house and fog signal, and their cost must be included in the estimates. They will, of course, be designed and constructed by the Light-House Board. From information kindly furnished by Lieut. Col. D. P. Heap, Corps of Engineers, engineer of the Third light-house district, it is believed that the cost of these works, including contingencies, may be estimated at about $60,000. ESTIMATE OF COST. The cost of constructing these works is estimated as follows: Stone in place, 941,931 net tons, at $1.20.. .. $1,130,317.20 Contingencies _... ... 114,682.80 Total for breakwaters ......... 1,245.000.00 Light house and fog signal. ._. 60,000. 00 Total . _ . ... . . ... 1,305,000.00 The profiles of the breakwaters and the details of the estimate are shown on figures 1 to 6, inclusive. OLCOTT. The existing harbor at this locality is situated at the mouth of Eighteen-Mile Creek, which is about 18 miles east of the mouth of the Niagara River. The exposure is much less than at Oswego. The movement of wave-worn material is from west to east. The bottom consists of sandstone overlaid by sand and gravel. The entrance to the harbor is formed by two parallel crib break¬ waters extending outward from the shore for distances of 850 and 873 feet, the width between them being about 200 feet. These works were constructed between the years 1867 and 1882. The harbor is excavated in the bed of the creek inside the shore line of the lake. The locations of the piers and harbors are shown on plate 17. The canal designed by the Board enters the lake a short distance DEEP WATERWAYS. 215 east of the existing piers. The channel connecting it .vith deep water in the lake will he about 3,680 feet long and 600 feet wide, and has a direction a little west of north. Piers for mooring and guiding vessels at the entrance are not provided, as there will be ample anchorage area and mooring facilities in the basin of the canal within the shore line. A crib pier on each side of the entrance to clearly mark its location is all that is considered necessary. The first lock is about 5,000 feet above the entrance. PLAN OF THE BREAKWATERS. The breakwaters are required to shelter the entrance to the canal from heavy seas and to protect the excavated channel from shoaling. As before remarked, it is not necessary to provide for anchorage areas outside the shore line. The breakwaters are designed to start from the shore at a distance of 300 feet from the channel, making the interval between them 1,200 feet. They are carried out parallel to each other for a distance of 1,840 feet from the shore, or half the distance from the lake entrance. From this point they converge on arcs of circles out to a depth of 35 feet, so as to give the entrance a clear width of 600 feet. This plan has the advantage of forming a stilling basin to moderately reduce the height of waves during storms. The aggregate length of the proposed breakwaters is 5,395 feet. Their locations are shown on figs. 7 to 12, inclusive. CHARACTER, CONSTRUCTION, AND CROSS SECTION OF THE BREAKWATERS. For the reasons fully stated in connection with the investigation of the breakwaters designed for Oswego it is proposed to construct these breakwaters of random stone, in accordance with the general method adopted for the breakwaters in Delaware Bay. The cross section and details of construction and the data employed will be the same as heretofore described for the works at Oswego, with the exception of the price of stone, which is taken at $1.35 instead of $1.20 per net ton, the works in this case being much farther from the quarries. ESTIMATE OF COST. The cost of constructing the works is estimated as follows: Stone in place. 388,670 net tons, at $1.35. $524,704.50 Contingencies. 55,295.50 Total for breakwaters. ... 580,000.00 Light-house and fog signal _ . 60.000. 00 Total. 640,000.00 The profiles of the breakwaters and the details of the estimate are shown on figs. 7 to 12, inclusive. DEEP WATERWAYS. 216 BREAKWATERS FOR A WATERWAY HAVING A DEPTH OF 21 FEET. The same general plans are adopted; hut the lengths of the break¬ waters, and consequently the cost of construction, may be consider¬ ably reduced. OSWEGO. The locations of the breakwaters proposed for the 21-foot waterway are shown on tig. 4, and the profiles and details of the estimate on figs. 5 and 0. ESTIMATE OF COST. Stone in place. 551,150 net tons, at $1.20... $661,380.00 Contingencies... .. ..... 66,120.00 Total for breakwaters . .. . ... ... 727,500.00 Light-house and fog signal .... 60,000.00 Total. ___ - .. 787,500.00 OLCOTT. The locations of the breakwaters proposed for the 21-foot waterway are shown on fig. 10, and the profiles and details of the estimate on figs. 11 and 12. ESTIMATE OF COST. Stone in place, 175,062 net tons, at $1.35...$236,333. 70 Contingencies..... 23,666.30 Total for breakwaters.. 260.000.00 Light-house and fog signal...... 60.000.00 Total 320,000.00 Respectfully submitted. C. W. Raymond, Lieutenant- Colonel, Corps of Engineers. The Hoard op Engineers on Deep Waterways. Appendix No. 4. SPEED OF SHIPS. SPEED OF SHIPS IN THE PROPOSED DEEP WATERWAY. The speed which ships can attain in the different sections of the proposed deep waterway is a question vital to a discussion of the facilities it would offer for economical transportation. For this pur¬ pose there is needed a definite schedule of speeds which a ship is likely to attain in each of its several sections. Previous discussions of the CM TOTAL LENGTH OF SUHfcRSr RIJCTURE 2630 feet X t o F c X CL O O •0 z UJ D W D — J D CM CD LO 05 r-H o o p W APPENDIX 3, FIG.3 CM CD ID rH o o p W 4 APPENDIX 3,fig. 03 CD LD a r-H o o Q W APPENOIX 3,FIG. CM CD m CD o o Q APPENDIX 3.FI6.6 O I- O T. CL 00 CO CM CO lO 05 o o p « APPENDIX 3, FIG. CM CD lD 05 T-i o o p W PROPOSEO BREAKWATER APPENDIX 3, FIG. 11 03 CD lD o> r-l o o p tn APPENDIX oj 0 u_ ro eeds of loaded boats, without tows , in St. Clair Flats Canal, in statute miles per hour. [Section of canal = 290 x 19.5 = 5.772 square feet. Deduct for shoal water near piers, say 72 square feet. Net cross-section = 5.700 square feet. Current in channel, 1.71 miles per hour.] Ship. Gross ton¬ nage. Net ton¬ nage. Draft. Beam. Area of immersed cross section. Ratio of cross section of ship to cross sec¬ tion of channel. Speed of ship cor¬ rected for current in the channel (V)J Back flow. > © ■±2 ce © ft w. With (W) or against (A) current. Fore. Aft. Ft. in. Ft. in. Sq. ft. Miles. Fred Mercur. 1.224 906 15 4 15 9 35 534 1:10.7 0. 48 0.67 7. 15 W. J. N. Glidden. 1,322 1,110 10 6 10 6 35 500 1:10.2 0.02 . 66 0.68 w. Topeka .. 1.370 1,111 10 10 16 10 36 588 1: 9.7 6.56 . 75 7.31 w. R. P, Fitzgerald_ 1,081 1.175 17 2 17 2 38 633 1: 9.0 9.41 1.18 10.59 A. Oceanica. 1,490 1.241 14 6 14 0 37 518 1:11.0 5.24 .52 5. 76 W. Vulcan.,. 1,759 1.300 17 4 17 4 38 643 1: 8.9 6.31 .80 7.11 W. Conemaugh. 1,009 1,453 10 0 16 0 36 558 1:10.2 5.83 .04 0.47 W. Lehigh ... 1,704 1.503 15 0 15 0 36 522 1:10.9 8.17 .83 9.011 A. R. E. Schuck. 1, 867 1.523 17 0 17 0 41 676 1: 8.4 0.89 .82 7.71 W. F. B. Prince. 2,047 1.547 14 3 14 3 42 577 1: 9.9 0.00 . 75 7.41 W. Joliet... 1,921 1,596 17 7 17 7 38 049 1: 8.8 8.54 1.09 9.63 w LaSalle... . 1.921 1.590 17 5 17 5 as 643 1: 8.9 5.83 .74 6.57 w Majestic. 1,985 1,609 10 9 17 0 40 600 1: 8.6 8.04 1.10 9.80 w. Mohawk_ 2.357 1,010 15 0 15 0 41 594 1: 9.6 8.80 1.02 9. 82 w. City of London_ 2,005 1,075 10 8 17 9 41 683 1: 8.3 7.63 1.05 8. 63 A. Mahoning. 2,189 1,704 15 0 10 0 40 620 1: 9.2 8.41 1.03 9.44 w. Republic . 2,310 1,877 17 6 17 0 40 680 1: 8.4 0.30 . 86 7.22 w. North Star. 2, 470 1,885 10 0 10 0 40 020 1: 9.2 0.81 .83 7.04 w. Northern Light_ 2.470 1,885 15 4 15 4 40 597 1: 9.5 11.41 1.35 12. 70 w. Onoko.... 2.104 1.9:33 10 0 10 0 38 608 1: 9.4 0.34 . 75 7.09 IV. Owego. 2.011 1.940 10 0 17 0 41 076 1: 8.4 7.86 1.00 8.92 w. Chemung .... 2,015 1.943 10 0 10 0 41 035 1: 9 (1 7 62 .95 8 57 A. Thomas Mavtham. 2.329 1.972 17 0 17 9 41 683 1: 8.3 5 16 .71 5 87 W 1 i'a-aba . 2,431 1,992 10 5 10 5 40 637 1: 9.0 0.21 .78 6.99 A. Kearsanre 3 092 2 721 17 (J 17 6 44 74S r 7 6 6 36 97 7 33 \Y 1 enobscot 2 864 17 0 17 6 44 726 1-78 5 11 w Senator . 4,048 3,178 17 2 17 2 45 ; 5( 1 1: 7.0 0. 70 1.02 7.72 w. Presque Tsle. 4.578 3.570 17 8 18 0 50 875 1: 0.5 6 b! 1.12 7.25 w. Crescent City_ 4.213 3.075 17 1 17 1 48 790 1: 7.2 7. 36 1.18 8.54 w. Passenger boats: Northland. 2.339 13 14 44 612 1: 9.3 9 99 1.20 11 19 w. Arundel (fast- est observed). 257 26 15.41 w. Idle wild. 284 5 8 5 8 26 134 1:42.5 13.02 .31 13.33 w. Note.— Sis inches deducted from maximum draft in calculating crosssection of ships. DEEP WATERWAYS. 227 The Suez Canal affords a precedent even more pertinent. In the original canal, having a bottom width of 72 feet, a speed of G.2 statute miles per hour was fixed, rather with regard to the effect of wash on the channel banks than to safety of the ship itself. If it were not for the injurious effects of wash on the banks a higher speed limit would probably have been fixed. There can be no doubt that a speed of 8 miles per hour in the regular channel of the deep waterway is relatively less than the speed permitted in the narrow channel of the Suez Canal. In the 300-foot channels excavated in open water in the St. Marys River, boats formerly ran 12 to 13 miles per hour. This is a crowded navigation, the number of boats per day being about 100. Collisions were frequent, and it was thought necessary to establish and enforce rules limiting speeds to 9 miles per hour. In the deep waterway the speed of the type ship in the 30-foot channel of 300-foot bottom width would be 10 —1.44 = 8.56 miles per hour (see Table III) and in the 21-foot waterway of 300-foot bottom width 10 —1.36 = 8.64 miles per hour (Table Y), both within the limit established in the St. Marys River. An inquiry addressed to Mr. Joseph Ripley, the assistant engineer in charge of the St. Marys River and St. Marys Falls Canal, elicited the following reply, extracted from a letter dated April 13, 1899: The highest permissible speed through the present cuts at Little Rapids and Hay La .e depends on number, size, and draft of boats, and whether single or with tows. For the commerce of last season, which averaged nearly 110 boats per day for that part of the river, with about 10 per cent of them over 400 feet in length, the speed should not exceed 9 miles per hour, slowing to about 6 miles when meeting boats. Under the above conditions there were two collisions in the Little Rapids cut last season. For width of (300 feet, with earth banks, the speed could be safely increased to 11 or 12 miles per hour, checking to 8 miles when meeting tows. The channel at the Little Rapids, to which Mr. Ripley refers, has proved troublesome on account of the strong current through it and cross currents caused by escape of water into lateral channels. Cur¬ rents as strong as these would occur in the deep waterway only dur¬ ing very unusual floods. These floods, rare as they are, occur almost without exception during the spring break-up, before tiie opening of lake navigation. The experience at the Little Rapids does not apply to the movement of ships in slack water or in very light currents— the latter in the conditions which will exist in the deep waterway—and under such conditions the examples which have been cited support strongly the speeds calculated for the type ships in their respective waterways. These speeds being within the limit of successful prac¬ tice need no further reduction. The ship of 19-foot draft in the 30-foot channel will have at least 10 feet of water under its keel when in motion in the waterway, and will steer as well, practically, as in open sea. 228 DEEP WATERWAYS. The speeds thus far discussed refer to a straight channel, with no meeting ships. The}’ need correction for reduced speed at meeting points and on curves. (G) Reduction of speed at meeting point's .—On the Suez, the Man¬ chester, and other ship canals, when two large ships are about to meet, one ties up until the other passes. 'This will not be necessary in the greatly wider deep waterway. The speed of both ships, however, will have to be reduced in the narrow channels. The proper speed at the meeting point will depend obviously on the width of the channel, the greater the width the greater the permissible speed at the meeting point. In a channel 600 feet wide no reduction of speed will be neces¬ sary. In the 300-foot wide channels of the St. Marys River a reduc¬ tion of speed of from !) to 6 miles per hour has been found sufficient, but in most places the channel is dredged through shoal water, and the escape of the water laterally from the channel may possibly facili¬ tate the passing of ships. While it has not been proven that the movement of ships is thus facilitated, it is believed some weight should lie given to this consideration. It will be assumed, therefore, that the speed would be reduced one-lmlf at the meeting point in a channel 200 feet wide at bottom. In wider channels the reduction will be taken as diminishing uniformlv with increase of bottom width, becoming zero when the bottom width is 600 feet. It would seem that if two ships of 19 feet draft should meet in the 30-foot channel of the regulated Mohawk a greater speed at meeting points would be permissible. This is certainly true, but to make it available it would be necessary to have different • regulations for dif- ferent classes of ships, which would be objectionable. The same law of reduction is, therefore, applied in this case as in the others. In order to measure the effect of delays at meeting points it is necessary to assume a navigation period and a traffic. For the pres¬ ent discussion the number of days of navigation per year will be assumed at 237. In calculating the amount of water required for the supply of the summit level it was assumed that the annual tonnage through the canal would be 25,000,000, carried in 8,333 ships, giving 35.2 ships per day, or 1 in 41 minutes. These figures are for the 30-foot waterway. For the 21-foot waterway the same traffic may be assumed to be carried in 10,000 ships, giving 42.2 ships per day, or 1 in 34 minutes. The maximum delay from meetings will occur if all ships are met singly, and if this is assumed the resulting delay will be greater than the truth and on the safe side. The number of ships assumed is also on the safe side, because it represents the full capacity of the canal with ships averaging 3,000 and 2,500 tons, respectively, in the two channels. The full capacity is not likely to be reached at once and the ships will probably be larger, thus reducing the number of ships and number of meetings for the assumed annual traffic. When ships are approaching a meeting point they will begin reduc- DEEP WATERWAYS. 229 ing speed when some distance apart, this distance depending on the speed at which they are moving and the reduced speed to be attained. In order to ascertain the time required to reduce or increase speed an informal inquiry was made of Mr. Ripley, who prepared a form of report as a circular for distribution among masters of vessels. About ICO of these reports have been made and furnished to the Board by Mr. Ripley. From these reports those of 17 of the larger boats, carry¬ ing loads and moving singly (without tows), have been used to deter¬ mine the rate of change of speed. The reports show both time and estimated distance required to attain full speed from a stop and also to stop from full speed; but in many cases the ships’ wheels were used to stop speed, which would generally be impracticable in a nar¬ row channel. Rejecting those cases where the use of the wheel was noted or was obvious, the rate of reduction of speed seems to be about the same as the rate of increase. For the reason that there still remained some doubt as to the use of the wheel in stopping it has seemed best to base the rate of change of speed wholly on the records relating to increase of speed. As might be expected, the estimates of time and distance made In¬ different sailing masters vary greatly. Without exact observations it is not easy to decide when a ship has attained full speed. It is believed, however, that a sufficient number of observations have been taken to eliminate any important error. Taking the entire set of 17 observations, the average rate of increase of speed from stop to full speed is 1 mile per hour while the ship traverses 440 feet. As the time was also noted, the mean time required for an increase of speed of 1 mile per hour is found to be nine-tenths of a minute. In this interval a ship moving 12 miles per hour—the average full speed of the boats observed—would traverse a distance of 306 feet. This agrees well enough with the former determination. Rejecting 8 of the observations—tin* lowest four and the highest four—the remaining 0. reduced in the same manner, give averages of change of speed of 1 mile per hour in 350 and 336 feet, respectively. This result is probably more nearly correct than the former. In order to be on the safe side the distance of 400 feet is here accepted as required for a change of speed of 1 mile per hour. This conclusion is based on the assumption that the change of speed varies directly with the distance traversed. It may seem that the higher the speed the greater tlie distance to be traversed for an increase of 1 mile per hour. This is very probably the case. On the other hand, when the engines are stopped it is probable that the rate diminishes more rapidly at first than after the speed has been much reduced. If this is true, the two variations from the mean tend to offset each other, and the average distance deduced above will not be much in error for the entire cycle of reduction and increase. In order to produce the minimum delay at a meeting point the ship’s 230 DEEP WATERWAYS. engines should be stopped at the exact moment required to permit the proper reduction of speed to be attained when the bow of the ship is opposite the bow of the meeting ship. The engines should be started at full speed at the moment when the ships are clear and not before, or when each ship has moved its length. For the type ship in the 30-foot channel this distance is 550 feet. It will not be practicable, however, to realize this ideal movement. The pilot will not be able to .judge with the required precision either the distance to the other boat or the speed of his own, and it is probable that the required reduced rate will be attained before the ships meet. As a basis for calculation, it is assumed that after the required reduction of speed has been attained the ship will continue under check at this rate for a distance of 1,000 feet, and will then proceed with engines at full speed. This assumption will be applied to both type ships. The cal¬ culations are made in the following manner: Required the delay to a ship in the standard 30-foot, channel: The full speed in this channel is 8.04 miles per hour. (See Table III.) The reduction of speed will be 600—203 4x 8.04 X (j(jQ_2Q0 = 3.99 miles per hour, and the minimum speed at meeting point will be 8.04—3.99 = 4.05 miles per hour. The speed is to be reduced from 8.04 to 4.05 miles per hour and to be maintained at this rate for a distance of 1,000 feet and then increased to 8.04 miles per hour. Feet. Distance to be traversed while reducing speed, 400x3.99__ 1,590 Distance to be traversed at minimum speed...... 1,000 Distance to be traversed while increasing speed 400x3.99. 1,598 Total...... 4,192 Minutes. Time required for 1,590 feet at 4 (8.04+4.05) miles per hour..3.00 Time required for 1,000 feet at 4.05 miles per hour....2. 81 Time required for 1,590 feet at (8.04+4.00) miles per hour......3.00 Total .. ...... 8.81 Time required to traverse 4,192 feet at 8.04 miles per hour.... 5.92 Time lost... .... 2, 89 This loss of tiir^e will occur at each meeting in the given channel. According to the assumption, meetings will occur every 41 minutes. 2.89 The loss of time is therefore of the total. The distance passed 41 — 2.89 38.11 over by the ship will be — n - -rr-of the distance it would move J 1 41 41 w APPENDI X 4. FIG. Od CD lD a r-H o o p W DEEP WATERWAYS. 231 in 41 minutes if there were no meetings. Its mean speed is therefore 38.11 , , —ot the speed it would have if there were no meetings, or 8.04 x 38.11 41 7.47 miles per hour = average speed of ship. In this manner the speeds in column 10 of Table III have been cal¬ culated. In column 10 of Table TV are the speeds in the 30-foot waterway in the Mohawk Valley if it forms part of a 21-foot naviga¬ tion. In column 10 of Table V are the speeds in the 21-foot waterway. The speeds in Tables IV and V are calculated in precisely the same manner as in Table III, except that meetings for the 21-foot naviga¬ tion are taken as occurring once in 34 minutes instead of once in 41 minutes. These are the average speeds on tangents and require correction on curves. The tabulated speeds in column 10 of the tables are reproduced graphically on fig. 1, where the abcissas are the bottom widths of channel and the ordinates are speeds. Table III.— Speed, in statute miles per hour, of ship 5^0 feet long over all, 58 feet beam, 27 feet draft, in 30-foot channel. 1 Bottom width of channel *> A 3 r 4 Speed of ship in open sea. 5 Loss of speed on en¬ tering shoal water. 0 Speed in shoal water of unlimit¬ ed width. ( Retarda¬ tion due to back flow in restrict¬ ed chan¬ nel. 1 8 Full speed on tan¬ gents. 9 Reduc¬ tion of average speed caused by meet¬ ings. 10 Average speed of ship on tan gents. CO £ 7,990 5.10 12.5 2. 5 10 1.90 8.04 0.57 7.47 210 8,200 5. 24: 12.5 2.5 10 1.91 8.09 .54 7.55 220 8,500 5. 43 12 5 2.5 10 1.84 8.16 .52 7.04 230 8,800 5.02 12.5 2.5 10 1.78 8.22 .49 7. 73 240 9, loO 5.81 12.5 2.5 10 1.72 8.28 .47 7.81 2:50 9.400 6.00 12.5 2.5 10 1.67 8.33 .44 7.89 200 9.700 6.19 12.5 2.5 10 1.62 8 38 .41 7.97 280 10,300 6.58 12.5 2.5 10 1.52 8.48 .37 8.11 300 10.900 6. 90 12.5 2.5 10 1. 14 8.56 .33 8. 23 320 11.500 7.34 12.5 2.5 111 1.36 8.64 .29 8.35 340 12, 100 7. 73 12.5 2.5 10 1.29 8.71 .25 8. 40 300 12, 700 8.11 12.5 2.5 10 1.23 8.77 *>» 8. 55 380 13,300 8.49 12.5 2.5 10 1. 18 8.82 .19 8. 03 400 13,900 8.88 12.5 2.5 10 1.13 8.87 .16 8. 71 420 14.500 9.20 12.5 2.5 10 1.08 8.92 .14 8. 78 440 15,100 9.64 12.5 2.5 10 1.04 8.96 .11 8.85 400 15,700 10.03 12.5 2.5 10 1.00 9.00 .09 8.91 470 16,000 10.22 12.5 2 5 10 .98 9.02 .08 8.94 600 19,900 12.71 12.5 2. 5 10 .79 9.21 .00 9.21 800 25,900 10.54 12.5 2.5 10 .60 9.40 .00 9.40 1,000 31,900 20.37 12.5 2.5 10 .49 9.51 .00 9.51 1 Equals velocity of back flow. 232 DEEP WATERWAYS Table IV. — Sjieed, in statute miles per hour, of ship 450 feet long over all, 52 feet beam, ID feet draft, in 30-foot channel. 1 Bottom width of channel. 0 A 3 r 4 Speed of ship in open sea. 5 Loss of speed on em tering shoai water. 6 Speed in shoal water of unlimit¬ ed width. i Retarda¬ tion due to back flow in restrict¬ ed chan¬ nel. 1 8 Full speed on tan¬ gents. 9 Reduc¬ tion of average speed caused by meet¬ ings. 10 Average speed of ship on tan¬ gents. Feet. 203 7,090 8.09 12.5 1.25 11.25 1.39 9.86 0. 76 9.10 210 8.200 8.30 12.5 1.25 11.25 1. 35 9.90 .73 9.17 220 8,600 8.60 12.5 1.25 11.25 1.31 9.94 .70 9.24 230 8,800 8.90 12.5 1.25 11.25 1.26 9.99 . 66 9 33 240 9,100 9.21 12.5 1.25 11.25 1 22 10.03 .63 9. 40 250 9,400 9.51 12.5 1.25 11.25 1.18 10. 07 .59 9.48 260 9.700 9.82 12.5 1.25 11.25 1.15 10.10 .56 9.54 280 10,300 10.42 12.5 1.25 11.25 1.08 10.17 . 49 9.68 300 10,900 11.03 12.5 1.25 11.25 1.02 10.23 .44 9.78 320 11.500 11.64 12.5 1.25 11.25 .97 10.28 .39 9. 89 340 12,100 12.24 12.5 1.25 11.25 .92 10 . as .34 9.99 360 12. 700 12.85 12.5 1.25 11.25 .88 10.37 .30 10.07 380 13,300 13. 46 12.5 1.25 11.25 .84 10.41 .25 10.16 400 13,900 14.07 12.5 1.25 11.25 .80 10. 45 >>> 10.23 420 14.500 14.67 12. 5 1.25 11.25 . 77 10.48 .18 10.30 440 15,100 15.28 12. 5 1.25 11.25 .74 10.51 . 15 10.36 460 15,700 15.89 12.5 1.25 11.25 .71 10.54 .12 10.42 470 16,000 16.19 12.5 1.25 11.25 .70 10.55 .10 10. 45 600 19,900 20.14 12. 5 1.25 11.25 . 56 10.69 .00 10. 69 800 25,900 26.21 12. 5 1.25 11.25 .43 10.82 .'JO 10. 82 1.000 31,900 32.29 12.5 1.25 11.25 .35 10.90 .00 10. 90 1 Equals velocity of back flow. Table V. — Speed, in statute miles per hour, of ship 430 feet long over all. 52 feet beam, 19 feet draft, in 21-foot channel. 1 Bottom width of channel. 2 A 3 r 4 Speed of ship in open sea. 5 Loss of speed on en¬ tering shoal water. 6 Speed in shoal water of unlimit¬ ed width 7 Retarda¬ tion due to back flow in restrict¬ ed chan¬ nel. 1 8 Full speed on tan¬ gents. 9 Reduc¬ tion of average speed caused by meet¬ ings. 10 Average speed of ship on tan¬ gents. Feet. 215 5.497 5.54 12.5 2.5 10 1.81 8.19 0.64 7.55 220 5,602 5.67 12.5 2.5 10 1.76 8.24 .63 7.61 2:10 5,812 5.88 12.5 2.5 10 1.70 8.30 .60 7. 70 240 6, 022 6.09 12.5 2 .5 10 1.64 8. 36 7.79 250 6,232 6.31 12.5 2.5 10 1.58 8.42 .53 7.89 260 6.442 6.52 12.5 2.5 10 1.53 8.47 .50 7.97 280 6.862 6.94 . 12.5 2.5 10 1.44 8.56 . 45 8. 11 300 7,282 7.37 12. 5 2.5 10 1.36 8.64 .40 8 24 320 7.702 7.79 12.5 2.5 10 1.28 8.72 .35 8.37 340 8.122 8.22 12. 5 2.5 10 1 22 8.78 .31 8.47 360 8.542 8.65 12. 5 2.5 10 1.16 8.84 .27 8. 57 380 8.962 9.07 12.5 2.5 10 1.10 8.90 .23 8. 67 400 9.382 9. 50 12.5 2.5 10 1.05 8.95 .19 8. 76 420 9.802 9.92 12.5 2.5 10 1.01 8.99 .16 8.83 44(1 10,222 10.34 12.5 2.5 10 .97 9.03 .14 8.89 460 10,642 10. 77 12.5 2.5 10 .93 9.07 .11 8.96 470 10,852 10.98 12.5 2.5 10 .91 9.09 .10 8.99 600 13,582 13. 75 12.5 2.5 10 .73 9.27 . 00 9.27 800 17,782 18.00 12.5 2.5 10 .56 9. 44 .00 9.44 1,000 21.982 22.25 12.5 2.5 10 .45 9.55 .00 9.55 1 Equals velocity of back flow. (7) Reduction of speed on curves .—The speed of the ship will be less on curves than on tangents, because the steering will be a little more difficult. Some compensation is made for this in designing the I DEEP WATERWAYS. 233 waterway by increasing the width on curves. The reduction of speed will be less as the width of the channel increases. The speed of meet¬ ing ships may also be reduced. This is hardly likely to be the case in the narrow channels, as the speed already allowed for is barely sufficient for steerage way, but in the wider ones the reduction of speed on curves at meeting points may be appreciable. The subject does not admit of analysis; there are no reliable precedents, and tlie best that can be done is to make empirical rules. The following are proposed: First. For channels of standard width, reduce the speed from Tables III IV, and V by 1.5 x degree of curve. Second. It is not believed that any reduction of speed will be required in channels 600 feet wide. For intermediate widths, multi¬ ply the reduction obtained by the preceding rule by the ratio: 600 — Bottom width of channel in question. 600— Bottom width of standard channel. SUMMARY. In the foregoing discussion two type ships have been taken, one 540 feet long, 58 feet beam, and 27 feet draft; the other 480 feet long, 52 feet beam, and 19 feet draft. The larger ship is assumed to be sea¬ going; the smaller ship is assumed to navigate tlie lakes and deep waterways to tide water, and both are to have a speed of 12.5 statute miles per hour in open sea. The discussion then takes up the principal causes of retardation when a ship passes from open sea into a restricted channel and treats them singly, making use of recorded observations where possible and of theoretical discussion where observations are wanting. The retardation when passing from deep open sea to shoal water of unlimited width is determined from observations in Lake Huron and Lake St. Clair. The reduction accepted for the discussion is somewhat greater than that observed, and, therefore, is on the safe side. The retardation due to back flow and the retardation due to end resistance are treated theoretically. The relation between speed of ship and injurious washing of slopes is treated theoretically, and the results checked by experience on sev¬ eral artificial waterways. At this point the difficulty of steering a ship in a narrow channel and the danger of grounding on the side slopes is taken up and the conclusion reached, by reference to experience in other waterways, that there will be no material difficulty in this respect on tangents of the deep waterway. The reduction of speed, which will be necessary when ships meet, 234 DEEP WATERWAYS. is discussed on a liberal basis of reduction of speed, and the time lost is determined from the results of recent observations in the Great Lakes. The reduction of speed when passing from a tangent to a curve depends as much on the skill or timidity of the pilot as on anything else, and can not be analyzed. Observations are lacking, and a purely arbitrary assumption is made. The theoretical discussions are extremely elementary and might be developed. It is not believed, however, that their usefulness for the present purpose would be enhanced. VALUE OF THE RESULTS. The theoretical portion of this discussion covers so large a part of the subject and is necessarily so incomplete, both in scope and treat¬ ment, that the results would have very little value unless checked by observations. An examination of the speed curves (fig. 1) will facili¬ tate an estimation of value of the results, and it will be advantageous to study each curve separately. First. Ship 480 by 52 by 19 feet draft in the 21-foot channel. A channel 1,000 feet wide is obviously the equivalent of a channel of unlimited width as regards the movement of a ship. The Lake St. Clair observations show that in the shallow water of that lake, where the depth of water exceeded by about 24 feet the draft of the boat, the loss of speed was considerably less that 20 per cent; 20 per cent was taken for this discussion to be on the safe side. The theo¬ retical determination of retardation due to back flow gave a loss of speed due to the narrowing of the channel to 1,000 feet of 0.45 mile per hour. There can be little doubt that this is excessive. The same cal¬ culation shows a loss of speed of 0.73 mile per hour when the channel is narrowed to 600 feet, and there can be no doubt that the allowance is sufficient. It is also obvious that the permissible speed will be reduced in increasing ratio as the channel is narrowed, and therefore the curve should be convex upward. If, however, this were not true, and the reduction of speed varied in direct ratio to decrease in width, the curve from its lowest point to, say, the point corresponding with a channel width of 600 feet would be a straight line. The maximum deviation of this straight line from the platted curve is less than 0.4 mile per hour. The speed for a bottom width of 300 feet is only 0.64 more than the speed allowed by regulations in the crowded St. Clair Flats Canal, and is much less than frequently attained there. This curve, then, being checked at its lowest point by observation—being shown by observation to be a little too low at its highest point and having a logical direction between with no possibility of material error—is entitled to confidence. Second. Ship 480 by 52 by 19 feet draft in the 30-foot channel. The difference in the position of this curve and the preceding one is due to the smaller reduction of speed for limitation of channel depth. DEEP WATERWAYS. 235 This ship would have 11 feet of water under the keel, and it should suffer little reduction of speed on this account. The value taken, 10 per cent, must he considerably excessive. With such a depth of water under the keel the ship would steer well. The lowest point of this curve is not so well connected with observations as the preceding one; but it may be remarked that in the Manchester Canal, which has a bottom width of only 120 feet, much higher speeds are permitted and attained by light-draft boats. While this curve is not as thoroughly checked by observation as the preceding one, it is believed to be trustworthy. Third. Ship 540 by 58 by 27 feet draft in the 30-foot channel. The relation between this type ship and its channel are very nearly the same as the relations of the smaller type ship to the 21-foot channel. The reduction of speed on entering shoal water of unlimited width is taken at 20 per cent, which is doubtless somewhat excessive. The calculated retardation due to back flow in a channel 1,000 feet wide at bottom, 0.49 mile per hour, is doubtless somewhat excessive; in a channel 600 feet wide, the calculated retardation is 0.79 mile per hour, which is doubtless adequate. In a channel 300 feet wide, the calculated speed is 8.56 miles per hour, which is less than frequently attained in the St. Clair Flats Canal. It is obvious that this curve should also be convex upward. Taking the portion included between abscissas corresponding to bottom width of 203 and 600 feet, the maxi¬ mum deviation of the speed curve from a straight line is less than 0.4 mile per hour-. The error in the curve must be less than this if the speeds at the extremities of the curve are correct. The curve agrees well with the speeds permitted in the Suez Canal after allow¬ ing for the difference in channel width, and corresponds so closety in form to the curve for the smaller type in the 21-foot waterway that it is presented with considerable confidence in its correctness. Respectfully submitted. Alfred Noble. The Board of Engineers on Deep Waterways. Appendix No. 5.—Part 1. COMPARISON OF WATERWAYS. The value to commerce of any waterway from the Lakes to tide water depends upon the amount and nature of the tonnage it is capable of passing, the speed at which rest ricted portions can be safely navigated, the sailing time between terminals for ships best adapted to its use, and the cost per ton-mile for transporting freight under normal conditions. From these considerations it is evident that from both economic and engineering points of view the importance of the waterway will DEEP W A T E R \V AYS. 23(5 depend upon ils depth, ratio of cross section to that of the type ship which will use it, the stability of its banks, its length and location, the cost to construct, operate, and maintain it, and the indirect bene¬ fits to the territorry tributary to the route. The waterway will consist of two distinct divisions. First, the lake waterways, connecting channels and harbors, and, second, the ship canal from the Lakes to the Atlantic. It has been shown elsewhere that the dimensions of the latter should be such as to furnish safe and easy transit for the type of ships best adapted for the waterways connected, from which it is evident that whatever depth is to be established for the lake harbors and water¬ ways should be made the basis of the design for any canal which may be constructed to connect such channels with the seaboard. The cost to carry a ton of freight 1 mile in the lake traffic is a function of the original cost of improvements, annual expenditures for operation and maintenance, and of the speed, carrying capacity, fixed charges, and daily expenses of the ship in which transported. It has been found from actual experience that the economical carry¬ ing capacity of ships depends upon the depths of harbors and chan¬ nels to be used, and upon the speeds at which the waterways are to be traversed. For the type of ships best adapted for lake traffic, speed in excess of about 124- statute miles per hour in the open lake involves a corre¬ sponding decrease in freight capacity, or an increase in amount of fuel per mile run, either of which will increase the rate per ton mile of the freight carried. The economical carrying capacity of a ship increases rapidly with the draft which the waterways will allow, and, since the cost of chan¬ nel and harbor improvements increase rapidly with the depths to be maintained, it is important that the depth of lake channels be deter¬ mined at which the cost of transportation of the lake freight will be a minimum for ships best adapted for the business and safe navigation. The depth of water at mean stage at the foot of Lake Huron, through Lake St. Clair, and at the head of Lake Erie is from 20 to 21 feet, and, therefore, for all navigable channels over 21 feet deep the cost per foot of improvement and for maintenance will l)e greatly in excess of that for waterways of less depth. The conditions at the entrance of the lake harbors are such as to produce a similar effect on cost of improvement for different depths. The cost in the past for deepening all the harbors and waterways of the lake system has been about $5,000,000 per foot of increase. From the best available data, it is estimated that for depths between 21 and 30 feet the cost for similar improvement will be about $7,000,000 per foot. Assuming that the Government pays 3 per cent for the money expended on improvements, and one-half of 1 per cent of cost for maintenance of improved channels, the annual increase in fixed DEEP WATERWAYS. 237 charges for eacli additional foot depth of waterways would be approx¬ imately -$175,000 for channels less than 21 feet deep, and $245,000 for channels between 21 and 30 feet deep. The actual freight carried on the lakes in 1809 was not far from 40,000,000 tons, making the average increase for the cost per ton trans¬ ported, $0.0044 and $0.0061, respectively, for fixed charges due to 1 foot improvement of channels. The actual average rate per ton charged for freight passing Sanlt Ste. Marie in 1899 was $0.87, from which it appears that the cost of improving the harbors and channels less than 21 feet deep will increase the rate of transportation one-half of 1 per cent, and for channels between 21 and 30 feet deep, seven-tenths of 1 per cent for each foot of improvement. Ships capable of steaming 124 statute miles per hour in the open lake make from twenty-three to twenty-five round trips between Buffalo and Duluth during an average season of navigation, or a round trip once in ten days, three days of which are due to terminal deten¬ tions. Referring to the cost and running expenses of type carriers best adapted to waterways of 21 and 30 feet depths, given in table on page 253, and assuming that these steamers make the same number of round trips per season in the respective waterways as the larger ships now in the service, the cost per ton-mile for transportation, not including sinking fund and profit to shipowner, would be 0.350 mills for the 21-foot channel and 0.338 mills for the 30-foot channel, from which it appears that the decrease in cost of transportation, due to increase of draft of ship, would amount to about one-third of 1 per cent of the transportation rate for each foot increase, indicating that for the type of ship considered there is no economy in improving lake channels where 21 feet deep, and that, with the depth of waterway increased to 30 feet, the interest on cost of improvement, with annual maintenance, would be at least twice the amount that would be saved to the producer by lower rates of transportation likely to result from deeper channels. It is true that the fixed charges arising from interest on the cost of construction, with annual expense for operation and maintenance, are paid by the Government; but whether paid from Government revenues or by a toll on traffic, the net result is the same. A draft of 19 feet is necessary for vessels best adapted for the com¬ merce and safe navigation of the lakes, and since the free movement of ships in shallow water requires at least 2 feet of water between the keel and bottom where the shallow channel is of considerable length, it appears that 21 feet is about the least depth which can be estab¬ lished for the lake waterways consistent with securing a minimum rate of transportation for ships best adapted for the lake service. The traffic capacity of the lake system of waterways is practically unlimited except at Sault Ste. Marie, where additional locks will be needed with the development of new commerce. The present structures at Sault Ste. Marie are adapted for passing 238 DEEP WATERWAYS. ■ the ships which will navigate the lakes, if the ultimate depth of har¬ bors and waterways be fixed at 21 feet, and it is probable that, with such limit definitely fixed, the construction of ships in the future would be adapted to the conditions and their dimensions become more uniform than at present. A system of 21-foot channels can be easily completed in the near future by regulating the levels of Lake Erie and Lake St. Clair above the present mean stage and improving the St. Marys River and the entrances of lake harbors to correspond. If, however, a greater depth than 21 feet be fixed for the navigable channels the cost will be enormously increased, a long time will be consumed in executing the work, and millions of dollars’ worth of vessel property will become obsolete by being forced out of business by larger freight carriers as the depths of connecting channels are gradually i ncreased. The otie thing above all others essential for securing a fixed mini¬ mum rate of transportation on the Great Lakes is to have the final depth to be given waterways definitely fixed, and to obtain such depth at the earliest possible date. From an engineering point of view it is also essential that final dimensions be fixed for the connecting channels of the lake water¬ ways, for the reason that each additional foot of depth given the channel decreases the total fall between the lakes and diminishes the depth of water on the lower miter sills of the locks at Sault Ste. Marie a like amount. If there is no economy in making the lake channels more than 21 feet deep for domestic traffic, the only legitimate reason for water¬ ways for more than that depth between the lakes and tide water would be to allow the best types of ocean freight carriers to reach the lake ports. For this purpose it would be necessary to make only the terminal lake harbors 30 feet deep, the cost of which, with that for deepening the connecting channels from 21 to 30 feet, would probably be about $50,000,000. A 30-foot waterway from the lakes to the Atlantic will cost approx¬ imately $100,000,000 more than a ship canal 21 feet deep, and there¬ fore if the producers of the United States are to be benefited by such a waterway the resulting direct and indirect annual benefits must exceed the difference in the fixed charges of the two routes due to interest on cost of construction and expenditures for maintenance and operation, which will amount to about $5,000,000 a year. The capacity of these waterways has been estimated at about 25,000,000 net tons per year, of which it is probable that less than 6,250,000 tons would be foreign commerce, from which it appears that to establish 30-foot navigation between the Atlantic and lake ports for foreign commerce will cost the Government about 80 cents for every ton of export freight transported when the canal is utilized to its full capacity. DEEP WATERWAYS. 239 It is true that increased transportation facilities will develop new industries and new commerce, but when we consider that it will be many years before the traffic of the waterway will reach 25,000,000 tons annually it is not reasonable to expect that the direct and indi¬ rect benefits arising from foreign trade of one-fourth this amount through a 30-foot waterway will exceed $5,000,000 the returns to be obtained with a 21-foot ship canal, even if all foreign commerce be subject to transfer delays and charges at New York. The maximum capacity of a 30-foot waterway with single locks 740 feet long and 80 feet wide is but little greater than for a 21-foot canal with smaller locks, for the reason that the time required for passing ships through locks increases with the dimensions of lock chamber, and while a larger class of vessels would use the deeper waterway, the number which could be passed during the navigation season would be less. If necessary to design a waterway of greater capacity than the two which have been considered, it will be far more economical to dupli¬ cate the single locks of the 21-foot canal than to increase the dimen¬ sions of canal prism to allow passage of ships of larger size. If the locks for the 21-foot waterways should be given larger dimen¬ sions than required for the type carrier boat adopted for the channel, for the purpose of developing shipbuilding industries on the lakes, the annual traffic capacity would be decreased and the relative capac¬ ity of the two channels become more nearly proportional to the ton¬ nage of the ships which would use them. From a careful study of the economical and safe speeds of ships in restricted waterways it is found that the cross section of the channel should be about five and five-tenths times that of the larger ships for which designed, which ratio has been used in fixing the dimensions of the prisms of the waterways. The dimensions of a ship canal to furnish transportation between terminals at a minimum rate must be such that the earning capacity of type carriers in time saved by greater speed in canals of larger sec¬ tion will be less than the amount of the interest on difference in cost and expense of maintenance, and greater for the time lost by slower speed in canals of smaller section than the amount saved in fixed charges for interest and maintenance. For the class of vessels which maybe expected to utilize waterways 21 and 30 feet deep between the lakes and tide water, the areas of the cross sections of canal prisms to fulfill the above conditions would be approximately 5,000 square feet in rock and 5,500 square feet in earth for the 21-foot channel, and 7,500 square feet in rock and 8,000 square feet in earth for the 30-foot channel. The prism for the 21-foot channel adopted by the Hoard has a bottom width of 215 feet, making the area of cross section 5,407 square feet, and corresponds closely with that of the St. Clair Flats 240 DEEP WATERWAYS. Canal, through which, there is an annual traffic of over 30,000,000 tons, transported in steamers and barges which traverse the canal with speeds ranging from 5 to 15 miles per hour. The Government regulations require that steamers shall not exceed a speed of 8 miles per hour in the canal, and from actual observation it has been found that the larger steamers, without tows, maintain speeds of from 7 to 10 miles per hour, depending upon whether running with or against the current of 1.71 miles velocity in the canal. In view of the great number of vessels traversing the canal, its exposure to the full force of the wind from all directions, its short length, and the danger from cross currents at its terminals, it is desir¬ able that the canal should be made at least 000 feet wide. The canal has practically the same surface width as that adopted for the standard cross section of the 21-foot waterway, but as the canal is excavated between piers its bottom width and cross section are a little greater. The dimensions of prism for the 21-foot waterway are intended for locations where there will be no currents in the chan¬ nel of sufficient velocity to materially interfere with navigation, and in rivers where such currents do exist estimates have been made for wider channels. Referring to Tables III and V of Appendix No. 4, it will be seen that the type ships of 19 feet and 27 feet draft, capable of steaming 124 miles per hour in the open lake, are estimated as being able to maintain 7.55 miles and 7.47 miles per hour, respectively, on tangents in waterways 21 and 30 feet deep. These speeds are based upon the theory of retardation of ships by shallow water and restricted waterways, discussed in Appendix No. 4, and upon the actual performance of the larger ships of the lake fleet in the Great Lakes and connecting waterways. The average round-trip time between Duluth and Buffalo of the 124-mile steamers in the lake service corresponds almost exactly with that resulting from computing the sailing time for the same routes by Table V of Appendix No. 4. A careful study of the performance of the vessels of the lake fleet in the St. Clair Canal indicates that, except at times of heavy winds, it can be safely navigated at a speed of at least 10 miles per hour, and therefore the speed of 7.55 miles per hour adopted for tangents on the proposed 21-foot ship canal, which would have practically the same width and depth as the St. Clair Canal, is certainly safe and practical. '14ie route adopted for these proposed channels through the lakes is practically the same as that now in use, except at the Limekiln Cross¬ ing of the Detroit River, and between Ilay and Mud lakes, on the St. Marys River. The Limekiln Crossing channels, as now constructed, have sharp curves and are dangerous to navigate, and should be straightened and made GOO feet wide. The estimates submitted with this report are DEEP WATERWAYS. 241 based on a channel 600 feet wide, constructed on a continuation across the reef of the tangent east of Bois Blanc Island, as shown on plates 12 and 13. For the improvement of the St. Marys River the estimates are for a channel 600 feet wide through the West Neebish branch of the river, which makes the route 1.2 miles shorter than the present route and can be improved for the required dimensions of channel for less cost than on the route now used. From an engineering point of view the change is desirable on account of the better alignment of channel which can be made and better oppor¬ tunity to execute the work without interference from passing steamers. THE NIAGARA SHIP-CANAL ROUTES. If, taking the cost of construction and maintenance into account, the commerce tributary to the lake system of waterways can be trans¬ ported to the seaboard more economically with channels improved for a minimum depth of 21 feet than if made 30 feet deep, as has been shown to be the case in Appendix No. 5, part 2, the comparison of waterways between the lakes and the Atlantic is reduced to the con¬ sideration of the different 21-foot routes which have been investigated. A careful reconnoissance made by the Board in advance of the field work showed that only two of the routes from Lake Erie to Lake Ontario were worthy of investigation, viz: The route from the Niagara River at Tonawanda to Lake Ontario at Olcott, and from the river at Lasalle to Lewiston and thence through the Niagara River to the lake. These were thoroughly investigated relative to volume and kind of material to be excavated, nature and dimensions of structures which will be needed, and character of foundations on which such structures will have to be erected. The difficulties to be overcome on the two routes are practically the same, and the real comparative merits of the waterways depend largely upon relative cost to construct and maintain them and the difference in time required by a type steamship to traverse the respective routes between points common to each. On the Tonawanda-Olcott route several alternative lines were exam¬ ined to determine the best location for locking down the escarpment near Lockport. The line down the so-called “ Gulf,” which has been followed in previous surveys, was found to involve the construction of a high dam and undesirable alignment of the waterway at locks and approaches, and a location was made on a tangent down the escarpment a little west of the “Gulf,” by which better location and foundations can be had for lock structures at less expense for both excavation and construction. At the Lake Ontario end of the canal an artificial harbor will have to be const ructed by build ing breakwaters outside of the entrance and enlarging the mouth of Eighteen Mile Creek for a distance of about a II Doc. 149-16 242 DEEP WATERWAYS. mile back from the lake, whereas on the Lasalle-Lewiston route the Niagara River constitutes one of the most capacious and safe harbors on the lake waterways. The question has been raised as to the advisability of constructing locks, which will cost several million dollars, as close to the boundary between the United States and Canada as will be the case at the Lew¬ iston escarpment; but when we consider the important lock and regu¬ lating structures which will be needed at the head of Niagara River, the deep channels already excavated in Canadian waters at the mouth of the Detroit River, and the locks and canals at Sault Ste Marie, it is difficult to conceive, if the Lewiston location is objectionable for mili¬ tary reasons, why similar reasons should not have prevented the improvement of the entire upper lake system of waterways. The Lasalle-Lewiston route has fewer important railroad crossings than the Olcott route, and does not interfere with manufacturing and private enterprises to the extent that the latter does in the vicinity of Tonawanda. A steamship of 19 feet draft, capable of steaming 124 statute miles per hour in the open lake, would traverse the Lasalle-Lewiston route from tlie proposed Buffalo regulating works to the junction of the two routes in Lake Ontario in nine hours and fifty-eight minutes, and between the same points on the Tonawanda-Olcott route in eleven hours and six minutes—making a difference of one hour and eight minutes, or over 11 per cent of total time of passage, in favor of the Lasalle-Lewiston route. The total estimated cost of the Tonawanda-Olcott route, including regulating works at the foot of Lake Erie, is $48,533,400, and for the Lasalle-Lewiston route, $42,472,900—making an estimated saving of $6,060,500 if the waterway should be constructed on the latter route. From an engineering and financial point of view, and from the less danger of delays and accidents to navigation in the comparatively short reach of restricted waterway on the Lewiston line, it appears to be the preferable location on which to construct a ship canal. WATERWAYS FROM LAKE ONTARIO TO THE ATLANTIC. The routes via the Oswego, Mohawk, and Hudson rivers and down the St. Lawrence River, Lake Champlain, and Hudson River were the only ones found worthy of complete surveys, plans, and estimates. Consideration was given to lines from Big Sodus Bay and Little Sodus Bay to the Mohawk line near Oneida Lake, but preliminary estimates indicated beyond question that these routes would cost much more to construct and would take much longer fora steamer to traverse than either of the others, and were not considered further by the Board. The conditions existing on the two routes investigated are so radically different that the comparison must depend more upon DEEP WATERWAYS. 248 the conclusions deduced from discussion of relative advantages in Appendix No. 5, Part 2, than upon the natural features of the sec¬ tions through which the lines are located. In the project by the Oswego and Mohawk rivers it is proposed to lock up from the ele¬ vation of 245.4 feet at Oswego Harbor to 379 feet for the low-level plan, and 416 feet for the high-level system on the summit between Oneida Lake and the Mohawk river, in both of which projects the water to generate power for operating the locks and for locking ships across the divide must be secured by storage in reservoirs located on the waterway or on adjacent watersheds. The lockage required to cross the divide with the low-level project will be 267 feet and for the high-level project 341 feet, making the route expensive to construct and slow to navigate. The route down the St. Lawrence through Lake Champlain and the Hudson River is a down-grade canal, having a minimum amount of lockage and an ample supply of water taken directly from the St. Lawrence River, but has the disadvantage of being partly in Canadian territory and about 208 miles longer than the Oswego-Mohawk route. The location and design of the waterway by both routes have been made with reference to interfering the least possible with manufac¬ turing interests at water-power sites. On the Champlain route the changes, except at Fort Miller, will be beneficial to the water-power interests, but on the Oswego and Mohawk a rearrangement of several of the power plants will be nec¬ essary. At Oswego and on the Mohawk route below Schenectady all interference of power rights has been avoided by locating the routes outside of the river valleys. On the Oswego-Mohawk route, the low-summit-level project, utiliz¬ ing Oneida Lake as a principal storage reservoir, will cost about $1,678,100 more than the liigh-summit-level project with water supply obtained from storage reservoirs in the valleys of the Black and Sal¬ mon rivers, but will have the advantage of not having a long feeder line to maintain, and with not interfering with water-power rights outside of the waterway location. The only grave engineering difficulty to be overcome in connection with the low-level plan is the liability of landslides in the deep cut, which in places is through soil which unless thoroughly drained may have a tendency to slide. A system of back drainage to prevent landslides has been provided for in the estimates, and in the opinion of the writer the low-level plan is feasible and is the preferable project to carry out in case the waterway should be constructed. Probably the most serious difficulty to adjust on either route, if the waterway should be constructed, will be to make satisfactory arrangements for railroad crossings. This is especially the case in the 244 DEEP WATERWAYS. Mohawk Valley, where the river is paralleled by the four tracks of t he New York Central and the two tracks of the West Shore railroads. Ample provisions have been made in estimates for either swing or bascule bridges for all crossings, but in the case of the New York Central crossing, the number of trains is so great that a higli grade crossing on a long embankment or an embankment and trestle may be necessary. With a slight readjustment of the railroad lines in the vicinity of Rome only one such high viaduct will be needed. The substitute location of line adopted, from Schenectady to the Hudson River below Albany, involves a deep cutting at South Sche¬ nectady, but will be in stable soil and will be safe construction. A waterway from Schenectady to the Hudson via the Mohawk Valley, would involve sharp curves, difficult and dangerous river crossings, and with damages to water-power privileges would probably cost over $20,000,000 in excess of the line located down the Normans Kill. From the mouth of Normans Kill to deep water in the lower Hudson the Champlain and the Oswego-Mohawk routes have a common location. The following table gives the details of the lengths of the channels of different dimensions, and amount of lockage on the two routes between Lake Erie and New York: BUFFALO TO NEW YORK. Oswego-Mohawk route. Cham¬ plain route. High level. Low level. Total distance.....miles.. Fall,regulated stage of Lake Erie to mean tide.feet.. Down lockage......do... Up lockage. do... Total lockage ..do... Number of locks, including guard locks. 477.04 574.5 476.94 574.5 685.21 574.5 742.6 170.6 705.6 133.6 547.2 0.0 913.2 40 1 839.2 38 1 547.2 21 2 Number of guard locks... Standard canal...miles.. Canalized river: 200 to 250 feet wide.do... 102.56 102.42 102.35 1.51 250 to 300 feet wide.do .. 20.38 12.37 2.59 13.90 8.99 39.15 277.10 20.38 12.37 2. 59 13.90 8.99 37.66 278.64 300 to 350 feet wide..........do... 350 to 41X1 feet wide...do... 38.97 400 to450 feet wide...do... 450 to 500 feet wide.......do... 500 to 1,000 feet wide ...do... Open lake and river...do... Total..... 8.08 11.59 73.65 449.06 477.04 476.94 685.21 It will be noted that of the 574.5 feet fall from Lake Erie to tide water at time of low stage, of rivers, 572 feet are overcome by locks and 2.5 feet by river slopes on the Mohawk route and 547.2 feet by locks and 27.3 feet by river slopes on the Champlain route. At times of high water in the rivers the slopes will be steeper and the total fall at locks less than at low stage. DEEP WATERWAYS. 245 The Champlain route is a down-grade canal from the lakes to the seaboard, and while the large amount of fall in the St. Lawrence River causes rapid currents, the maximum traffic is in the same direc¬ tion, and therefore is subject to no delay from this condition. The estimated cost of these routes from Lake Erie to the Atlantic are as follows: Champlain route .... $183,420, 000 Oswego Mohawk route: High level...._. 197,718.200 Low level...... 199.396, 300 A study of the length of the ice season on the lakes and connecting- channels indicates that open navigation can be maintained on the Champlain route 230 days and via the Mohawk route 245 days for average years. Referring to Table II of Appendix No. 5, Part 2, it will be noted that the round trip from New York to Chicago, including terminal detentions, will take sixteen days and nine hours for the Champlain route and fifteen days and eight hours for the Mohawk route, or, for a full navigation season, a steamship capable of steaming 124 statute miles per hour in the open lake would be able to make fourteen and sixteen round trips in the respective waterways. That is, the ship on the Mohawk waterway would make fifteen round trips in the same time that fourteen trips on the Champlain route could be made, while the greater length of season on the Mohawk route permits still another round trip. A steamship could therefore transport freight from Chicago to New York over the Oswego-Mohawk route for 93 per cent of the rate it would be necessary to charge on the Champlain route to make the same annual profit. Assuming that freight can be transported over the improved water¬ way between New York and Chicago for an average rate of $1 per ton, the saving by using the Mohawk route, with an annual traffic of 20,000,000 tons, would amount to $1,400,000. If the more expensive low-level Mohawk waterway should be con¬ structed, the first cost would be about $15,975,700 in excess of what it would cost to construct the Champlain route. If 3 per cent interest be assumed as the rate paid by the Government for the money used in constructing the waterway, the saving in the annual fixed charges due to the difference in cost of the waterways would be $479,270 and for the difference in cost of operation and maintenance $642,740, making a total of $1,122,010 in favor of the Champlain route, or about 80 per cent of the amount which would be saved on transportation rates by using the Mohawk waterway. For any greater transporta¬ tion rate between Chicago and New York than $1 per ton the compar¬ ative economy of the Mohawk route will be greater and for any lower rate it will be less Steamers operating solely in the lake trade and DEEP WATERWAYS. 246 going- out of commission in the winter season would make two more round trips per year on the Mohawk waterway than could be made by the Champlain route, making the net earning capacity of steamer one-seventh greater for Mohawk route, in which case, if the volume of traffic over the proposed waterway should exceed 8,000,000 tons annually, the saving on transportation rates by using the Mohawk route would exceed the difference in fixed charges due to the less cost for construction and maintenance of the Champlain route. The estimated cost of these routes has been based upon the same standard cross-section and unit prices for similar conditions, the speeds which can be maintained in the waterways have been deducted in a similar way for each route, and the rates of transportation between terminals determined from the sailing time of a type steamer as deduced from the actual performance of similar ships in the lakes and connecting channels. It is therefore evident that errors which may have been introduced by erroneous assumptions will be common to both routes and will not materially affect the conclusions deduced from the comparisons. A large portion of the commerce which will be transported in these waterways, if constructed, will be for domestic consumption, and must necessarilv be delivered at the home market where needed, and as a result the volume of export commerce with lake ports will follow the same transportation lines as used for domestic trade. The choice of routes should therefore be made with reference to reaching the New York market and the ports on the Atlantic coast accessible by the steamers which will be constructed for the lake and seacoast trade. M any of the South Atlantic ports will not admit ships of over 20 feet draft, and therefore the expenses for transfer of domestic freight, if carried in deep-draft ships, would probably be fully as large as for transfer of export traffic through the canal if transported in vessels adapted to 21-foot channels through the connecting waterways of the Great Lakes. From a careful consideration of the type ship best adapted for car¬ rying the lake commerce, and of the depth of channels and waterways of the lakes and from the lakes to the Atlantic required for such ships, it is evident that the proposed 21-foot waterway will furnish better returns for the transportation of domestic and foreign com¬ merce than can be obtained by constructing a waterway 30 feet deep from the lake ports to the seaboard. A careful comparison of the engineering difficulties to be overcome in constructing a waterway from the lakes to the Atlantic by the Oswego-Mohawk route and by the Champlain route, erf the cost to construct and operate them, and of the time required for steamers adapted to the lake and coast trade to make round trips between ter¬ minals, strongly indicates that the former is the preferable route to adopt if the waterway is to be built. DEEP WATERWAYS. 247 The probability of extensive shipbuilding industries being devel¬ oped on the Great Lakes by the construction of a deep waterway to the Atlantic may make it desirable to adopt locks 80 feet wide for pass¬ ing war ships instead of 00 feet wide, as estimated in this report. Such construction would add 84,221,000 to the estimated cost of the Mohawk route and 82,560,000 to that of the Champlain route. The annual capacity of each route would be diminished by the addi¬ tional time required for filling and emptying locks and the time for round trips correspondingly increased, but as these changes would apply to both routes the relative comparison of the two waterways would not be materially changed. The changes would, however, make the comparison between the 21 and 30 foot waterways less favorable to the former. Respectfully submitted. Geo. Y. Wisner. The Board of Engineers on Deep Waterways. Appendix No. 5.—Part 2. RELATIVE ADVANTAGES OF THE 21 AND 30 FOOT WATERWAYS. The sundry civil act of July 1, 1898, definitely fixes the depths of the waterways to be investigated by the Board at 21 and 30 feet respectively, and requires a statement of the relative advantages thereof. The following investigation has been made to determine, as far as may be found possible, the relative advantages of the two waterways. These advantages may be either direct or indirect. The direct advantages consist of the relative returns in value received for the expenditures incurred in the construction, maintenance, and oper¬ ation of the two waterways respectively. The indirect advantages arise from the influence of the waterways upon the commerce of the country. RELATIVE DIRECT ADVANTAGES. In order to investigate the relative direct advantages of the water¬ ways, it is desirable, for the sake of clearness, to state in a condensed form certain general considerations which relate to all lines of trans¬ portation. When an article of merchandise is transported from one point to another, its least value at the point of delivery must be the sum of its value at the shipping point plus the cost of transportation; otherwise the transport would not generally be effected. If v represents the difference in the value of the freight unit at the terminals, Q the total number of freight units transported in a given period (say, one year), and K the total cost of transportation in the same period, 248 DEEP WATERWAYS. U = Qv — K may be taken as an approximate measure of the direct utility of the transportation line. The cost of transportation (Iv) is composed of two parts, (1) the toll or charge for the use of the line, and (2) the transport proper or charge for carrying the freight over the line. It is necessary to dis¬ tinguish between these two charges since they are often paid to differ¬ ent parties. The toll consists of the annual interest on the cost of constructing the line and the annual cost of its maintenance, operation, and super¬ intendence. This charge is paid by the users to the proprietor of the line. The transport proper consists of the annual interest on the cost of carriages, the annual cost of their maintenance, depreciation, and oper¬ ation, and such terminal and other charges as may attach to the traffic. These charges are paid by the users to the carrier. If T represents the toll and P represents the transport proper, the equation of utility becomes TJ = Qr — T — P.(1) In this equation U represents the amount of benefit which should be annually divided between the parties interested in the traffic after the payment of toll and transportation charges. It is, however, in some cases only an approximate measure of utility, for it is based on the single consideration of the difference between terminal value and cost of transportation. There are, however, other important elements involved in the determination of the measure of utility. The speed and regularity with which the transport is effected are often impor¬ tant elements of value, and the capacity of the line for transportation may limit the volume and cost of the traffic. If I w represents the annual interest on the cost of constructing the line; M w the annual cost of its maintenance and operation; I m the annual interest on the cost of carriages, and M m the annual cost of maintenance, depreciation, and operation of carriages, including all charges connected with moving the traffic, we have— T = I w -f M w and— T = Im 4- M m and— U = Qv — I w — M w — I m — M m .(2) It will be observed that, in the general case, there are three parties interested in the transport movement—the carrier, the proprietor of the line, and the shipper and receiver using the line. When these parties are independent their interests conflict and each will endeavor to get as large a part of Qv as possible. The carrier must receive M m , otherwise the freight will not be moved. He must also receive a fair value of I m , for although he DEEP WATERWAYS. 249 might move the freight for a while without receiving any interest on the cost of his carriages, he would refuse to invest money in new carriages to replace those which were worn out and the traffic would soon cease. The proprietor must receive M w , otherwise he will cease maintain¬ ing and operating the line. He need not, however, receive a fair value of I w , for having invested liis money in the line, he can not remove it, and it will pay him to maintain and operate the line as long as he receives M w and a very small value for I w . Indeed, he may continue the operation of the line without receiving any value for I w , in the hope of future gains under more favorable traffic conditions. The users or general public will employ the line if they receive a very small value for U, so that it is to the interest of the carrier and proprietor to make IT as small as possible by making the freight charges as heavy as the traffic will bear. But the total amount received depends upon the volume of traffic Q, and this may often be increased by diminishing the transportation charges. It is, therefore, not always to the advantage of the proprietor and carrier to impose high charges upon the traffic. This distinction between toll and transport proper exists in all cases, although it is not easy to separate the two charges where the proprietor and carrier are one and the same party, as in the case of a railroad. In such a case the toll is simply the difference between the total cost of transportation and tin* cost of moving the freight over the line. When the three parties interested in the traffic movement are inde¬ pendent of each other, it is evident that after each has received a rea¬ sonable portion of the proceeds derived from the traffic, the remaining part of U should be divided between them. The proprietor is gener¬ ally a company, the shares of which are held by many persons, and the carrier often represents a large number of owners. These parties are as much entitled as the users to a share in the extra earnings of the line. In dividing these extra earnings the risk of loss assumed by each party should be taken into consideration. Thus, as has been before remarked, the proprietor can not withdraw his money from the line even if the traffic fails to pay him any returns for his investment. He should therefore receive the largest proportional share of the extra earnings. The carrier should receive the next largest proportional share, for he risks the interest upon the value of his plant and the cost of operating it, which may turn out to be greater than the amount received for transportation. So long as the charge for transport proper does not fluctuate, the user runs no risk from the use of the line and therefore should receive the smallest proportional part of the extra earnings. When the Government is the proprietor of the line, the user (or general public), the proprietor, and the carrier become one and the 250 DEEP WATERWAYS. same party so far as the cost <>f construction, operation, and main¬ tenance of the line is concerned, and if the charges for transport proper can he restricted within reasonable limits the toll may be abolished as a direct charge upon the traffic, since it will be recovered by the people in the increased value of U. Moreover, as will be ex¬ plained hereafter, the indirect benefits expected from the establish¬ ment of the line may fully justify the assumption of the toll by the Government. In the case of a line where the •carrying business is a monopoly (such as a railroad), the tendency would be for the carrier to take as much as possible of the benefit resulting from the abolition of the toll. In the case of a line open to the competition of all car¬ riers (like a great national waterway), the law that where there is free competition the charges for transportation must closely approx¬ imate the net cost will operate. This is one reason why it may some¬ times be an economical advantage for the Government to assume the cost of construction, maintenance, and operation of a waterway, while it would not generally be an advantage to the public in the case of a railroad unless the Government also conducted the business of carrier. The special problem which we are required to consider consists in the determination of the relative advantages of two lines of water transportation of different depths extending from Lake Superior and Lake Michigan to the Atlantic tide waters. The line adopted will compete with the railroads for freight transportation over a large part of the distance. Where the cost of transportation is relatively large, in comparison with the value of the transported commodity at the start¬ ing point, speed will generally be of less value than cost of transport, and the traffic will generally be by the cheaper line without regard to time. If the Government assumes the toll, as is proposed in the cases considered, the cost of transportation will be much less over the waterway than over the railroad, but the railroad will generally have the advantage of speed. Hence we may make the following assump¬ tions : 1. The toll on the waterway is to be assumed by the Government. i’. The traffic will consist of the movement of bulky freight, the cost of the transportation of which is relatively large in comparison with its value at the starting point. In the comparisons to be made it will be unnecessary to consider speed as a direct element of value. Indi¬ rectly, however, it is an element of importance in the determination of the unit cost of transportation. For this case equation (1) becomes U = Qi>—P ........ (3) If U , Q , and P represent the values of U, Q, and P for a second waterway, we have U' = Q'c — P' DEEP WATERWAYS. 251 If p and p' represent the average cost of moving the freight unit over the two lines respectively, we have and and P = Q p and P' = Q 'p' U = Q (v-p), TT' = Q' ( v-p') IT and U' are the annual values returned to the people in compen¬ sation for the assumption of the toll by the Government. If C and C' represent the cost of construction of the two waterways respectively and R and R' the annual returns upon $100 expended in construc¬ tion, after the payment of the costs of maintenance, operation, and transport proper, in the two cases compared, we obtain and U = CR + M v 100 rj'T?' U' = P-^ + M\ 100 The values of R and R' are measures of the relative direct benefits derived by the public in the two cases. Substituting these values in equation 5, we obtain R + 100 Q' C r Q ( v ~P \v-p ) 51 100 TV 51 v ( 6 )* By substituting the values of IT and P in equation 3 we obtain R 100 c [Q ( v — p) — 5I W ] R' If R y =—^ , equation G may be written T3 QYv — p'\ fC 100 5I W \ 100 51 w ' Q \v -p J VC' + C' R J r R (7) THE WATERWAYS. The waterways to be compared will be about equally well adapted to navigation by small vessels engaged in way traffic. The determi- *The value of R may also be placed under the followin _Q' U n, A , 100 Q R = C' form: Q (p-p ) This form exhibits very clearly the influence of the various elements of the problem upon the value of R'. Thus the first term of the second member depends principally upon the traffic capacities of the waterways per unit cost of construc¬ tion, the second term upon costs of maintenance, and the third term upon costs of transport proper. It is not so well adapted to numerical discussion as equation 8, since it requires the assumption of a separate value of b> in the third term for each waterway. 252 DEEP WATERWAYS. nation of relative direct benefit will therefore be limited to the con¬ sideration of through traffic conducted in the most economical carriers. The waterways have two main branches of through traffic extend¬ ing from the head of Lake Superior and the head of Lake Michigan to New York. These will be considered separately and their terminals will be assumed to be at Duluth and New York and at Chicago and New York, respectively. The ocean line may be left out of consider¬ ation in the comparison, since the same class of carriers is assumed to affect the transport over it in all cases. Part of the traffic will be to domestic markets, some of which will be at the coast ports, and part of it will be to foreign markets. For each line the comparison will therefore be made for two separate cases, viz, (1) for domestic mar¬ kets and (2) for foreign markets. The investigations of the Board show that for both waterways the most favorable route from Lake Erie to Lake Ontario is via Lasalle, Lewiston, and the Niagara River. This route is therefore adopted as part of every line investigated. The routes from Duluth and Chicago to New York, which will be compared, are as follows: 1. Thirty-foot waterway via Lasalle, Lewiston, St. Lawrence River, and Lake Champlain. 2. Tliirty-foot waterway via Lasalle, Lewiston, and the Mohawk Valley, high-level plan. 3. Thirty-foot waterway via Lasalle, Lewiston, and the Mohawk Valley, low-level plan. 4. Twenty-one foot waterway via Lasalle, Lewiston, St. Lawrence River, and Lake Champlain. o. Twenty-one foot waterway via Lasalle, Lewiston, and the Mohawk Valley, high-level plan. <>. Twenty-one foot waterway via Lasalle, Lewiston, and the Mohawk Valley, low-level plan. THE TYPE CARRIERS. The characteristic feature of the waterway having a depth of 21 feet is that it can be navigated by lake vessels, so that freight does not have to be transferred to other carriers in passing from lake to canal or from canal to lake. The depth, however, will not be sufficient for navigation by vessels of the most economical type to cross the ocean, and it is therefore assumed that freight destined for ports beyond sea will have to be transferred to other carriers at the seaboard. For reasons which will be given elsewhere, it is assumed that the type car¬ rier adopted for this waterway will be adapted not only to lake and canal traffic, but also to economical navigation along the coasts of North and South America. r I he characteristic feature of the waterway having a depth of 30 feet is that it can be navigated by both lake and ocean-crossing vessels, and so transference of freight will not be required. The traffic on either waterway will consist principally of the move- DEEP WATERWAYS. 253 ment of bulky freight, such as grain, coal, lumber, and ores, to domes¬ tic and coast markets or to markets beyond the sea. Probably this freight will be carried for a time in a great variety of vessels, but there will be a gradual tendency to develop a type of vessel adapted to the most economical service in connection with the waterway con¬ sidered. It is, of course, impossible to predict with certainty what these types will be for the waterways in question, but for the purposes of this investigation type carriers have been selected which, it is believed, will transport freight at the lowest cost. The following table gives data for various carriers and the cost of transportation per ton-mile in open water computed therefrom. This cost is determined by dividing the daily cost of the vessel, including interest, depreciation, repairs, and insurance, by the product of the number of tons in the full cargo and the number of miles traveled in twenty-four hours in open sea. It takes no account of detentions, profits, shore expenses, or final absorption of the capital invested, and therefore should not be confounded with the actual cost of transpor¬ tation. It is employed only to determine the relative economy of transportation of the different vessels in open water. The table includes vessels of 10, 23, and 27 feet draft, lengths from 480 to 550 feet, breadths from 52 to 60 feet, and speeds of 12-^ and 15 statute miles per hour in open water. These data have been prepared for the Board by Mr. Frank E. Kirby, the eminent marine engineer, and therefore may be accepted with great confidence. The carriers are modern steel vessels with water ballast when light. Nos. 1 to 8 are single-screw steamers, and Nos. 9 to 14 are twin screws. Number of carrier. . 1 3 5 4 9 11 13 Length overall. -...feet.. 180 480 480 500 520 540 550 Breadth..do _ 52 52 52 54 56 58 60 Draft..do_ 19 23 27 27 27 27 27 Speed, statute miles per hour. 121 12) 121 121 124 124 121 Indicated horsepower.. 2,200 2,480 2,800 2,930 3,100 3,200 3,330 Coal consumed per hour.. .pounds.. 3,850 4,340 4,900 5,120 5,450 5,600 5,830 Carrying capacity......net tons.. 8,600 9,600 11,760 12,600 13,300 13,980 14,100 Cost of ship for. lake business ouly, dollars Cost of ship for ocean and lake business, 360,000 420,000 504,000 558,000 657,600 710,050 750,000 dollars . .... Pay roll per day, including subsistence, 387,000 462,000 554,400 612,400 705,600 771,400 828,400 dollars. Percentage of cost for repairs and depre- 60 60 62 62 80 80 82 ciation .. ... 5 5 5 5 5 5 5 Percentage of cost for insurance... Incidental expenses per day (coal, waste. 41 41 44 41 41 44 44 etc.).....dollars.. Cost of transport in open water per con- 117 129 143 149 155 159 165 mile (ocean and lake carrier)_mills.. 0.128 0.129 0.121 0.120 0.129 0.130 0.136 Number of carrier... .. 2 4 6 8 10 TT 14 Speed, statute miles per hour ... 15 15 15 15 15 15 15 Indicated horsepower.... 3,650 4,130 4,700 4,850 5,200 5,300 5,500 Coal consumed per hour.pounds.. 6.400 7,230 8,230 8,500 9,100 9,300 9,700 Carrying capacity ... .. _net tons.. Cost of ship for ocean and lake business, 7,650 8,800 9,600 10,000 10,800 11,600 12,000 dollars.... .. .. Pay roll per day, including subsistence. 410,000 507.200 595,000 661,200 760,000 836,000 885,000 dollars..... . Percentage of cost for repairs and depre- 72 74 76 76 96 98 102 ciation....... .... 5 5 5 5 5 5 5 Percentage of cost for insurance... Incidental expenses per day (coal, waste, 44 41 41 41 44 44 4i etc.)...dollars.. Cost of transport in open water per ton- 154 174 198 204 218 224 232 mile.mills.. 0.141 0.142 0.148 0.151 0.158 0.157 0.159 254 DEEP WATERWAYS. The carriers which can navigate the 21-foot waterway are Nos. 1 and 2. They have a draft of 10 feet, a breadth of 52 feet, and a length of 480 feet. These dimensions are adopted for the 21-foot, waterway in this investigation because they are believed to conform to the best present lake practice. The vessels are not quite as long as the long¬ est vessels employed with the same draft, but they have nearly the greatest beam yet adopted. The cost of transport per ton-mile in open water is much greater for carrier No. 2, which has a speed of 15 statute miles per hour, than for carrier No. 1, which has a speed of 124 miles per hour, and the difference will be increased in a restricted channel. No. 1 is therefore adopted as the type carrier for the 21-foot waterway. All the carriers can navigate the 30-foot waterway. It appears from the costs per ton-mile given in the table that the carriers having a speed of 124 miles per hour are much more economical than those having a speed of 15 miles per hour. Of these carriers, No. 7 is the most economical in open water, and also the most economical in a restricted channel as compared with the other carriers having a draft of 27 feet. The diminution of speed in a restricted channel varies considerably for vessels of different draft. To determine whether No. 7 is more economical than No. 1 for navigation in the 30-foot waterway, the loss of speed due to channel restriction and the result¬ ing costs of transport per ton-mile have been calculated for the two cases. For carrier No. 1 the cost of transport per ton-mile in the 30-foot waterway is 0.308 mill, and for carrier No. 7, 0.300 mill. Car¬ rier No. 7 is the more economical under the assumed conditions, and is therefore adopted as the type carrier for the 30-foot waterway. THE DATA FOR COMPARISON. In order to apply equation 8 to the determination of the relative direct benefits to be derived from the waterways under consideration, values for the constants entering the equation must be computed or assumed. The methods employed in determining these values will now lie described. The 30-foot waterways will first lie compared with each other, water¬ way No. 3 being taken as the standard. The 21-foot waterways will then be compared with each other, waterway No. G being taken as the standard. The best 30-foot waterway will then be compared with a 21-foot waterway, the latter being taken as the standard. In equation 8, Q, M w , R, and p relate to the standard waterway and Q', C', M' w , and_p' to the waterway compared therewith. In the tebles the primes are omitted, except where the data refer to both waterways. For convenience the notation employed is here repeated: C =cost of the construction of the waterway. M w -= annual cost of the maintenance and operation of the waterway. Q —number of net tons of freight annually transported over the waterway. DEEP WATERWAYS. 255 p = average cost of moving one ton from one terminal to the other. v = average difference in value of freight unit at terminals. R = annual return upon $100 expended in construction of standard waterway. R' = annual return upon $100 expended in construction of waterway compared. 1. Constants of the 'waterway .—The cost of constructing each water¬ way (C) is obtained from the detailed estimates made by the Board. In connection with the 30-foot waterways it will be necessary to increase the depth of the harbors at Duluth and Chicago to 30 feet to afford necessary terminal facilities. The cost of deepening Duluth Harbor to 30 feet, not including maintenance, is estimated by Maj. C. B. Sears, Corps of Engineers, at $4,607,500, and the cost of establishing the same depth in Chicago Harbor is estimated by Maj. J. H. Willard, Corps of Engineers, at $5,000,000. For the 30-foot waterways these items are added to the estimates of the Board. Major Sears estimates the cost of new piers and the protection of old ones at Duluth at $1,000,000. This has not been included in the estimate for harbor improvement, as it is assumed that the work would not be done at the expense of the Government. The estimate for Chicago Harbor does not include the cost of acquiring property nor the protection of existing structures, which. Major Willard says would be enormous. The annual cost of maintenance and operation of each waterway (M w ) has been determined by the Board after a thorough study of the conditions in each case and the results of experience on existing water¬ ways, the estimates being based upon the following assumptions: For annual repair and maintenance of all structures, such as locks, dams, and bridges, 1 per cent upon the first cost. For annual repair and maintenance of the canal prism, one-half of 1 per cent upon the first cost. For annual operation of single lift, single lock, $24,740. For annual operation of double locks, $19,358 to $38,645 per lift, depending upon the number of lifts combined. These estimates include the estimated cost of the general supervi¬ sion of the whole line. 2. Constants of traffic volume .—The values of Q and Q' can not be estimated with certainty, but for the purpose of comparing waterways of the same depth and navigated by the same type carrier, it may be assumed that these quantities are proportional to the length of the average season of navigation. In determining the round-trip load, it is assumed that the vessel will carry its full load in its eastward trip and only one-third of its load on its westward trip. As will be shown hereafter, the volume of DEEP WATERWAYS. 256 the westward traffic is about one-third that of the eastward traffic at the present time. The number of days in the average season of navigation has been determined by the Board from the records of the St. Marys Falls and St. Lawrence canals for recent years. 3. Constants of transport .—The average cost of moving the freight unit from one terminal to the other ( p ) is computed by dividing the cost of maintaining and operating the type carrier (including interest on first cost) during the round trip by the number of tons in the round-trip load. The cost of maintenance and operation during the round trip is determined by multiplying the daily cost of carrier by the number of days in the round trip and subtracting from the result the estimated value of the coal saved during detention. On each day of detention it is assumed that 20 tons of coal are consumed, which is valued at 12 per ton. The value of p thus determined is not the freight rate. It does not include the cost of loading and unloading at the terminals, the insur¬ ance of the freight during transit, nor the shore expenses. These elements are omitted because they are assumed to be equal increments of r, p, and p ', and, therefore, do not affect the value of the coeffi¬ cient of transport , v-P' v-p ■ In the case of carrier No. 1, however, 25 cents is added to the value of p for foreign traffic, this being the esti¬ mated average cost of transferring the freight unit to a deep-sea carrier. To determine the value of the coefficient of transport v-P' v—p it is nec¬ essary to assume a value for v. The minimum value of v is the actual price paid for moving the freight unit from one terminal to the other, plus the cost of t ransferring it to a deep-sea carrier. In 1898 the freight rate on a ton of wheat from Chicago to New York (by lake and canal, including transfer) was $1.61, and from Duluth to New York, $1.77. To these must be added $0.25 for transfer to the deep-sea carrier. The minimum value of v for routes from Chicago to New York is there¬ fore assumed to be $1.80, and for routes from Duluth to New York at $2.02. The maximum value of v can not be determined; but as v increases the coefficient of transport rapidly approaches a limit at which it becomes sensibly equal to unity, and the terms in equation 8 containing R become insignificant. The values of R, corresponding to extreme values of v, may therefore be obtained from equation 8, with the minimum values of v above given and from the equa- Q' C tionR ; = -Q . yr>. In the comparisons given hereafter the minimum values of v are employed, because the values of p and p' are minimum values, and because it is desirable to give the cost of transport its full I DEEP WATERWAYS. 257 effect in the formula. The possible variation in the value of R„ due to a change in the assumed value of v, may be readily computed from the data given in the tables. It will be found that the possible change in R, is in every case very small, being generally in the third decimal place. It is evident that the coefficient of transport r-p V ~P measures the relative beuefit derived from the line per unit of freight transported, leaving out of consideration costs of construction and maintenance. 4- Constant of return for the standard icaterway .—Changes in the value of R will produce little effect upon the value of R,, determined from equation 8. It will be seen from equation 7 that R depends upon the annual volume of traffic actually moved over the standard water¬ way. Assuming Q =25,000,000 tons (which is the value assumed by the Board for the discussion of technical questions affecting the water¬ way), values of R for foreign and domestic traffic may be computed from equation 7. Equation 8 is applicable to any part of the traffic (foreign or domes¬ tic) if we assume that it is in the same relative proportion to the total traffic in the cases compared. The values of the constants as computed by the methods above described, and the principal data upon which they are based are given in the following tables: H. Doc. 149-17 Table I.— Thirty-foot waterways. [Type vessel No. T. ] DEEP WATERWAYS 258 Table II.— Twenty-one-foot roaterways. DEEP WATERWAYS. 259 3 O OP V ’5 s=j •-H © c a 8“§ ce . &Crv S O 1; p. A ,fi O w r-> ’$ ® 6B z g ^ S ’ 313 1 t " Ij ?u® c ij«iioooaC as — o o fl>5 i.--- a* 2 u ^ ^ ^ I> - g5 &CtU0 O 'd 3 S) ’© o ^ d S sp g g o o r » 3“!C? olb il° « K 260 DEEP WATERWAYS. COMPARISON OF THE WATERWAYS. 1. Thirty-foot waterways .—The values of R, obtained from equation 8, using the data from Table I, are given in the following table, water¬ way No. 3 being the standard for comparison: Route. New York to— Domestic. Foreign. R/. Mean. R,. Mean. r Duluth.. 0.926 0.927 1.009 1.010 }• 0.927 } 1.010 }■ 1.000 0.926 0.925 1.010 1.009 — - }■ 0.926 | 1.010 1.000 (< 'hieago.. .. (Duluth ... Mohawk high level-- Mohawk low level (standard) .. (Chicago... (Duluth.. If the assumptions upon which this investigation is based are accepted, the above values show that the return of direct benefit from the 30-foot waterway via the St. Lawrence River and Lake Champlain (No. 1) is less for both foreign and domestic traffic than the return from either of the Mohawk Valley routes. The two Mohawk Valley waterways give practically the same returns. The difference between the Champlain and Mohawk Valley routes seems too great to be acci¬ dental. It corresponds to a change of about 11 per cent in the relative cost of construction. 2. Twenty-one-foot waterways .—The values of R, obtained from equation 8, using the data from Table II, are given in the following table, waterway No. 6 being the standard for comparison: Domestic. Route. New York to— R,. Mean. Champlain__ (Duluth . 1.013 1.013 } 1.013 (Chicago ____ Mohawk high level. /Duluth. (Chicago . 1.005 1.004 } 1.005 Mohawk low level (standard)... (Duluth._. } 1.000 /Chicago __ No. 4 5 6 Foreign. R. 1.012 1.012 1.004 1.003 Mean. \ 1.012 1.003 1.000 The above values show that the return of direct benefit from the 21-foot waterway via the St. Lawrence River and Lake Champlain (No. 4) is theoretically a little greater than the return from either of the Mohawk Valley routes. The Mohawk Valley routes give nearly equal returns. Practically, however, the three routes give the same return of direct benefit, the difference between the Champlain and Mohawk routes corresponding to a change of about 1 per cent in the relative cost of construction. 3. Twenty-one and 30 foot waterways .—The value of R, as deter- Q' mined from equation 8 depends largely upon the value of v , that is upon the relative volumes of traffic upon the waterways compared. In the preceding comparisons there was no difficulty in making a DEEP WATERWAYS. 261 reasonable est imate of this quantity, since the waterways considered were of the same dimensions and the traffic was assumed to be con¬ ducted in the same type carriers. For the purpose of comparing a 30-foot waterway navigated by carrier No. 7 with a 21-foot waterway navigated by carrier No. 1 it is necessary to determine a reasonable value for this quantity. In the paper on Locks (Appendix No. 1) it is shown that the practical maximum annual traffic capacity of the waterway is limited by the rate at which vessels can passthrough the locks. With a single lock of 20-foot lift, the maximum annual traffic capacity of the 21-foot waterway (Q) will be 25,150,000 net tons, and the maximum annual traffic capacity of the 30-foot waterway (Q') will be 26,359,000 net Q' tons. For this case we have ^ =1.048. Assuming waterway No. 6 as the standard and comparing there¬ with waterway No. 2 (which is the best 30-foot waterway), we obtain the values of R ( given in the following table: Waterway. Roxite. New York to— Domestic. Foreign. R,. Mean R,. Mean. No. 2 (30-foot)_ No. 6 (31-foot)_ Mohawk high level... /Mohawk low level (stand- ( ard). (Duluth. (Chi lago . -. (Duluth. (Chicago. 0.069 0.687 J 0.078 }- 1.000 0.839 0.871 | 0.850 1.000 If we suppose the single-lift locks to be duplicated, the traffic capacity of the waterway will be determined by the lockage capacity of the Lewiston flight, and we have Q 1 = 34,405,000, Q = 35,801,000, Q' and q- = 0.961. In this case the traffic capacity of the larger water¬ way is actually less than that of the smaller one. The corresponding values of R / (neglecting changes in relative cost and maintenance due to the additional locks) are given in the following table: Waterway. Route. New York to— Domestic. Foreign. R„ Mean. R,. Mean. No. 2 (30-foot) .... No. 6 (21-foot) .... Mohawk high level. (Mohawk low level (stand - l ard). fDuluth. (Chicago . i Duluth... (Chicago.. 0.613 0.039 { 0.631 j- 1.000 0.757 0.793 } 0. 775 1.000 The above values show that the return of direct benefit from the 21-foot waterway is much greater than the return from the 30-foot waterway. It should be remarked, however, that the lock dimensions adopted by the Board for the 30-foot waterway are too large for the most eco¬ nomical results from a purely freight traffic, these dimensions having been chosen to provide for the passage of ships of war and large ocean DEEP WATERWAYS. 262 vessels built on the lakes. With locks designed for the most econom¬ ical freight traffic only, somewhat better results would be obtained. RELATIVE INDIRECT ADVANTAGES. The preceding discussion has been confined to the investigation of the comparative direct value of the waterways considered simply as instruments of commerce for the economical transportation of bulky low-priced freight. From this restricted point of view it appears that the 21-foot waterway is much superior to the 30-foot waterway. This is the view which would necessarily be taken by a private pro¬ prietor looking only to the direct gains which might be derived from the traffic. But when the Government is the proprietor, the problem can not be limited to such narrow conditions. The indirect benefits derived from the establishment of the line may be of such importance in their influence upon production, commerce, and the general pros¬ perity of the people that the question of a greater or less return of direct value may become comparatively insignificant. It is of vital importance to the private proprietor that he should obtain a reason¬ able money return for his investment; but the Government may often wisely expend large sums for the production of general results, even when no direct return of value can be expected. In forming an estimate of the relative indirect advantages and dis¬ advantages of the waterways, questions must be considered which can not be stated in mathematical formulas. For a clear understanding of ihese questions it will be necessary to consider briefly the amount and character of the existing lake traffic and its past and probable future development, to point out the distinguishing peculiarities of transportation lines of different character and capacity, and to indicate the objects which the proposed waterways are intended to subserve. THE LAKE TRAFFIC. The demand for increased facilities and diminished rates of trans¬ portation from the region of the Great Lakes to the interior of the country and to the sea is based upon facts which are believed to be established by the history of the development of the productive resources of this part of our territory. The commodities forming the bulk of the traffic for which provision is desired are grain (including flour), iron ore, lumber, and coal. The movement of these commodities comprises about 90 per cent of the total freight movement on the lakes. As will be shown hereafter, t he greater part of this traffic goes to the domestic markets of our coun¬ try, but still an important part is destined to foreign markets. The volume of these products has increased rapidly with every increase in the facilities of transportation and with every permanent decrease in transportation rates. It is claimed that further increase in facilities DEEP WATERWAYS. 2G3 and reduction in rates is absolutely necessary if we would hold our place in foreign markets in competition with the products of other countries. For the purposes of this inquiry full and accurate statistics of the lake commerce as it now exists are unnecessary, and indeed would be of little value. The problem involves conditions which will exist after a deep waterway has been established, and the quantitative effects of these conditions can not be determined from any existing data. Only such figures, therefore, are given as relate directly to the questions under consideration. The following table 1 shows for the year 1898 the traffic for each of the four leading commodities referred to above, the eastward traffic (that is, the traffic east of Detroit), and the quantities destined to domestic and foreign markets, respectively: [Quantities in net tons.] Commodity. Total traffic. Eastward traffic. Total. Domestic. Export. Grain (including flour). Iron ores..... 12,03(5,013 13,650,788 4.5+0.000 8,722,667 12.030,013 11.028,321 2.531,180 2,888,829 11,028,321 2,531,180 a 9,147,184 Lu mber.. Coal. Total.. 38,949.4t5S 25,595,514 16,448.330 9,147,184 a Exports from Montreal, Boston, New York, Philadelphia, and Baltimore. To indicate the magnitude of the past development of this com¬ merce it is only necessary to say that flu* total lake traffic for the year 1871 has been estimated at 14,283,000 tons. Since that time trans¬ portation facilities by rail and water have been greatly increased, new locks around the falls of St. Marys River have been constructed, the Welland Canal has been deepened, the lake harbors and channels have been improved, steam vessels have taken the place of sailing vessels, and the population of the country has about doubled. These are the principal causes of this enormous expansion of the volume of traffic. FUTURE DEVELOPMENT OF LAKE TRAFFIC. The population of the country will surely continue to increase rap¬ idly, and this must be accompanied by an increase in the volume of the lake traffic. It must not, however, be inferred that the eastward traffic will develop in direct proportion to the increase in population of the country, for about one-half of our population is situated in the 1 The figures for grain (including flour) are compiled from a report entitled The Grain Trade of the United States, published by the Bureau of Statistics of the Treasury Department, January, 1900. The other figures are based upon data obtained from the admirable tables which accompany the report of the commit¬ tee on canals of New York State, 1899. DEEP WATERWAYS. great Mississippi Basin, where the rate of increase is much greater than in our Eastern territory. The future demands of this part of our country upon the products of the lake region will doubtless reduce the relative amount of Eastern traffic. Nevertheless, it does not seem unreasonable to believe that the ratio of demand to supply will continue to be as great as it is at the present time, even should the facilities for transportation be very largely increased. The assumption that the difference in value of the freight unit at the terminals is constant for tin* t ransportation lines considered, which forms the basis of our equation of utility, seems, therefore, to be justified. It appears from the table given above that only about one-third of the east-bound lake freight is exported to foreign countries, the re¬ mainder being distributed to domestic markets. Practically the entire exports of commodities transported on the lakes and received from the lake region consists of grain and other food products. GRAIN. As regards the future development of the production of grain in the region tributary to the lakes, it is only necessary to point out that the rapid increase of our population will imperatively demand the utilization of all our food-producing areas in the near future for the supply of our own markets. It has been stated by Hon. John Hyde, Chief Statistician of the Agricultural Department, that within the short period of thirty years more than the entire wheat production of the country will be required for consumption by our own people, to tin entire exclusion of our export trade. 1 Even should this view' not be accepted by all, it must be admitted that the ratio of the export trade to the domestic trade in food products must rapidly diminish. IRON ORE. The movement of iron ore, which forms at the present time so large a proportion of the lake traffic, is principally from Lake Superior to Lake Erie ports, from which the ore is sent by rail to the great coal and iron region of which Pittsburg is the center. As the undeveloped resources of the Lake Superior region are enormous, this traffic may increase greatly under the demands resulting from increased popula¬ tion. Should adequate facilities for water transportation lie provided, it is possible that a considerable part of these products may be carried to points within the interior of the State of New' York, wdiere conven¬ ient limestone and the saving in cost of transportation both of the crude material and finished product may compensate for the advan¬ tage of the Pittsburg district in its greater proximity to coke and coal. 2 ‘“America and the wheat problem." Published in The Wheat Problem, by Sir William Crookes, F. R. S. 2 Report of committee on canals of New* York State, 1899, p. 15. DEEP WATERWAYS. 265 None of this ore is exported at the present time, nor is it probable that much of it ever will be except in the form of finished material. LUMBER. Of the four leading commodities considered lumber forms the small¬ est proportion of the lake traffic, and its movement is rapidly dimin¬ ishing. The reasons for this rapid decrease are fully and clearly stated by Prof. George G. Tunell in his able report on lake com¬ merce. 1 It is largely due to the destruction of the forests on the shores of the lakes and on the banks of the tributary streams. Lumber is now principally obtained at points so far in the interior that it is generally cheaper to saw logs at local mills and transport the product by rail than to carry or float them to the water and transship them. More¬ over, there is a strong and increasing competition in northern markets from southern lumber. The exports of lumber from the lake region are now insignificant, and they must cease in the near future, as much more than our entire product will soon be needed for our own people. COAL. The total volume of eastward traffic on the lakes greatly exceeds that of the westward traffic. The lake movement of coal, which is entirely westward, is therefore of great importance, not only because it supplies the necessities of the territory west and north of Lakes Michigan and Superior, but also because it furnishes a return freight for the lake carriers. Professor Tunell states that during 1890 coal constituted about three-fourths of the west-bound traffic through the Detroit River and 80 per cent of the west-bound traffic through the St. Marys Falls Canal. Most of this material is shipped from the ports of Lake Erie to Duluth and Superior, at the head of Lake Superior, and to Chicago and Milwaukee, at the head of Lake Michigan, the shipments to Lake Superior being much greater than those to Lake Michigan, as in the latter case the conditions are more favorable for railway competition. At the present time none of the coal transported on the lakes is sent to markets beyond sea, but if a deep waterway to the seacoast were constructed it would probably become an important factor in our export traffic. SUMMARY. To summarize the above statements, the freight traffic of the Great Lakes, already amounting to at least 40,000,000 tons per year, 2 may be expected to increase greatly and rapidly with increase of popula¬ tion and the extension and cheapening of facilities for transportation, 1 Doc. No. 277. House of Representatives. Fifty-fifth Congress, second session. s The registered tonnage of the lake traffic for 1898, as given in the report of the New York State committee on canals, is 62,028,000. A large percentage of this is the registered tonnage of passenger steamers. DEEP WATERWAYS. 266 but this traffic will tend more and more to domestic markets and less and less to foreign ones. CHARACTERISTICS OF TRANSPORTATION LINES. These conditions appear to fully justify the establishment of new facilities for transportation from the lakes to the sea either by the General Government or by State or private enterprise. At the pres¬ ent time by far the greater part of the traffic between lake and ocean is by railway, only about one twenty-fifth of the volume transported going by canal and river. If a new line for water transportation is to be established it must be done by the General or a State Government, not only on account of the great expenditure involved, but also because such a line is not so desirable for private ownership and operation as a railway, upon which the carrier business can be monopolized by the owner, and therefore it probably would not be constructed by private enterprise. In order that we may clearly understand the consequences involved in the proposed change of the greater part of the traffic from rail to water transportation, we must now briefly point out the prin¬ cipal characteristics of railways and waterways considered as instru¬ ments of commerce for the transportation of freight. It is frequently asserted that water transportation is always much cheaper than transportation by rail, but this statement can not be accepted without qualification. If it is intended to mean that the cost of transport proper is generally less in the case of the waterway than in the case of the railway, the statement is doubtless true; but if the toll is included in the cost of transport for the waterway as well as for the railway, the cost of transportation will often be less for the rail¬ way than for the waterway when the latter is an artificial channel of moderate dimensions. As a line of communication between the same terminals, the rail¬ way is for obvious reasons almost always shorter than the water line- Moreover, it carries passengers, and a considerable part of its freight is of large value in proportion to its bulk. The passengers and high- class freight are made to bear a large proportion of the mean cost of transportation. A distinguished authority 1 on this subject finds from a study of experience on French railways and waterways that between two given points the mean net cost of transportation by rail is gener¬ ally lower than the cost of transportation by water; but it must be re¬ membered that the canals of France are of small dimensions and not well adapted to economical traffic. In short, no general rule on this subject can be laid down. Each case must be separately investigated, and the relative economical advantages of the rail and waterway must lie determined in accordance with the existing special conditions. Even then it is not easy to make a satisfactory comparison, owing to 1 C. Colson. Ingenieur des Ponts et Chaussees, Maitre des Requetes au Conseil d'Etat. Transports et Tarifs. Paris, 1890. DEEP WATERWAYS. 267 characteristic differences in the methods of conducting transportation by the two lines. Generally the railway carries passengers and a great variety of high-class as well as low-class freight, so that it is ex¬ ceedingly difficult to determine the average cost of transportation of any assumed freight unit. One of the most important differences between the railway and the waterway arises from the fact that in the case of the former the pro¬ prietor of the line and depots for receiving and shipping freight and the carrier are one and the same party, while in the case of the latter these interests are generally in different hands. It results from this that railwav service is much more regular and efficient than water service, because it is under a centralized management. The skill and efficiency with which the railway service is managed and improved and the lack of improvement and efficient manage¬ ment in canal transportation have often been pointed out, but it does not seem to have been observed that these differences are largely inherent in the different character of the organizations of the two serv¬ ices. The management of the railway is as much interested in the shipping, receiving, and movement of the traffic as in the toll, while in the case of the waterway each interest is concerned with the others only so far as may appear to be for its own direct benefit. In the case of the waterway, especially when it is of small dimen¬ sions, delays are more liable to occur from accidents and crowding than in the case of the railway. The railway has generally the great advantage of speed, which secures for it all the traffic in which time of transport is an element of importance. Finally, the railway is available for traffic during the whole year, while the waterway must be closed during the season of ice. M. Colson rein irks that experience shows that generally these advantages of the railway cause it to be preferred for the movement of merchandise of moderate value when the rates do not exceed those of water transportation by more than 20 percent. 1 This deduction, however, is doubtless based upon a study of the traffic upon tin* rail¬ ways and small canals of France. The net cost of transportation upon the waterways herein consid¬ ered, for both domestic and foreign traffic, would, of course, be very much smaller than on a railway or combined lake and railway line, even should the toll be included. Moreover, it is important that the facilities provided for increased traffic movement should be fully ade¬ quate to meet all possible future demands, and the waterways have a traffic capacity exceeding that which could be furnished by railways at the same cost. The total freight tonnage of the New York Central and Hudson River Railroad in 1898 was 23,403,439 tons, which is much less than the maximum traffic capacity of the 21-foot waterway. 'Tarifs et Transports, p. 310. DEEP WATERWAYS. 268 All important advantage of the waterway over the railway results from the characteristic feature of its organization which has been already pointed out—that the various interests of line manager, freight shipper, and receiver and carrier are in different and inde¬ pendent hands. The maximum amount of benefit is derived from the traffic by the users of the line (or general public) when the toll and transport proper are made as small as possible. In the case of a rail¬ way, where the entire system is controlled by a single management, the natural effort is to obtain for the proprietor and carrier as much as possible of the value derived from the traffic; in other words, to make the traffic pay what it will bear. In the case of a large water¬ way open to the use of all carriers, the element of free competition regulates the rate of transport proper, and under these circum¬ stances the charge for transportation must tend to approximate the net cost. But it is not merely from the reduction of rates that benefit is derived. One of the most injurious effects of the lack of free compe¬ tition in railway traffic has been the variation of rates through a wide range, resulting from alternate competition and combination of trans¬ portation lines. It has been found difficult, if not impossible, to con¬ trol these variations by law; but the influence of a large waterway, open to the use of all carriers, could not fail to prevent large fluctua¬ tions in railway charges upon bulky freight during the season of its operation. It has already been pointed out that this characteristic feature of waterways is a disadvantage so far as regards regularity of service and efficiency of management, and this is one reason why it may be considered desirable for the Government to own and manage the waterway and assume the toll. Under these circumstances the public will receive all the benefit derived from the traffic after the carrier has been paid his charges, and these charges will be kept from large fluctuation and near the net cost of transport by the action of free competition. It would, at first sight, seem unfair for the Government to assume the toll on one transportation line to enable it to compete to advantage with other lines constructed and operated by its own citizens; but it is claimed that the increased demand for a higher class of freight, created by the business and prosperity which would inevi¬ tably follow the construction of a great waterway, would more than compensate the railways for their loss of the low-class traffic. It would not be to the public interest to have the high-class traffic diverted from the railways to the waterways, but high-class freight is generally package freight not readily handled by mechanical devices, and therefore not likely to go by water. This characteristic feature of water transportation controls not only the movement of the. freight, but also its shipment and delivery. In the case of the railway, stations are established at which freight must DEEP WATERWAYS. 2()9 be handled under the direction of the management. In the case of the waterway, every point upon its banks is a possible station. The result must be an active competition, which must control and cheapen the cost of handling and develop points of shipment and delivery best ' suited to economical receipt and distribution. It is claimed as a great advantage of waterways of sufficient dimen¬ sions for navigation by ships that they permit of the transport of the cargo through to domestic or foreign ports without transfer from one carrier to another, thus saving the time and cost of handling and loss by waste. This is an advantage of the ship canal as compared with the barge canal of moderate dimensions as well as with the railway. It is, however, considered by high authorities very doubtful whether a vessel can be so constructed as to navigate successfully and eco¬ nomically the ocean, the lakes, and the canal. The ocean vessel must be stronger than the lake vessel and more costly in construction, operation, and maintenance, and it must be fitted with expensive appliances which are not required in the lake traffic. In consider¬ ing this question it must be remembered that under existing condi¬ tions the lake vessel is compelled to be idle during about one-third of the year, while if it had free access to the sea and were constructed for foreign or coast navigation it could be earning money all the year round. Mr. Kirby estimates the cost of our type vessel No. 1, when designed for lake and ocean business, at $387,000, and when designed for lake business only, at $360,000. The daily cost of maintenance and opera¬ tion, including 5 per cent on first cost, is, in the first case, $331, and in the second, $404. Such a vessel, when destined for lake and ocean business, could carry a full cargo from Duluth or Chicago to New York, and, owing to the additional buoyancy of sea water, then take on all the coal required for the ocean voyage without overloading. The benefit to commerce which would result from giving access to shipping from the lakes to the sea, thus rescuing the lake fleet from enforced idleness during one-third of the year, would, of course, be enormous if the problem of constructing a vessel economically adapted to both kinds of service can be satisfactorily solved. This is a bene¬ fit which is peculiar to the waterway, and can not be derived from the extension of railway facilities. It is further stated that if adequate water communication with the sea were provided a great industry in the construction of iron and steel ships would be immediately developed on the lakes. This industry is already an important one, no less than 1,258 vessels hav¬ ing been constructed at the lake ports during the last ten years; but as there is no access to the sea for vessels of more than about 13 feet draft, the business is almost exclusively confined to the construction of ships for the lake service. It is claimed that nowhere in tin; world are the conditions for the 270 DEEP WATERWAYS. economical construction of steel vessels more favorable than at some of the lake ports. Cleveland, for example, is the center of a great steel manufacture. It is further from the coke-producing districts than Pittsburg, but this disadvantage is counterbalanced by its advantage of receiving ores by direct and cheap water transportation. The opening of a deep waterway to the sea would enable the ship¬ yards of the lakes to compete with those of the seacoast in the con¬ struction of vessels for the ocean traffic. Finally, the argument has often been advanced that a deep water¬ way connecting the lakes with the sea would be of great military value in connection with the defense of the northern frontier of the country Such a waterway would enable ships of war to pass between the sea and the lakes, and it would also permit the economical con¬ struction of such vessels at the lake shipyards. The preceding brief statement of commercial and transportation conditions and of the benefits which may be expected to result from the establishment of a deep waterway from the lakes to the sea is intended only as a basis for the comparison of the relative advantages and disadvantages of the 21-foot and 30-foot waterways. Under the provisions of law, it is not the duty of the Board to report upon the general question as to whether the requirements of commerce justify the construction of a deep waterway at the expense of the General Government or to compare the advantages of such a waterway with those of one of moderate dimensions requiring transfers of freight at both its terminals. Nevertheless, before making the comparison required by law, it seems desirable to invite attention to two important points. The first point is that if any project is undertaken by the Govern¬ ment it should be fully adequate to the present and future purposes which it is intended to subserve. It has been remarked by a distin¬ guished authority that the life of any public work is practically coincident with that of the generation which began it. This is especially true in a country like our own, where population increases and commerce develops with amazing rapidity. The reason why it is true is because in the construction of such works future necessities are almost invariably underestimated. For example, the first canal and locks at St. Marys Falls were completed in 1855, at a cost of about $1,000,000. To meet the necessities of the increasing traffic, a new and much larger lock and canal were commenced in 1870 and completed in 188', at a cost of $2,171,000. This was soon found to be insufficient for the requirements of the lake navigation, and still another and larger lock was commenced in 1887 and completed in 1890, at a cost of about $4,700,000. The volume of the lake traffic has so greatly increased that at the present time the construction of a new lock is under consideration. In 1870 no one could have been bold enough to suggest the construction of a lock of the size and cost DEEP WATERWAYS. 271 of the Poe lock, recently completed; and yet if such a lock had been then constructed the results would have been a large saving to the Government and a great benefit to the commercial interests of the lakes. It is, therefore, of the highest importance that any waterway con¬ structed by the Government should be fully capable of meeting every possible commercial demand which may arise in the future. A lack of capacity for future commerce might necessitate its entire recon¬ struction at enormous cost, and require an adaptation of vessels and traffic to new conditions, involving great loss to commercial interests. The second point is that any project undertaken by the Government should be of a national and not of a local character, benefiting many and varied commercial interests and exerting its influence over as great an extent of the country as possible. It is easily conceivable that a barge canal of moderate dimensions, requiring transfers at Buffalo and New York, might be of more direct benefit to the State of New York than a canal of sufficient dimensions for the uninterrupted passage of ships, but much of this benefit would be at the expense of the producers and shippers of other parts of the country. Moreover, with such a canal the large interests of shipbuilding and winter traffic for the lake fleet would be unprovided for. It appears from the investigations of the Board and the preceding discussion that the most favorable route for a 30-foot waterway from the lakes to the sea is from Lake Erie to Lake Ontario via La Salle and Lewiston, and from Lake Ontario to tlie Hudson River via Oswego and the Mohawk Yalley, on the low-level plan, and that the same route is practically as favorable as any for the 21-foot waterway. The high-level plan gives a slightly greater return of direct value than the low-level plan for both the 21-foot and 30-foot waterways; but the difference is insignificant, and the Board considers the low-level plan preferable for engineering reasons. This route is entirely in our own country and has a longer season of navigation than the more north¬ erly line. The problem ol its defense is, of course, much simpler than it would be were a part of it in a foreign country, and it is available as a line of communication for ships of war. In the following com¬ parison of the 21-foot and 30-foot waterways this route will alone be considered. Cost of construction. —The estimated cost of the 21-foot waterway is $206,358,000; the estimated cost of the 30-foot waterway is $317,284,000, to which should be added about $9,607,500 for the neces¬ sary deepening of the harbors at Duluth and Chicago, making the total cost $326,892,000. Cost of maintenance and operation .—The annual cost of mainte¬ nance and operation is estimated at $2,286,189 for the 21-foot water¬ way, and $2,883,158 for the 30-foot waterway. 272 DEEP WATERWAYS. Cost of transport proper. —The theoretical cost of moving the freight unit, exclusive of toll, from one terminal to the other on the lines considered, is given in the following table: Route. 21-foot waterway. 30-foot waterway. Domestic. Foreign. Domestic. Foreign. Total cost. Cost per ton-mile. Total cost. Cost per ton-mile. Total cost. Cost per | Total ton-mile. cost. Cost per ton-mile. New York to— Cents. Mills. Cents. Mills. Cents. Mills. Cents. Mills. Duluth. 45.2 0.31 70.2 0.48 45.4 0 31 : 40.9 0.28 Chicago. 42. :i .31 67.3 .49 42.7 .31 1 38.2 .28 Mean. . 310 . 485 . 310 . .280 It must be remembered that these values are purely theoretical, and are not given as the probable freight rates; but they are believed to be proportional to the latter and may, therefore, be taken as relative measures of the cost of transport proper for the waterways compared. The table shows that the cost of transport proper on the 21-foot waterway is the same for domestic traffic as on the 30-foot waterway. For foreign traffic the 30-foot waterway shows a much lower cost of transport than the 21-foot waterway. Traffic capacity. —The maximum annual traffic capacity of the 21- foot waterway (when the single-lift locks are duplicated) is estimated at 36,319,500 net tons, and that of the 30-foot waterway at 34,903,000 net tons, the traffic on the smaller waterway being greater than that on the larger one owing to the difference in time expended in lockage. Speed. —The average speed on the 21-foot, waterway is 10.67 miles per hour. The average speed on the 30-foot waterway is 10 miles per hour. Adaptability to traffic conditions. —Our vessel No. I, which is the type vessel adopted for the 21-foot waterway, has a draft of 19 feet, and can enter all the important lake harbors as well as navigate along the seacoast. It is therefore much better adapted to domestic traffic than vessel No. 7, the type vessel for the 30-foot waterway, since the latter has a draft of 27 feet and can not enter the lake harbors. The smaller vessel is not so well adapted to deep-sea navigation as the larger one. Regularity of service. —In the 30-foot waterway navigation would be freer, and for smaller vessels a little more rapid than in the 21-foot waterway, and there would be less danger of delay from accidents and crowding. The time required for vessel No. 1 to make a single trip from Duluth to New York on the 30-foot waterway is six days and three hours, while the same journey on the 21-foot waterway would require two hours longer. Influence on railway rates. —As both waterways furnish low rates for large traffic volumes, there seems to be little choice between them in this respect. DEEP WATERWAYS. 273 Outlet for the lake fleet .—Even should a 30-foot waterway he estab¬ lished between the lakes and the sea, it is probable that the number of vessels of large draft in the lake service would be comparatively small, since such vessels could not enter most of the lake harbors, and would be adapted only to through and principally foreign traffic. The 21-foct waterway would therefore be practically as good as the 30-foot waterway as a means of access to the sea for the lake fleet. Route for ships of war .—In the very improbable event of a war with Great Britain, every large ship of war possessed by this country would be required on the high sea. Such vessels would be unnecessary on the lakes, since the greatest depth of the Canadian waterways is only 14 feet. For purposes of naval defense, the 21-foot waterway appears to offer ample facilities. Shipbuilding .—The 30-foot waterway would enable the shipbuilders of the lakes to construct seagoing vessels of the largest size both for commercial and naval purposes. With the 21-foot waterway this industry must be restricted to the construction of vessels of not too great dimensions to pass the locks. CONCLUSION. As the result of this investigation, it.appears that the 21-foot water¬ way promises a much greater return of value relatively to its cost than the 30-foot waterway. The main advantages of the 30-foot waterway are that it would furnish the lowest cost of transport proper to for¬ eign markets and permit the construction of the largest seagoing ves¬ sels on the lakes. Respectfully submitted. C. W. Raymond, Lieutenant-Colonel, Corps of Engineers. The Board of Engineers on Deep Waterways. Appendix No. 0. LAKE ERIE REGULATION. The act of Congress providing for a board of engineers on deep waterways requires that surveys and examinations be made on which to mature a project for controlling the level of Lake Erie, as recom¬ mended by the report of the Deep Waterways Commission, transmitted by the President to Congress January 18, 1897. In compliance with this requirement, I have the honor to submit for the consideration of the board the following discussion of lake levels, with plans and esti¬ mates of cost of regulating works designed for the purpose of main¬ taining the level of Lake Erie near its high-water stage during the navigation season. II. Doc. 149- 18 274 DKEP WATERWAYS. The regulation of the level of a lake implies the maintenance of its surface at or near some fixed stage, to accomplish which the discharge must be so controlled that it will be at all times approximately equal to the difference bet ween the supply of water to the lake and the evapo¬ ration from the surface. In the Great Lakes system the watershed is about 2.4 times the area of the lake surfaces, and since the variation in the annual precipitation on the entire basin is approximately 50 per cent of the rainfall for a minimum year, and as the per cent of run-off from the watershed increases rapidly with increase of precipitation, it is probable that the actual supply for years of maximum rainfall is more than double that for years of minimum rainfall. The surfaces of Lakes Michigan and Huron rise and fall at times at the rate of two-thirds of a foot per month, corresponding to a change of reservoir supply of 320,000 cubic feet per second. Assuming the discharge of the St. Clair River to be 190,000 cubic feet per second at the time that these changes occur, 1 it is evident that when the lakes are rising the supply at times exceeds 510,000 cubic feet per second, and when falling the supply becomes 130,000 cubic feet per second less than the evaporation from the lake surface, making the actual supply a negative quantity. It is apparent, therefore, that with the evaporation from the lake surfaces at times largely in excess of total supply—which supply, including evaporation, has a range of over 600,000 cubic feet per second—any attempt to maintain the level of Lakes Huron and Michigan at a fixed stage would necessarily be a failure. The storage capacity of Lake Superior amounts to 28,000 cubic feet per second annually for each foot in depth on the lake surface, and since the time of maximum discharge into Lake Huron is that when a large supply is necessary for the maintenance of the level of Lakes Huron and Michigan, it is apparent that any material modification of the range of water levels of Lake Superior would be an injury to the entire waterway system, and therefore the natural conditions on that lake should be maintained. The storage of water in Lakes Superior, Michigan, and Huron, and the consequent fluctuations of the levels of those lakes, is absolutely essential for the maintenance of a flowthrough the connecting water¬ ways sufficiently uniform for navigation purposes. While absolute regulation of the level of these lakes is an impossibility, a decrease of the fluctuations may be permissible, and will be considered elsewhere in connection with the indirect effect of the regulation of the level of Lake Erie on the depths of connecting waterways. 1 The discharge of the St. Clair River is approximate, but is probably about 190,000 cubic feet per second for mean stages of Lakes Huron and St. Clair. The volume of discharge depends upon the level of both Lake Huron aud Lake St. Clair, and is not constant for a given stage of the former. DEEP WATERWAYS. 275 The large areas of the water surfaces of the upper lakes serve as stor¬ age reservoirs during years of surplus rainfall on the lake basins, from which it is gradually discharged through the outlets during years of less than average precipitation. A variation in level of 1 foot for Lakes Superior, Huron, and Michigan is equivalent to a change in actual supply of 68,300 cubic feet per second for an entire year. The average rainfall on the lake basins tributary to Lake Erie is 31.64 inches per year, and is equivalent to 618,000 cubic feet per second for the same period, of which, approximately, 340,000 cubic feet per second either falls on or flows into the lakes, and 278,000 is either absorbed into the land on which it falls or is evaporated from vegeta¬ tion and surfaces of ponds, streams, and marshes. About 120,000 cubic feet per second of the average annual supply to the lake reser¬ voir system is taken up by evaporation and 220,000 cubic feet per second discharged through the Niagara River. The run-off and evaporation are approximate quantities and are based on the result of observations made where the conditions were somewhat similar to those of the lake basin. So far as this discussion is concerned, a knowledge of the absolute values of these quantities is not necessary, as the relation of net supply and storage to the flow through the connecting waterways is so apparent that the approximate values used are sufficiently accurate to show the limits within which the storage capacity of the reservoir system may be modified without material injury to the navigable channels. The evaporation is to a great extent a function of the temperature of the air and water, and of the force of the wind, and may differ materially from results obtained at experimental observation stations. The Rochester observations indicate that the evaporation for Lake Erie is probably 30 to 36 inches annually, while on Lake Superior the best data available indicates that the evaporation is less than one-half this amount. The amount of water which the ground of any given watershed will absorb is about the same for each year, provided the supply is ade¬ quate, and therefore the run-off from watershed is a function of the difference between the precipitation and the amount absorbed, and not a direct percentage of the average rainfall for a series of years. The annual precipitation on the Lakes Huron-Michigan basin varies from 27 to 40 inches, while the average absorption on the watershed is about 22 inches, from which it will be seen that the run-off during a wet year might be from two to three times that for a dry year, or a volume equivalent to over 3 feet in depth on the entire lake surfaces. This supply will not, however, produce a rise of 3 feet, for the reason that the volume of discharge through the outlets increases rapidly with increase of stage in the lakes. The maximum change of stage for any one year on Lakes Superior, Michigan, Huron, and Erie does not exceed 2.25 feet, while the extreme change during any series of years is about twice that amount. DEEP WATERWAYS. 27 (i This extreme change of level fixes the maximum stage necessary for taking care of tlie flood waters of the lake basins during a series of wet years, and the minimum limit which can be fixed for the fluctua¬ tion of any lake is that which will permit sufficient storage to insure an adequate flow for navigation through the connecting waterway during any series of dry years which are likely to occur. The following table gives the area of the surfaces of the Great Lakes and watersheds, as determined by the best authorities: Authority. Area of watershed. Lake— L. Y. Scher- merhorn. L. E. Cooley. T. Russell. Sq. m iles. 31,200 Sq. m ties. 31,800 S'q. m iles. 32.060 Sq. miles. 48,600 Michigan.... 22, 450 22,400 22. :136 45,7(H) 23,800 23,200 22,978 52,100 495 503 6,320 Erie .. . .. ... 9.960 7,240 10,000 9.968 24.480 Ontario.... 7,450 7,243 25,530 The values given by Mr. Cooley have been used in this discussion with the exception of that for Lake Erie, the area of which has been redetermined in this office and found to be 9,932 square miles. The average annual rainfall on the lake basins, as furnished by the United States Weather Bureau, is as follows: Inches. Lake Superior _....... 28 Lake Michigan ...........33 Lake Huron........32 Lake St. Clair.. ........ .. 35 Lake Erie .. ........... .36 Lake Ontario. ....... 33 Assuming the run-off from the watersheds for a year of average precipitation to be 42 per cent for Lake Superior and 33 per cent for Lakes Michigan, Huron, and Erie, the distribution of the flow through the waterway system will be, approximately, as follows: Lake Superior (area of watershed 1.53 times area of lake surface): Depth on lake. Rainfall on lake surface. .feet.. 2.33 Run-off from watershed..do_1.50 Total supply.......do... 3.83 Evaporation from lake surface..‘...do_1.25 Discharge through St. Marys River (72,000 cubic feet per second) ..do... 2 . 58 Total....do... 3.83 Lakes Michigan and Huron (area of watershed 2.15 times the area of lake surfaces): Depth on lakes. Rainfall on lake surfaces.... ...feet.. 2.7 Run-off from watershed....... do... 1 .9 Inflow from St. Marys River (72,000 cubic feet per second)...do... 1.8 Total supply..... do... 6.4 DEEP WATERWAYS. 277 Lakes Michigan and Huron —Continued. Depth on lakes. Evaporation from lake surfaces ...feet.. 1.7 Discharge through St. Clair River (190,000 cubic feet per second). .do_ 4.7 Total .......do_ 6.4 Lake Erie (area of watershed 2.46 times area of lake surface): Depth on lake. Rainfall on lake surface..-..feet.. 3.0 Run-off from watershed___.___do... 2.5 Inflow from Detroit River (195.000 cubic feet per second).do_22.2 Total supply......do_27.7 Evaporation from lake surface....do_ 2.6 Discharge through Niagara River (220,000 cubic feet per second)..do_25.1 Total....... .. _.do.__ 27.7 The variation in the annual rainfall on the lake system amounts to 12 inches over the entire basin and is equivalent to a depth of 27 feet over the surface of Lake Erie, whereas the maximum storage which has occurred in any one vear on Lake Erie only amounted to 2.16 feet in depth, with an average of 1.17 feet, or less than 5 per cent of the mean discharge of the Niagara River, and only one twenty-third of the total variation of the annual rainfall on the drainage basin tribu¬ tary to the lake. EFFECT ON LEVEL OF LAKE ONTARIO AND ST. LAWRENCE RIVER. To regulate the level of Lake Erie so as to maintain its surface near some fixed plane of reference will require such control of the outflow through Niagara River that the storage which would naturally occur in the lake will be discharged during tin* first half of the year and the outflow be diminished a like amount during the last half of the year. This modification of outflow will not materially change the total vol¬ ume of discharge for any entire year, and will amount to only one-fifth of the variation of discharge of the river for different years under present conditions. The effect of this modification of flowthrough Niagara River on the level of Lake Ontario will be to slightly increase the rate of rise in the spring and make the date of maximum stage a little earlier. No reliable data exist from which the increment of discharge of the St. Lawrence River for 1-foot change of stage can be determined, but since the average annual fluctuation of the lake is 30 percent greater than for Lake Erie, while the area of surface is only 25 per cent less than that of Lake Erie, it is evident the increment for the St. Law¬ rence does not differ much from that of the Niagara, a condition which would have a tendency to make any change of levels due to modified inflow very small. 278 DEEP WATERWAYS. The following table shows the stage and rise of Lake Ontario from January to June, and the inflow from Niagara River each year since I 860: 1 Year. Elevation in— Rise. Average inflow from Niag¬ ara River. January to July, per second. Elevation in— Rise. Average inflow from Niag¬ ara River, January to July, per second. January. June. Year. January. June. Feet. Feet. Feet. Cubic feet. Feet. Feet. Feet. Cubic feet. 1865. 247.17 247. 75 0.58 212.388 1882. 245.82 247.62 1.80 245,758 1866. 5.55 6.01 . 45 212.818 1883 . 5.41 7.58 2.17 227,523 1867. 6.04 8.57 2 53 222,805 1884 _ 6.60 8.18 1.58 243, :?56 1868 . 4. 60 2. 03 208,909 1885 . 6.23 7.53 1.30 223,576 i860 5.31 1.75 212,885 1886 ... 7.69 8.53 .84 239 381 1870_ 7.35 8.72 1.37 237.676 1887 . 6.26 8.25 1.99 245,243 1871. 6.15 7.15 1.00 224,819 1888 _ 5.53 6.37 .84 217,906 1872 4.82 5.38 197,013 1889 5. 71 1.01 214 043 1873. 4.40 7.01 2.61 208.451 1890 _ 6.34 8.25 1.91 232,920 1874. 6.44 7.35 .91 235.603 1891. 6.28 6.92 .64 216,876 1875 4.82 5.96 1. 11 204, 366 1892 . 4.60 5.90 1.30 202,288 1876 5.40 8.39 2.99 247,328 189 i. 4.96 2.50 207,470 1877.... 5. 98 t; 52 .54 223. 787 1894 . 5.65 6.89 1.24 208,594 1878. 5.57 7.06 1.49 237,887 1895 . 4.58 4.97 .39 189,048 1879 6.90 6. 92 .02 220,832 1896 .. 3.89 5.43 1.54 189.141 188(1 5 41 6. 60 1. 19 2/5.757 1897. 3.97 5.71 1.74 203,471 1881_ 4.83 6.30 1.47 216.054 1898 . 4.73 6.22 1.49 211,834 Comparing the water level of Lake Erie for 1876 with that for 1895, it will be seen from table on page 287 that the actual supply to Lake Erie varies by 50 per cent of that for a minimum year, and that the discharge of the Niagara River varies 30 per cent for extreme stages of the lake. If an absolute variation of 30 per cent of the supply to Lake Ontario from the Niagara River does not produce any serious results, a change of 5 per cent of the flow from the first to the last half of the year would certainly have no material effect on depth of waterways. On plate 81 a curve showing the level of Lake Ontario since 1865 has been platted from the monthly mean elevations of the lake, and another curve showing the monthly inflow from the Niagara River, in which the unit of the ordinates represent the number of cubic feet per second necessary to raise the level of the lake 1 foot in one month. Assuming the discharge through the Niagara and St. Lawrence rivers at any given time to be approximately the same, the monthly difference of ordinates of Niagara discharge curve will represent the effect of the change of volume of inflow on the level of the lake, and the difference in the ordinates of the two curves will represent the excess of local supply over evaporation when the lake is rising, and the excess of evaporation over local supply when the lake is falling. It is very evident, both from table and from the curves, that any small modification of the flow of the Niagara which does not materially increase or diminish the total annual discharge through the river can not affect the depths of the Lake Ontario and St. Lawrence River waterways to any material extent. 'All elevations given in this report are referred to mean tide at New York and are based on the elevation of bench mark at Gtreenbush, N. Y., which is 14.73 feet. DEEP WATERWAYS. 279 If it should he found desirable to control the discharge of Ihe St. Lawrence River within such limits as to reduce the fluctuation of 5.5 feet on Lake Ontario, under present conditions, to one-half that amount or less, it can be easily and cheaply accomplished by regu¬ lating works at the head of the Galop Rapids. The St. Lawrence River above the Galop Rapids is of large cross- section and lias nearly the same slope for all stages of the river. At the head of Galop Island the river separates into two main channels, with heavy rapids in each. The channel on the north side of the river can be enlarged at the head of the rapids and provided with regulating works, which would greatly reduce the annual fluctu¬ ation of the lake levels. The physical features and conditions at this locality are discussed in report on the St. Lawrence River surveys. EFFECT ON THE CONNECTING WATERWAYS OF THE UPPER LAKES. The effect on the Lake Ontario and St. Lawrence River waterways which would arise from the control of the outflow of the Niagara River would be practically that due to distributing about 5 per cent of the inflow from the Niagara River over a different portion of the year than under natural conditions, but the effect of Lake Erie regu¬ lation on levels of Lake Michigan, Lake Huron, St. Clair and Detroit rivers will be of an entirely different nature, arising from the fact that the low-water levels will be permanently raised, the river slopes decreased for any given volume of discharge, and a redistribution of the flow due to a greater variation of the St. Clair River slopes than those arising from the natural change of lake levels. The average annual fluctuation of Lake Erie is greater than that of Lake Huron, and consequently tin* slope of the connecting waterway is greater for the low-water stages of the lakes than at high stages, a condition causing a greater low-water discharge and a smaller high- water discharge, through the St. Clair and Detroit rivers, than would be the case with the surface of Lake Erie regulated at some fixed elevation. During the season of navigation the fall of the St. Clair and Detroit rivers has an average annual variation of less than 0.50 foot each,with a total variation between Lake Huron and Lake Erie of about 1 foot, from which it is apparent that, the variation of the outflow from Lake Huron is largely a function of the change in area of cross-section of outlet, due to change of stage. During the winter season the freezing over of the St. Clair River diminishes the flow into Lake St. Clairtosuch an extent that the level of that lake usually falls considerably with reference to both Lake Huron and Lake Erie, making an abnormally steep slope in the St. Clair and a very low slope in the Detroit River. (See plate s:j.) 280 ✓ DEEP WATERWAYS. Previous to 1880 the average fall from Lake Huron to Lake Erie was 9.2 feet, after which date the slope gradually diminished until 1890, since which time it has had an average of 8.8 feet. This decrease of slope was caused by the deepening of the river channels at the St. Clair Flats and at the Limekiln Crossing by the Government, and by the great increase from natural causes in depth and cross-section of the St. Clair River through the rapids at the outlet of Lake Huron. A survey made at the request of this board in December, 1898, by a party under the direction of Lieut. Col. G. J. Lydecker, Corps of Engineers, United States Army, shows that the gorge at the head of the river now lias a central deptli of 66 feet and a cross-section of 86,000 square feet, whereas at the date of previous survey, in 1867, the central depth was only 48 feet and the cross-section 80,000 square feet. While there are no records to show when the deepening of channel through the rapids occurred, a study of the water levels and slope curves shown on plate 83 indicates that the erosion was probably started in the spring of 1886 by the abnormal fall (7.5 feet) of the St. Clair River at that time. The average fall of the surfaces of the rivers previous to 1886 and since 1889 was as follows: St. Clair River. Detroit River and St. Clair Flats Canal. 1873 to 1886..... Feet. 5.7 5.0 Feet. 3.5 3 3 1889 to 1898. Decrease in slope__ . 7 •> It is probable the 0.7-foot decrease of slope in the St. Clair is due to the change of cross-section through the rapids at the head of the river, and that the 0.2-foot change of slope in the Detroit River and St. Clair Flats is due to the increased depth of channels from Government improvements. To determine what the effect of change of stage in Lakes Huron and Ei •ie has upon the slopes of the connecting waterway a line of precise levels was run, under the direction of this board, from Lake Erie to Lake Huron, and connections made with 15 different gauges at critical points on the rivers, which were read simultaneously for a week at two different periods, when the stage of the lakes had a difference of 0.66 foot. The profiles of these slopes are shown on plate 82, which, in connec¬ tion with curve of the monthly mean slopes of the Detroit and St. Clair rivers from 1873 to 1898, shown on plate 83 indicate that if the low-water stage of Lake Erie be raised and maintained 3 feet above its natural elevation the corresponding low-water stage of Lake St. Clair would be raised 2 feet and that of Lake Huron 1 foot, making the resulting DEEP WATERWAYS. 281 low stage for Lakes Huron and Michigan approximately what it was before being lowered by the deepening of the river channels. A rise of 0.66 foot in both Lake Huron and Lake Erie produced 0.56-foot rise in Lake St. Clair, with a largely increased volume of discharge through the waterway. If the change of level in Lake St. Clair had been produced by raising the level of Lake Erie without changing the volume of flow, the sur¬ face of the former would have been raised only 0.45 foot, or a change of 3 feet iu the level of Lake Erie by regulation will raise the level of Lake St. Clair 2 feet. An increase of 3 feet in the low-water level of Lake Erie would wipe out the entire fall now existing in the Detroit River for that stage, and the level of Lake St. Clair would rise until the slope and cross-section of the Detroit River was sufficient to main¬ tain the low-water discharge, which existing data indicate would be about 2 feet. If the low-water stage of Lake St. Clair should be increased 2 feet, the flow of the St. Clair River would be diminished to such an extent that Lake Huron would rise 1 foot before the normal low-water flow would be established. The increase of the cross section of the Det roit River for the regu¬ lated stage over that at its natural low-water stage would be such that two-thirds of the low-water slope would be sufficient to produce approx¬ imately the same volume of flow in each case. The total variation of slope would still be about 1 foot, as under present conditions, and would limit the fluctuation of Lake St. Clair to the same amount. The total variation of the slope of the St. Clair River when not atfected by ice is about 1 foot, and since the cross section of the pro¬ posed regulated waterway would be greater than at the natural low- water stage, it is safe to assume that the slopes necessary to produce the same annual volumes of discharge as in the past will be less, and taken in connection with the fact that Lake St. Clair will have a fluc¬ tuation of about 1 foot, would indicate that the levels of Lakes Michi¬ gan and Huron will seldom have a variation of 2 feet, except in years of excessive rainfall on the lake basin, when the natural high-water stage would likely be reached. But since the deepening of the chan¬ nels of the Detroit and St. Clair rivers has permanently lowered the levels of Lakes Huron and Michigan by about 1 foot for all similar conditions of supply and discharge, the high-water stage of those lakes for the future will be at least a foot lower than for similar seasons previous to 1890. The regulation of Lake Erie will, therefore, raise the present low-water stage of Lakes Michigan and Huron by about 1 foot and diminish the fluctuation of the levels the same amount. The slope of the St. Clair River and the resulting discharge will be less in winter under the proposed conditions than at present. This, however, only applies to the ice period on the St. Clair River. Under present conditions the slope of the waterway from Lake 282 DEEP WATERWAYS. Huron to Lake Erie decreases slightly as the lakes rise but with the level of Lake Erie maintained at a fixed stage the slope would vary with the stage of water in Lake Huron, and consequently the discharge would become a function of the hydraulic head at t he outlet of the lake, a condition requisite for taking care of a maximum variation of supply to the lake reservoirs with a minimum fluctuation of the surfaces. The area of Lake Erie is so small compared with that of Lakes Superior, Huron, and Michigan that its reservoir capacity is only one- ninth that of the upper lake system and if eliminated from the system by regulation the storage capacity remaining would be ample to main¬ tain the connecting waterways so that the low-water stages would always be 3 feet higher in Lake Erie, 2 feet in Lake St. Clair, and 1 foot in Lake Huron than it is under the present conditions. If tin* channel from Lake Huron to Lake Erie should be made 30 feet deep, the cross section of the waterway would be increased 600 square feet at the shallow places for each foot in depth that the channel is deepened at the respective shoals, which would decrease the slope requisite for any given volume of flow and slightly lower tin* low- water plane for both Lake Huron and Lake St. Clair. The portion of the waterway over which this improvement would be distributed is comparatively short, and the total reduction of fall between the two lakes would probably be less than 0.3 foot, but is not susceptible of exact determination. REGULATION OF LAKE ERIE. To determine the limits within which the level of Lake Erie is sus¬ ceptible of being controlled by properly const ructed regulating works, the Board has caused careful surveys and examinations to be made to determine the topography, hydrography, and character of material on which structures would have to be founded at foot of the lake and in the head of Niagara River, and a long series of observations to establish law of flow for the discharge through Niagara River for dif¬ ferent stages of the lake. A series of observations has also been made for the board at the hydraulic laboratory of Cornell University, by Prof. Gardner 8. Williams, to determine the coefficients of the formula for discharge over submerged weirs when the depth of water on crest of weir was greater than that for which the coefficients have heretofore been determined. The two different plans which have been generally advocated for controlling the levels of the lakes are, to construct a dam, with regu¬ lating sluices, across the Niagara River below Tonawanda, N. Y., or to construct a submerged weir in connection with a set of regulating sluices at the foot of the lake, just below Buffalo Harbor. A preliminary study of the problem and estimate of cost of regulat- ing works, based on these surveys and examinations, developed the fact DEEP WATERWAYS. 283 that the first of these plans would require an expensive dam with lock and waste weirs in the Niagara River on each side of Grand Island, the excavation of over 5,000,000 cubic yards of material in the head of t he river, the purchase of at least #0,000,000 worth of property which would be ruined by the works and high water along the river front, and the construction of several miles of dikes to safely maintain t he impounded water above the level of adjacent country. The distance from Lake Erie to the site where dam would have to be constructed is 12 miles, on which the high-water slope of the river is about 8.5 feet. With the river improved by regulating works and enlarged cross section of channel through the gorge, this high-water slope would be reduced to about 2.5 feet and the low-water slope to 1.5 feet, making the fluctuation of the lake due to change of slope in river for different volumes of discharge approximately 1 foot, which would be increased 0.5 foot by change in velocity head at foot of lake, or a total probable fluctuation of 1.5 feet when the discharge of river is controlled by regulating works for maintaining the river at a fixed stage at a point 12 miles below outlet of lake. The total cost of the project, including damages and the necessary drainage channel for taking care of Tonawanda Creek and the water from adjacent country, would be over #12,000,000, which, with the fact that the lake would still have considerable fluctuation, practically eliminates all chances of the plan receiving favorable consideration. If a deep waterway should ever be constructed from Lake Erie to Lake Ontario via the Tonawanda-Olcott route, the improvement of the river by regulating works below Tonawanda would diminish the cost of the canal about #6,000,000, which would still leave a balance of #0,000,000 chargeable to the project. Regulation of the lake levels by means of controlling works in the foot of Lake Erie will require either a submerged weir of such length that the change of discharge over crest of weir, due to a few inches variation of stage of lake, will be equivalent to a variation of outflow through the gorge at the head of the river due to 3-foot change in depth of river, or a short submerged weir in connection with a set of regulating sluices so designed that, with the sluice gates all closed, the low-water flow for the regulated stage of the lake will be discharged over the fixed submerged weir, and with the sluice gates all open the additional volume of outflow necessary to maintain the lake at nearly the same level will pass through the sluices at times when the lake is receiving its maximum supply. The surveys and examinations indicate that a combination of a fixed weir and regulating sluices is better adapted for an economical and complete control of the lake level than by means of a fixed weir, and the plans and estimates submitted are for such a project. In order to properly proportion the height of the weir and the width of sluices, and determine limits within which such works will control 284 DEEP WATERWAYS. the level of the lake, the observations for determining the volume of discharge of tlie Niagara River were made for as many different stages of the lake level as possible, and Ihe following equation estab¬ lished, showing the outflow for any given elevation of the lake, viz, Q = 168,812+17,762 y+ 1,409 if. In which Q=the volume of discharge and i/ = the elevation of the lake at Buffalo in excess of 570 feet above tide water; dr y= stage of lake—570 feet. In connection with the direct observations made to determine the outflow from the lake at different stages continuous readings were made with gauges from which the slope of the foot of the lake and the river surface has been computed for the corresponding volumes of discharge, which, with the carefully measured cross sections of the river shown on plate 84, furnishes a check on the equation of discharge given above and shows a discrepancy of less than 3 per cent for maximum volume of outflow used in this discussion. It has been found from an examination of the records of the lake water gauges at Buffalo, Cleveland, and the mouth of the Detroit River that the relative elevations of the gauges have been incorrectly established, and that the zero of the Cleveland gauge is 0.30 foot high, relative to the gauge readings at Buffalo, and 0.48 foot high as com¬ pared with that at the mouth of the Detroit River. There is some uncertainty in regard to the stability of the gauge at, Buffalo previous to 1896, and since the level of the lake is least affected by winds during the months of June, July, and August, the compari¬ son of levels at Cleveland and Buffalo is based on the gauge readings for these months during 1896, 1897, and 1898, which are as follows: Elevation of lake. Date. Buffalo gauge. Cleveland gauge. Differ¬ ence. Feet. Feet. Feet. 1896.•. 571.57 572.01 0.44 1897.. 572.23 572. tiU .43 1898. 572.25 572.69 .44 Mean... .... .44 Correction for slope in foot of lake.... .13 Zero of Cleveland gauge too high... .31 A series of observations was made with gauges established in the open lake outside of Buffalo Harbor during July and August, 1898, which, compared with the record of the Government gauge at Cleve¬ land, makes the zero of the latter gauge 0.30 foot too high, relative to the elevation of the lake at Buffalo, which amount has been used in computing the following table of elevations of the lake from Cleveland gauge records: DEEP WATERWAYS 285 Monthly mean elevations of Lake Erie at Buffalo, N. Y., for years 1865-1898, inclusive. Year. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. Means. Jan. to June, , inclu- i > ear - sive. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. 1865 .... 571.80 571.22 571.54 572.26 572.84 572.82 572. 78 572.70 572.66 572.36 571.98 571.84 572. 08 572.23 1866 _ 1.57 1.41 1.80 2.38 2.60 2.86 2.97 2. 72 2.66 2.65 2.41 2. 42 2.10 2.37 1867 .... 2.13 1.81 2.21 2.53 3.05 3.36 3.17 2.86 2.47 2.13 1.63 1.41 2.52 2.40 1868 .... 1.21 0.83 1.42 2.25 2.70 3.09 3.06 2.54 2.27 1.82 1.66 1.45 1.92 2.02 1869 .... 1.44 1.37 1.85 2.15 2.70 3.09 3.37 3.27 3.00 2.55 2.09 2.44 2.10 2.44 1870 .... 2.68 2.91 2.68 3.33 3.54 3.51 3.55 3.50 3.25 2.87 2.57 2.45 3.11 3.07 1871 ... 2.24 1.91 2.36 2.84 3. 11 3.11 3.12 2.91 2.74 2.07 1.89 1.45 2.60 2.48 1872 .... 1.37 1. 13 1.04 1.24 1.68 2.05 2.04 2.01 1.78 1.61 1.28 1.05 1.42 1.52 1873 .... 0.95 0.96 1.03 2.31 2.9S 3.06 3.04 2.98 2.58 2.28 2.08 2.45 1.88 2.23 1874 .... 2.84 2.89 2.92 3.09 3. 18 3.25 3 28 3. 12 2 . 66 >> »>«> ISO 1.59 3.03 2. 74 1875 .... 1.36 1. 19 1.33 1.73 2.20 2. 63 2. 76 2. 75 2.61 2 12 1.97 2. 19 1.74 2.07 1876 .... 2.15 2.71 3.36 3.88 4.20 4.31 4.20 3.90 3. 73 3.20 3.28 2.94 3.44 3.49 1877 .... 2.54 2.38 2.15 2.58 2.83 2.91 3.15 3.01 2.93 2 53 2. 45 2.53 2.57 2.67 1878 ..... 2.61 2. 75 2 . 88 3.30 3.56 3.56 3.56 3.32 3.19 2.84 2.64 2. 72 3.11 3.08 1879 .... 2.30 2.16 2. 19 2 .55 2. 70 2. 79 2.82 2.60 2. 27 2.04 1.57 1.83 2.45 2.32 1880 .... 2.33 2.37 2.51 2.67 2.94 3.05 3.14 2.90 2.67 2.23 2.15 1.81 2.65 2.56 1881 .... 1.40 1.51 1.83 2.53 2.93 3.17 3.12 2.80 2. 45 2.40 2 22 2.43 2.23 2. 40 1882 .... 2.90 2.90 3. .35 3.57 3. 77 3.92 3.85 3. 71 3.44 2. 99 2.67 2.16 3.40 3.27 1883 .... 2.07 2.28 2.47 2.59 3.05 3. 75 3.95 3.89 3.58 3.26 2.88 2.91 2. 70 3. (Hi 1884 .... 2.58 2.84 3.03 3.58 3.85 3.93 3.71 3.55 3.12 2. 79 2.31 2.24 3.30 3.13 1885 .... 2.06 1.85 1.71 2.53 3.26 3. 77 3. 73 3.74 3.59 3. 49 3.37 3.32 2.53 3.04 1886 _ 3.34 2 . 61 2.42 3.30 3. 60 3.70 3. 68 3.47 3.23 3.00 2.59 2.64 3.16 3.13 1887 .... 2.41 2.83 3.64 3.66 3.84 3.87 3. 63 3.31 3.08 2.49 »> 2.24 3.38 3.10 1888 .... 2.06 1.79 1.89 2.52 2. 77 2.90 3. 05 2.95 2.51 2.14 2.20 2.08 2.32 2.40 1889 .... 2.10 1.94 1.78 2.13 2.31 2. 74 2.94 2.63 2.24 1.82 1.55 1.81 2.17 2.17 1890 .... 2.17 2.46 2.58 3.07 3.41 3.78 3. 40 2. 96 2. 77 2.58 2.55 2 . ft2 2.91 2.84 1891 .... 2.10 2.08 2.54 2.41 2.23 2.37 2.27 2.00 1.82 1.44 1.00 1.07 2.29 1.94 1892 1.10 0.89 0.93 1.49 2.29 3.05 3.17 2.82 2.50 1.94 1.61 1.34 1.62 1.93 1893 .... 0.96 1.04 1.26 1.99 2.83 3.02 2.74 3.40 2.02 1.67 1.27 1.35 1.85 1.88 1894 .... 1.63 1.51 1.54 1.94 2 . It! 2. 63 2 52 2.15 1.98 1.66 1.42 1.35 1.93 1.89 1895 .... 1.02 0. 79 0.80 1.05 1.27 1.36 1.25 1 17 1.07 0.59 0. 49 0.65 1.05 0.96 1896 .... 0. 75 0. 67 0.62 1.07 1.45 1.72 1.60 1.81 1.49 1.25 0.88 0.91 1.05 1.19 1897 ... 0.88 1.08 1.45 2.00 2. 33 2.43 2.42 2.26 1.98 1.49 1.36 1.33 1.70 1.75 1898 .... 1.38 1.58 1.84 2.42 2.57 2.60 2.38 2.18 1.80 1.60 1.48 1.31 2.07 1.93 Mean 1865 to 1898, inclusive. 572.36 572.40 The uncertainty in regard to the stability of the Buffalo gauge pre¬ vious to 1890, together with the excessive fluctuations of the lake levels at Buffalo, make the Cleveland gauge record much more reliable, and it therefore has been used in determining the mean monthly elevations of the lake. By substituting the monthly mean elevations of Lake Erie from the above table in the formula for discharge of Niagara River, the follow¬ ing table has been made, giving the mean discharge of the river for each month from 1865 to 1898, inclusive: Monthly mean discharges of Laice Erie at Buffalo, N. Y.,for years 1865-1898, inclusive. 286 DEEP WATERWAYS. • CO tO -h 05 "O ® -r tO ».C tO ® >0 — -+• 71 CO tO -t- 4 - 3*. — ® X' X) CO CO ® X X 40 71 —' ~ ~ fa os fa —’ fa -1 ^ » s i-1- s i i x a * i- r. t ; ; $ ir £- 3J ~ i ■ 4 i — — 7 i — -^cococococo — — co c: _ pc®ac _ J . ^J W M fl M 01 —i 7! fl 71 71 71 71 71 717 * 7 l 71 7! 7$ 71 71 71 71 71 N 71 71 71 7! 71 © £>■ *© O +J <, > iO 5 7. 7D7 -t - C X C <7 7. *1" -f - 7. M Oi ^ 7. C C7 J3 7! • C »7 71 O X ^ C 4 , CO ■— 1 ^ ‘7 r7 O 7 f C7 7. X ^-t< a 71 75 C 3“. X CD 30 CO 7! X 7 * CO t- *5 71 05 CO *0 tp »0 X '*"» oe ®®' 05* -r 7* 71 CO’ 7 » -T ®* -* — >7 -r CO tO ®' « - —’ CO 7 <- tO ‘O -*1 -* 71 CO -+* ».7 O a.71 7J 71 71 71 71 71 71 7* 71 7! 71 7* 7* 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 C3 c ft- ft ci *"7 a3 I 3 i Eh © a) P i - i - X — —< t- X X 05 1 - 71 CO ® — 1 - ® —I X *7 t- tO *® CO < - >17 —< CO tO J— 4 - JD- - “ 71 ~ CO > 7 —<0 — < - ~ — 7. 7 71 7! 7. -r 7 ! i-C 7 ^ i7 X tO ~ *7 71 <7 71 71 X X -** 71 ® ^\ J CO^X)'^COOaiOt-^D-H^i-C002»7COiC^-'-H-^71-fO: 71 *.7 711- X *7 71 — X * v "- 7 -r td tp CO N X x 71QCO CO CO 711-* -* -7 --7 -t*’ >7 —* 77 — 1 ' - — CO -r —' -r »7 ®' — ' i-i-7 iCi-7» 7 X *7-fa 71 X 77171 i7 - 7 5 -tCCI - 7 XOC"— -f «0 CO CO 71 1 -- 7 ' r 71 71 CO 71 71 CO r- 71 71 ^ CO -r CO CO 71 CO CO -H CO 71 CO CO 71 71 7 1 —■ —< 71 71 ■*7* ^ tO 05 05 CO CO O CO 05 Q t- 05 -f 71 — *^ < - X> --♦* 71 71 »7 t- 71 t-- 0 CO X 71 71 CO CO — 1 Q O' •. *5 10 CO -f 7 : CO t—I Gi r—I CO —> Ih- O CO <■- v^ o CO —I 71 O t-H- 75 CO 77 CD »7 ^ CO w ^ O' 71 3 *-7 »7 »7 i— »7 O t~- C0-0 7ll'-*70*77ll--- 'jD CO »7> ZQ X *7> ID -O *—• CO' CO O: •—i 05 *7 4 cT 77 t- ~ ON aTo o ^CO71 I^T® »C O CO 7l >7 co'>7 *7 tT»7 »7 >7 ~ O -0 -* ^ • 7 71 7 7. 71 71 7. X 7! 7 H- 7 71 71 7 7 7! - CO - 7 71 *h 7 - X O 7. 7. X> X) 7. 7. ~ 7 -? t 71 f-i »— 71 7 1 r—( 71 71 7l 71 71 71 71 71 71 71 71 71 71 71 71 71 71 ^ O * O ^ N c; O 71 7 O -O QQ O 55 -f CQ O 1-- ID C -0 »0 niCC: Q X 71 »-• CO 71 O <- 71 CO X X ^ •. 3 Oi-iXQl-HOiO^NrfOiNNi-i cr. X OC1-CCC CO i - X O-t(-OCCNQ0 ^ ^ifiXiC'-CN^XCCCO?17H-iOr-i>CH7lOC0 071^NN7lJ.OOXCOO»i:-' '4 3 : 05 r-t 71 71 COt-CO»-H»OC:71Q»OOCO».0*-Or-ti-^^»i0^05COl-^COO <- »0 tf5 X ^ • O - O C -- 71 O O - C i O 71 71 C - H- 71 CC - 7- 7! - - o. C! X C 35 a N X 7^ 7. i 3 0 7171 71 71 71 71 71 ^ 71 71 71 71 71 71 71 71 71 71717171717171’—'71 — 71^- — —— O C ^^COOcp-HN05Cl*7Qr-'ai7lO'C0HC0X'O(-h 05 O X O 7. O O O 71 CO ON X 1 ^ '. X xo ON O - 7 ^ 7 ONN7MCO I.C - C. ^ ONNN 7 - -N 3 X X 7! C CO ^ ^i7NCX71C0O5o-XONO3:TP(7iCO«7.i-N71XOC7i.CO-(-71OX l . 4 00 »o co >0 co'—- — —h to 40 7i 7i cTo »o 01 '-H — a;' i- -t-' —^ co »o < - x oi oi acox -> . — 71 — “ 71 CO — — — —■ -+ 71 CO — — — CO -► 71 -t- CO 71 — 3 71 05 OC C ONJ35 05 O i 3 ^ 71 71 71 71 71 71 71 71 71 71 71 71 71 71 7l 71 71 71 71 71 71 71 71 71 71 71 — 71 71 71 — 71 0 a, _ "^^^7*70100540—i-HOSTlOOCCSSiO^gCOQONNQQCO'MN-pCOMNMNNO: ^ O CO CO X 07 1- 7lC0 35'HC0NN4.CNOX05C‘.O'tC0XXNN7lC7l'+CC0pC^ ^ « 7lCOl-t * XOONOO CCC7H-iCO‘C5NXXClOXXC^'!:O«'O».CC0 ^ . ^•X)ti 1 -®^-X'tO0O'+710Ol0pO0N0C0 71 *0 X »0 71 O Ol 05 X ® »0 ^ • 7171 71—ICO -H 71 O 7! 71 71 40 CC -t — 71 71 -r< *0 -O »0 -r CO 7 1 — 71 7C 71 — _ X 05 — ~ ^ 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 —< — 71 71 0 a, "8 1 1 - -h 7. CO 05 71 N N X N Tf CO iO X ri 71 CO X N CO 40 CO >7 47 -O 71 X N O O N O’ - ^ •. -1- »OCO 71 *0 CO CO O 40 -T- — — -f — — 1- o: O 7! 71 »0 — i- N CO t - ~ »0 — 711-40 CO ^ w 40 — — 05 7! -1* 71 71 05 CO 40 O CO 40 —< 40 r-H 71 ID C5 ^ O -f* 7! i- 05 —I *0 40 40 40 — 7 } 4 N t- -h CO ^ X 1 71 O O 1 - X 05 *0 CO 71 05 -t* 05 05 i- CO 02 40 0? 05 ~ 05 CO — 40 50 -r • 71 71 CO 71 -r CO — CO CO 7 1 iO CO -r< 71 CO 7l 40 »0 -+• »0 — CO 71 CO — CO ^ —< 05 C: — — £ 71 71 71 71 71 71 71 71 71 71 71 71 7! 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 — 71 71 71 0 a. _ -Jr ^3 71O —' 40 X CO CO CO 71 X 40 X —1 x ^ 05 CO O 05 i - 1 - H 40' O —I 71 O X '05 X ^ X 05 X O N 71 *t CO CO N O’t O 1 O X *f X iO w N 7 »C 05 -h 05 05 X tp 71 71 CO -r *0 ^ ^ O 05 Cl W to CO O 05 X 71 lie Cl N 05 ri a C 05 ri O T 1 ! X C 71 Ht CC 7! C 4.C 71 CO c c .' 4 05~CO~05 CO 05* N O 40 of X X X Oi Q x' 1 - x' Q P CO —* CO CO 40 to' 05 ' x' 71 co '~ ~ 05 ' - . 71 CO CO CO ^ CO —< CO -+ 71 CO CO cO CO CO 40 cB 4.0 40 40 40 CO CO -t* —t CO 71 71 05 * 71 — JT r 7171 71 71 71 71 71 71 0 40 — X ^ ® 71 i - *•# co ® CO 05 X 05 CO X ® 7! 40 1 - CO X 71 05 ® i- t- CO 05 ^ ^ — I40tPt—(C0"H5t140^05C0®71O71C0XX®t-hO®X®C07140 407140 O ^ -t N N x X m o -H i.c rH 71 7 o O O O 15 O 1 C CC X 7t X O X O 4 .C «C iC CC C *t • CO CO CO co -+ CO H CO -+- 7 1CO -f 71 CO ‘-O' ** tO *0 »0 t.O CO 7 * *0 — CO cO 71 05 ® 7! 71 ~ O 7{ 7\J 71 7{ 71 71 71 71 71 71 71 71 71 71 71 71 7l 71 71 71 71 71 71 71 71 71 71 71 71 71 —< 71 71 71 0 a, _ ''r'd X t- — —• 05 — 7! CO X X CO 40 — -+• CO CO t- -t* 40 t- 1- — to -H t- to 05 40 CO 05 ® ^ % 71 —< 05 -+■ ^ X C2 *0 —< ® ® ® c: ® ® X 05 —1 05 71 CO CO COt-'O^-fCO-ft- ^ V O 40 C C O CO O O 71 X N Cl o 5. O’ 71 05 X O O O' O N X CC N -t X CO X CD 40 X N . H+*ONNO.N7lHf 07>X005NC071>00 XrH-NXN >0 40 ©QNCONJ-CO ^ • CO 71 CO' 71 7* H-f CO C. CO CO —< tO CO -+ 71 CO CO »0 CO »0 -H ‘O »0 71 — -H —« —< CO -h 05 05 — 71 C* 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 7171 71 71 71 71 71 71 71 — —' 71 71 °9,_ ^3 X 71 71 CO l- -* 1- — —• ® 71 05 —< —< 71 71 05 ® 71 X 71 O CO ® CO 05 CO 71 05 ^ ' .Otpi-- — X 7! ~ X 4.0 CO -t< —U-l- X<- X ® COt-i-® 71 ^71® ® t-—JCOt^-^ ^ — C 1 - ® 40 tO ® 3 CO — t- ® ® < - 71 711- —< 71 -1* l- l- ® 40 CD CO X -f 1- tO ® 05 O . 75 CO oTofto' co'cO ® co'l—'i- co X*tH OICO to 71 ® P ®' 71 7 1 71* 71 CO CO 05 00 of QO oT05 05 ®' -o • —1 71 — — h-H CO ® — CO ® *0 71 71 71 7140 71 iO 71 -f 1 IO 71 r-• CO 05 ® Q X X ® 71 ^ Ol 71 71 71 71 71 71 r-H 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 —l 71 71 H — 71 71 _ "tr J '^3®®®t^®C0®7l4O7l^HC0®4OO5C0«0*—tCO®cO®t'-'^<^05rH7171®COXCO® — -h -tM - 05 CO X”^ ® 05 CO ® —1 O CO i - 00 CO X i - ® CO — 05 — 71 40 CO —< 7 V 0 CO to ^ 40 CO 05 X -* 40 40 X ® ® ® -f *0 CO 71 C — 71 *0 CO ® —< ^ X ® ® tO ^ 40 ® CO »0 71 . 41 0 . »C O *0 to X X X 7M< CO - ’O 71 O 4* r- lO CO ® 71 * CO ® CO Oi CO ® i - O - • a P — 05 ® 71 — X X CO 05 -f — CO — 7! C 71 CO ® 71*0 ® ® 71 71 X 05 ® X X 05 ® r ~ ^ — 7171—1717171— 1 —t 71 — 71 71 71 71 71 71 71 71 71 71 71 71 75 71 71 71 »-* —< —t — —. — 71 ^ a _ "C — ® i- X CO 7! 05 40 40 CO to to X t- to —< to to ® CO — CO CO t- 71 ® X —I 40 2 X *0 i z 71 ® CO X ® — -11 05 to H 40 71 -111- CO 71 05 l- CO 711- CO tO CO —1 HN 71 ^ ® ^ ^ 55 to *0 40 i - -f X ® —. 05 05 7! C CO 1- X X —1 t2 ® ^ 1 1- CO ^ 40 ® X t- X X t- CO *® CO • * 2J ■£ iQ zT *2 71 io Q —I —i t- 05 x'co' x' x 71 to cf to -*' ®’ 4 o'x -h to'x x co —' oT c* 2 •„ Oi ® ® X 05 CO C 05 X CO 05 71 — 7! — —t 05 CO — fO ® 71 CO ® ® 71 —< X X 05 X X X ^ ^ ^ 711— — 71 Cl — —i 71 —i 71 71 71 71 71 —1 71 71 71 71 71 71 71 71 71 71 *-• —i T—t —( -h — 71 ^ a« "Q ® -^1 ® X ® 71 CO — -fix® — ®X®®^H®O;-liCJ5C0'*t , ' , ^ , 71-ti4O4O®l'-®®® ^ *• rt i-- ^ ® — CO i - ® to 7 n - — 71 — -O -t- l - — — X o C X7105C140CO—iO5C0C0w ^ CO — ® CO C040 ® 1- 05 to 40 4-0 ® < - — X -f H CO ® CO X X CO CO 05 CO ® —i tO CO 05 40 ® • A S — 71 JR *3 ® ~ iq co co -o t- 1 - to’ oi’ — —Tco 05 tico ti o i- P oo 7i to to 2 J ^ r, ^ -• i! r 31 X ? 35 * 1 * 1 ^ ^ ~ 7! -I H+ r- - -< M ^ o: X ® 00 QO X a ft ^ 7 t 7l 7 J — — 71 7! — —1 71 — 71 71 71 71 71 —1 71 71 71 71 71 71 71 71 7171 —- 71 —- — — — ^ ft, ft a; 01 bn < 7 £ 13 0.90 i 10 0.94 6 3 13 0.96 3 9 i 6 0.80 4 5 4 0.91 (5 3 0 0.87 4 8 A 8 0.90 A I Q 10 0.92 11 0.93 5 9 A 11 0.95 6 11 0.90 1() 12 10 0.94 Transverse curves. Vertical curves. Num- Number Coeffi- Span Sta- of curves cieutfor No. tion. ob- vertical served. served. curves. 6 10 A 13 0.88 4 T2 1 0.85 ti T2 8 0.90 A 5 0.92 T2 8 0.80 10 £ 6 0.84 4 3 3 0.93 e 3 6 0.91 8 i i 18 0.95 i 10 0.94 6 14 0.94 9 6 1 10 1.01 Total 73 236 For the coefficient for vertical curves for intermediate stations it is necessary to interpolate. In span No. G the vertical curves were taken closer together, for the reason that there is a sunken caisson in the bed of the river about 75 feet below the discharge section. The caisson is located directly below station T 4 ¥ , and its effect is readily traceable in the cross section and curve of average velocity. A mean transverse curve is formed from the individual transverse curves of each span by averaging the velocities at the ^ depth JULIUS BIEN & CO PrfOTO UTH H Doc 149 56 2 - j \ y>\\, U >■% ' “ ■ DEEP WATEKWAYS. 313 point, these velocities in some cases being corrected for direction when near the piers. This mean transverse curve is reduced to a curve called the average-velocity curve by multiplying the average velocity at the t 3 q depth point at each station by the respective coeffi¬ cient for the vertical curves at that station. This curve is shown in plate 88 and gives the average velocity in any vertical plane and at any point of the cross section. The stage of the river corre¬ sponding to this curve in each span is taken as the mean of the observed stages during the times of the individual transverse curves. Mean-velocity coefficients .—The observations for discharge, as already stated, were made at the T 3 ^ depth point at meter stations, one in each one-lialf or one-third part of span. Assume for the present that the mean velocity in the one-half or one-third part of span will vary directly with the observed velocity at the T 3 ¥ depth point at the meter station. The problem then is to find the ratio of these two quantities so as to change from observed velocity to mean velocity. In connection with the transverse curves it will be remembered that observations were taken at the T 3 F depth point at the meter stations. Compute the discharge for each one-half and one-third part of span, corresponding to the mean stage of the transverse curves. To do this, draw a discharge curve (plate 89), the points of the curve being obtained by multiplying the average velocity at any point by the corresponding depth at that point. The discharge equals the area of the figure and is easily found by means of the planimeter. Obtain the area of cross section for each one-half or one-third part of span, corresponding to the mean stage of the trans¬ verse curves. Discharge divided by area gives the mean veloctiy. Then— Mean velocity_Mean velocity i or A span, Coefficient — Mean of observed velociries at T 3 „ depth point at meter station. It has been assumed that this coefficient would be constant. To investigate the subject, consider the discharge from each one-half or one-third part of span as forming a solid, the cross section being the base and the velocities the ordinates. The coefficient and form of the vertical curve at the meter station are known from observation. The coefficient for the vertical curves in the width of one-half or one-third part of span will not vary more than 2 or 3 per cent, and may be taken as constant for this discussion. Velocities located by percentages of the depth will all vary in the same proportion. The mean-velocity coefficient would be constant under the following conditions: 1. The cross section remaining constant while all the ordinates of the solid increase or decrease in the same proportion. 2. The cross section to remain of the same form while ordinates of the solid, located at similar points of the cross section, remain con¬ stant, increase or decrease in the same proportion. 314 DEEP WATERWAYS. 3. The velocities located by percentages of the depth to remain constant, while the volume and area of cross section increase or decrease in the same proportion. A change of form under condition 1 is possible, and as all ordinates change in the same proportion it is evident that the mean-velocity coefficient will remain constant. A change of the cross section to one of a similar form can not take place when the sides of the cross section are vertical. In general, then, we must expect a slight change in the mean-velocity coefficient when the stage varies. Condition 3, followed or preceded by a change under condition 1, covers all cases where the mean-velocity coefficient can remain con¬ stant. The simplest case in which these two conditions are satisfied is that of a rectangular cross section. To find the effect of a change of stage on the mean-velocity coefficient, for the particular solids we have, a range of stage of 5 feet has been considered, 24 feet above and 24 feet below the mean stage of each span, corresponding to the trans¬ verse curves. Assume the velocities at precentages of the depth to remain constant. Any requisite change in depth can be made in this way with some resulting change in the area of the cross section and volume of the solid. Then with the area of cross section remaining constant all ordinates can be increased or decreased proportionally and will not affect the mean-velocity coefficient. So the only change that it is required to investigate is the change of area of cross section and volume, while the velocities at percentages of depth remain con¬ stant. If these two quantities increase or decrease in the same pro¬ portion, the mean velocity-coefficient will remain constant. For a rise of river stage the area of cross section will be increased by the product of the width and increase in depth of water. The volume of the solid has already been computed for each one-half or one-third span from the curve of average velocities. This will be increased by an amount represented in one of the vertical curves (fig. 1) by A and decreased by the amount represented by B. The volume represented by A is obtained as follows: Find the area of one-lialf or one-third span of the curve of average velocities and divide this by the Average velocity in vertical plane Surface velocity obtained from the mean vertical curve at the meter station. This will give the area of the base of the solid A; the height will be the increase in the depth of water. The volume of the solid represented by B is obtained as follows: The form of the vertical curve at any point can be determined from the vertical at the meter station. It can be easily proven that the areas represented by B are proportional to the horizontal projections of the vertical curves at the respective points; but the horizontal projections are in the same ratios as the DEEP WATERWAYS. 315 average velocities. So we have the area represented by B at any point is proportional to the average velocity at that point. The volume can then be easily computed. In general for the solids we have both area of cross section and volume increase in about the same proportion. The following table gives the results of the computations for the mean-velocity coefficients: Table No. 4 .—Mean velocity coefficients for computing discharge measurements. Elevation of water surface at discharge sec- tion above mean tide at New York. Span No. 564. 565. 566. 567. 568. 569. [East 4. 0.80 0.82 0.85 0.89 0.90 0.92 1 \West 4 .... .63 . 6S . 62 .62 .61 .61 |East 4.. * .93 .93 .93 .92 .92 .91 IWest 4 ... .95 . 95 .95 .95 .95 .95 [East 4.. .91 .90 .89 .88 .86 . 85 iWest 4. .82 .82 .82 .82 .81 .81 (East 4. .91 .91 .id .91 .91 .91 4 Middle 4 .. .91 .91 .91 .91 .91 .91 1 West 4 ...... .86 .86 . 80 .86 .85 .85 East 4 . .94 .94 .94 .94 .94 . 94 5 ■(Middle 4.-. .91 .91 . 92 .92 . 92 .92 1 West 4. .... .91 .91 . 90 .90 . 90 .90 East 4...... .86 . 80 .87 .87 .87 .87 6 1 Middle 4. .88 . 88 .89 .89 .89 .89 |West4 ...... .81 .81 .80 .80 .80 .80 [East 4. .80 .80 .79 .79 .79 .78 1West 4 .... .91 .91 .91 .91 .90 .90 [East 4 ... .93 . 93 .93 . 93 .92 .92 i West 4 ... .93 .93 .93 .93 .93 . 93 |East 4 .. 1.15 1.15 1.14 1.12 1.12 111 \West 4. 1.17 1.17 1.17 1.17 1.17 1.17 The observations of the individual vertical and transverse curves being approximately at the mean stage of the river, 566.5, the mean velocity coefficients for this stage are practically independent of any assumption in regard to the change in the form of the vertical curves when the stage varies. The mean velocity coefficient does not follow any regular law as the stage changes, but depends simply, as stated, on the relative increase or decrease of the quantities and of cross section and volume of solid when the velocities at percentages of the depth are assumed to remain constant. The effect of a triangular cross section in general is to make the mean velocity coefficient decrease as the stage increases, but the form of the solid may be such as to more than balance this effect and even reverse the law. It will be seen that the variation of the mean velocity coefficient for a change from low stage to high stage does not exceed about 2 per cent, and generally decreases. Curves of fall on Niagara River {plate 89 ).—This plate shows graph¬ ically the fall, in feet, from Lake Erie to the various gauges which are read. The curves were obtained from simultaneous gauge readings taken at ten-minute intervals throughout the day. From these gauge readings daily means and means of one-third part of the day were worked out and used as individual observations for determining the 3] 6 DEEP WATERWAYS. fall. Thus the daily mean is the mean of GO readings, and the mean for one-third part of the day the mean of 20 readings. The falls computed from these means were then platted and a straight line or curve fitted to the points. Considerable care has been taken in the selection of observations to see that the lake gauge 1 L or 2 L was not fluctuating too much and that the slope conditions remained fairly stationary. The following table gives the data in connection with the curves: Table No. 5 —Curves of fall of Niagara River . Number of obser¬ vations (see In¬ dex). 36 56 67 56 56 12 92 16 67 92 1 Erie Basin. The distances from gauge No. 2 have been measured along the cen¬ ter line of the river in about the mean direction of flow. The range in fall of the individual observations in gauges Nos. 1, 2, and 2 C was about 0.04 foot; in gauges 2 B, 2 A, and 2 C, about 0.10 foot, and in the remaining gauges about 0.15 foot. Within the limit of the observations the curves are thought to be quite reliable; beyond these limits they should be taken simply as an approximation. Computations for discharge .—The mean velocity for each one-half or one-third span is obtained by multiplying the observed velocity at yV depth by the mean velocity coefficient. This mean velocity multi¬ plied by the area of cross section of the one-half or one-third span for the observed stage gives discharge of this part of the span. The sum of the partial discharge gives total discharge. The following tables give the reductions of one of the measurements: Gauge No. Distance from gauge No. 2. 1 Feet. (>) 2 C 0 Q u 0 2D 1.220 2 A 2.550 2 B 3.260 2 E 3,780 3 4,370 3 A 5,550 4 7,070 6 12,250 rr 1 12,250 8 15,250 Equations of curves: x being fall from lake level in feet. y being elevation of lake above 570. x- x= fe x= x= X- x= X- e {£ !i. ii ■ 6 - 0.13. 0.678+0.010 (y). 1.11914—0.1714 /3' 14014-2/ for y < 2. 70. i -0.1776 +0.35531!/) for?/S3.70. 1.326 +0.037 ( y ). 1.606+0.037 (y) . ... 1.976 +0.037 (?/). 2.636+0.081 (y) . 2.981+0.081 (2/)..-. 4.5846-0.4714/3.09389-!/ for y < 2.45. 3.486 +0.294 (y) for y -t 2.45. 4.8126-0.4714/3.09389-2/ for y 5 2. +5. 3. 714 +0.294 (?/)_for y > 2.45._ 5.2444 -0.4714 /3.09389 -y for y <> 2.70. 3.9334 +0.376 (y) __for ?/ > 2. 70. 5.3244 -0.4714 /3.09389 - 2 / for y ^ 2. 70. 4.0134+0.376(2/) for y >2.70. 5.5304 -0.4714 / 3.09:489-2/ for y <, 2.70.. 4.2194 +0.376 ( y) for y > 2 . 70. Variation in y covered by observa¬ tions ob¬ tained. 1.8 to 2.6 1.8 to 2.7 1.8 to 2. 7 2.0 to 2.7 1.8 to 2.7 1.8 to 2.7 2.0 to 2. 7 1.25 to 2.7 1.8 to 2.7 1.8 to 2.8 1.25 to 2.8 1.8 to 2.8 1.8 to 2.8 DEEP WATERWAYS 317 Table No. 6 .—Discharge measurement No. 13. [Direction meter A, No. 12. December 3, 1897.] Station. Time of day. Elevation of water surface at sec¬ tion. Velocity observ¬ ed at A depth. Mean velocity coeffi¬ cient. Mean ve¬ locity. Area of croSs sec¬ tion. Dis¬ charge per sec¬ ond. Dis¬ charge by spans. /§. 8.12 565.85 1.63 0.85 1.39 408 Cu. feet. | 1,585 1 ||. 8.18 2.44 .62 1.51 1,018 i,7a5 2,095 8,304 8,663 13,136 21,512 17,408 18,048 16,257 14,579 12,324 12,451 6,439 6,703 6,946 5,071 3,990 817 ff.. 8.30 2.59 .93 .95 2.41 720 | 3,830 2 \f. 8.50 2. 71 2.57 815 ri. 8.55 4.75 .89 4.23 1,963 2,024 2.623 3,890 3,103 3,101 3,005 2,646 2,426 2,638 1,460 1,710 j- 16,967 | 3 11 9.00 565.77 5 22 .82 4.28 (A. 9.14 565.77 5.51 .91 5.01 4 9.18 565.77 6.08 .91 5.53 1 52,056 1^8. 9.20 565. 77 6.52 .86 5.61 [*£.. 9.26 6.19 .94 5.82 1 5 111----------------- 9.31 9.40 565.79 565.78 5.88 6.12 .92 .90 5.41 5.51 [ 48,884 (A.. 9.46 565.77 5.84 .87 5.08 | 6 9.53 565.76 5.30 .89 4.72 - 31,214 HI . 9.57 565.77 5.51 .80 4.41 10.02 565.78 4.96 .79 3.92 ]• 13,649 7 . If. 10.06 565.79 4.46 .91 4.06 1.711 8 /I. 10.10 565.80 4.35 .93 4.05 1,252 1,124 295 | 9,061 10.14 565.80 3.82 .93 3.55 9 10.18 565.80 2.43 1.14 2.77 | 875 . 11 U.43 1.17 1.67 a5 58 1 Approximate. Total, 178,121. Talle No. 7 .—Discharge measurements of Niagara River, 318 DEEP WATERWAYS. — pejjjuui pe^daocce snopBAaasqo , TS >1-3 m O »3 z - s £ a o fe'S ® P S§' 3 «S P, 5 T3 > o ^ I IIII IS ifiOOQOiflOCOQONiOO © I I I I I I I I I I I OOiCOOOONONCMO^ ^(5O!CC^tZ;H^!2iXC/iU3^C0 © rri > tJj . S •0* 2 © O Q S o S J x u x (- * c /. x 5 - « 7! r >t ? N ?. C X ?. X N C lC ?. X N ri L' x ir; vt GO © ofto to to ao co of t- oToT x"t- X X JiO 05 — :!>7 —< © 1 0 H-7! — t>- i- © O SHrHNNr-i Ol Oi 01 Cl Oi Oi Oi — — Oi Ci o CO — 01 © «0 < - C f © — »0 Of Ci ?. — CO -t< to 40 -r © 05 —1 »o i - — to —• uc 2ii 4- if 4 ;2 4 2 2 2 4 - 2 Ol 01 Cf Of Of 01 Oi Oi Oi Oi Oi Ci Of 01 © ria a • I* £ + i © fc£ b£ T. rf r* r^.Z-S: O 09^ Sf£fa ojvj uuds — O —I ~ '©< + + +I : I (NOS “ + + : .++ •ok uudg 3: 05 -*• -+ -f TO © co -* 05 SO «5 OS -* OS — O'tco^ioosHoacs ».o 1 - . x co oi a Oi to 05 1- © Oi co CO Oi Oi co r» + I I I I + ! ++ I + ! +++ I 1 O! N N t- --C +++ 1 + 1 + 1 + 111 1 T •OK uudg Relation of eleva tion of lake at Buffalo to eleva¬ tion of lake at Cleve¬ land, above +, below —. CO © © © © »-0 Oi © CO CO ^ © H jOrt>fiC’l ‘5 0 riOu 7 N 3 Xl>Oi .7 JrtOHrtH-dsitiHHHQOCd ^;+++1+++++E+i1++ 1l+il+tT 1 +++T+ Eleva¬ tion of lake at Cleve¬ land above mean tide at New York. ..^^^^SSSSSlSSsSsSSeSeSSSS coco^o?o 5 coc?§octico^c 6 c 6 ?i Fee 1 571 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . 1 . oi oi oi oi oi Oi oi oi oi oi oi oi oi oi © 0 ,o oj • M O u Eleva¬ tion used. |Q~*©C 5 XXOCOX—ia»f 5 - H t" 0 J m _.X 4000 05 © 05 c 5 »CCO©©CO» 0 ©©X ^ ©5 <—f oi oi o: oi oi co co oi co oi © © »-h ^ © t- t- ^- © t- L- t- t- i- t- L- b- t- L— k, 0 iot'-x©©c 2 x-**o-*xoii-© 1—i*-H'^'OiOl©©Oir—«cOCO"!t < ©'-^ oi oi oi oi oi oi oi oi oi oi oi oi -4 oi as Com¬ puted from gauge No. 3 . 1 © 05 XrH©OlOCOXr-t©X-H^-©CO ^CO-f©*—IXXHLCCO^OCJCO^XX • 1 1 • 1 1 • 1 • • • 1 • 1 • iiiiiiiitiiit M | rij ^ C5 -h oi co © 1 -h oi co co oi co oi © © -h -h 1 1 1 1 1 1 1 t 1 1 1 1 1 1 « 1 1 1 1 1 1 1 1 1 1 1 1 1 • » 1 1 1 1 1 • 1 • • 1 1 • 1 1 1 1 1 ’• 1 1 1 1 1 • t • 1 •• 1 1 1 1 1 1 1 1 1 1 ® M s> CC cc © f—>r^ *8 53 zH P 3 Is > © s a © 2 be § S P S3 o o&E&zi © 05 »-H Oi Oi 05 Oi Oi T3 - ©_; ® c bn °3 ? S 3 a ONOiOOOJNCONNa.N'f .C.lOOrHHLCriXlOXrH’tLCOCJO © ci r-5 co’ co o oi oi co' co oi co oi © © —* *-5 o £ © cC *£ cC rj © CX T3 o 3 ■2 =3 S ® *a o® o J-S '■£ co cC « > © ©42 'o 3 •eg © © • ,3 ■SP S o CJi . »s X CO ©5 Oi Oi co »o © O JO 00 *-< to ■*>• o x h 01 ^ 0 "*t o oi c h cf x o r. 5 ©. ... j»lOOXCOiOCONCOOO»XN«:iOCON COCO C l4 t- P- l- t- t- 1 ^. t- i>» N t- t- P O Of O O »C X l 7 O (N iC CO ^ O O O '•t CO O N X Of X 1 - X 1.7 O X Oi CJ 4 *.0 .© iCNXXlCNNOCXCOCONcbdN N N O CO O Oi O O O CO ‘O 05 Oi ».o © M-^XXOU'C5XC5ffiCX-H t-4*4 t3i>I cc" 1313 cc i- X) ' K ' i 1 *ri ~r> ^ eef ci Oi ot' — r-i oi q; *-t ' Of-t'i’ON i.- XNCCCOrHL7CC*.-'2-OXCi 0 cs o5 o o ^ os o o oi 01 oi si co GO X 05 05 05 05 x’ 05 05 05 * 05 05 00 05 x x o; oi co 05 x» -1.7 c. - c -c 'f rt< t- ^ o 05 CO EC© © N CO i> t- L- t- t- {- ^o6^oooo^ot’4i-ow’fE7»x -^c5^Hoicoo5oioicocooicooio©^*-i N t- N © i-» t- {.'« t'- t’. i— (>. *0 _ iot-opoocox^io^xoii^© h h -? Oi oi to © Oi ^ co co ^ a IIOI^'BAJ0SqO JO 0TBQ -o ^ Oi CO © ;© ; ;rHOi , , K - § > t£ > > 3 ^ > . ■£ .j^ -iOCO o 00 . . 0 © ■© . £££ : :£Q 'Q : • 2 P K O 5 © *—< oi co *Q x -( Oi Oi Oi Oi Oi Oi ol -y >3 r^> J>> b>j *^JrQ K^j ^^'•O^^ '^ 4 ^ . >>>>>>2 >> ■aoij'BAJ0sqo jo o^j f-f Oi CO ^ uo © 1 -^ X © © — Oi x »*o © t- x © cr —101 cc »o © x o: © »-* —< ^ 51 oi 01 oi 01 oi oi oi oi 01 co DEEP WATERWAYS. 319 gfcffieasssgssssssssessaaisasasBspfeeSssgssssssesss t^t^t^:dj^t^fc^t^sccdccpcdcdcd€dcocda6cdt^t^:dcd«didtfStf5»dididr^?dcdsocdedGC go cc cd 2ScS CC GO 00 3S3a£&3833;322SS:g!885'S@;gS3S£SSC£SsS8$3S;33g£S53 05 05 05 GO 00 CO 00 CO* 00 CO CO* GO GO l- i- t- N®CXCOXGOXCOd GO CO 50 & 05 GO GO GO CO GO O C5 p* S85S8 SSi aS&SSSSSSSS&SigS c$SS88 :s^sg[G :r:n:uunr:§§ § ' oco ■ *C i- ?r • *.c t- :2?]$ > ^ > r$ :: :!s i! -Us :35 ■: iSSSl :!::!::£! :I£I33: o o r o f d SSSeSSSioSS^^S^SS^SSSSaSSiSS&SSSsSSSSSioSSS^ a .rements Nos. 34 to 73 were furnished by the United States Lake Survey.] 320 DEEP WATERWAYS. Table No. 7 gives a list of all the discharges. Columns c, d, e, f,g, and h give the weighted elevation of the water surface at the various gauges at the time of the discharge measurement. The elevations were determined from simultaneous gauge readings taken at ten-min¬ ute intervals. The gauge readings were worked out which correspond with the time of observing for discharge in each span. These read¬ ings were then weighted in the proportion of the discharge of the respective spans, thus giving a weighted mean elevation of water surface during the time of the discharge. Gauges 1 L and 2 L were taken as the lake gauge. They gave results for lake level which agreed within 0.01 foot, so either could be taken. In some instances it will be noted the lake gauge has not been read, and in such cases it was necessary to use some other gauge as a base from which to com¬ pute the lake level. Columns i, j, and A’ give these computed lake elevations. The working up of the gauge records showed that gauge No. 1 was not a reliable base from which to compute lake level. The gauge is located in Erie Basin, and the entrance is protected by break¬ waters. With the lake surface fluctuating rapidly the water surface in Erie Basin would lag behind that of the lake surface. This, together with local currents around the breakwater resulting from winds and Buffalo Creek, seemed to justify the throwing out of read¬ ings on this gauge as furnishing a base for lake level. Lake elevations computed from gauge No. 2 are given the preference where this gauge has been read. The law for the fall from lake to gauge No. 2 is well determined within the limits of the observations for fall, and even beyond these limits, with slope conditions fairly stationary, the error of a computed lake elevation would probably not exceed about 0.05 foot. The error from gauge No. 3 under the same conditions would probably not exceed about 0.10 foot. Column m gives the elevations of the lake at Cleveland for the dates of the dis¬ charge measurements. These elevations are a good approximation to the mean elevation of the lake for the respective dates, and.have been taken as such. Column n gives the amount the water was raised or lowered from the mean level under the influence of wind. Columns o and p give the fluctuations of the level during the measurement. In working up the results of these measurements, in place of consid¬ ering all observations, and with equal or unequal weights, it was thought a more reliable law of the relation of discharge to lake level could be obtained by taking selected observations. To this end the following arbitrary limits were used in the selection of discharge observations: Elevations of Lake Erie at Buffalo: Not to exceed 0.60 foot above elevation of lake at Cleveland; not to exceed 1.20 feet below eleva¬ tion of lake at Cleveland. Fluctuation of lake level during observation: Not to exceed 0.50 foot. Wind velocity: Not to exceed 15 miles per hour (fresh to brisk). DEEP WATERWAYS. 321 It will be noted from the table of discharges that with strong winds the slope conditions are very changeable and gauge fluctuating. With the wind from the northeast or southwest the water is raised or low¬ ered, respectively, from the mean level of the lake, and there is danger of the discharge being influenced by the conflict of the return current and current in the river. It is thought that, with the limits used, only the most reliable discharges have been included. Out of the 72 dis¬ charges, 13 have been accepted, and cover a range in lake stage of about 2.3 feet. A curve has been adjusted to lit these 13 observations by the method of least squares, giving each observation a weight of unity. The equation of the curve is— x = 168812 + 17762 ( y) + 1109 (y)~ where x is the discharge in cubic feet per second and y the elevation of Lake Erie in feet above 570. This curve has been plotted on plate 89, together with the discharges taken from Table No. 7. In the measurements, where the lake level has not been directly observed there may be a slight error in the values used, but the changing of these elevations by 0.05 or 0.10 foot will not affect the position of the discharge curve by more than one-fourth of 1 percent. Measurement No. 28, while complying with the limits for the selec¬ tion of observations, seems so far outside the range of the other meas¬ urements as to indicate some unusual condition in the slopes, and it has accordingly been rejected. The greatest error of an accepted observation from the discharge curve is about 6,000 cubic feet per second, or about 2.7 per cent. The range in the discharge measurements at any lake stage is about 5 per cent. It should be remarked that outside the limit of the obser¬ vations the curve can only be expected to give an approximate dis¬ charge. In closing this report, the writer wishes to acknowledge the valua¬ ble Assistance of Mr. (4. B. Mitchell, who has assisted in making the reductions. Very respectfully, C. B. Stewart, Assistant En g ineer. The Board of Engineers on Deep Waterways. Appendix No. 8. REGULATION OF LAKE CHAMPLAIN. The proposed project for a ship canal from Lake Ontario to the Atlantic via the St. Lawrence and Hudson rivers requires that the supply of water for locks south of Lake Champlain and for power for the operation of gates, be taken from Lake Champlain at times of low H. Doc. 149-21 322 DEEP WATERWAYS. water, and in order to secure an adequate flow from Whitehall to Fort Edward the level of the lake must be maintained at about 100 feet elevation above tide water. Any less elevation than 100 feet for the summit level will add greatly to the cost of construction, and any rise above the natural high-water stage will cause damage to private prop¬ erty around the lake and make it necessary to construct regulating works to control the flow through the canal. The greatest known range of the lake levels is about 10 feet and the greatest range of the monthly mean levels about G.5 feet. The fall of the Richelieu River varies in a distance of 22 miles from 0.4 foot, when the lake is at elevation of 93.5, to 2 feet for lake at elevation of 101.5 above tide water. The discharge of the Richelieu River for different elevations of Lake Champlain has been computed from the corresponding depths on the crest of the dam at Chambly, and is as follows (see figs. 1 and 2): Elevation of lake at Fort Mont¬ gomery. Discharge of the Richelieu River at Chambly (cubic feet per second). Elevation of lake at Fort Mont¬ gomery. Discharge of the Richelieu River at Chambly (cubic- feet per second). 94 . 5,000 8,500 12,000 15,500 98 ... 19.500 24.000 29.500 30,000 95 . 99 . 90 . mo .....:_ 9T ... 101. The net supply to the lake for any given period is equal to the sum of the discharge and storage for the same time. The area of Lake Champlain is 43(5.7 square miles, and a rise of 1 foot per month is equivalent to storage at the rate of 4,632 cubic feet per second. The following tables give the monthly mean elevations of the lakes, the corresponding discharge of the Richelieu River, and the monthly mean supply to the lake when rising, from 1875 to 1898, inclusive: CM CD lO C5 rH o o Q W ' APPENDIX 8, FIG.2. CM Monthly mean elevations of Lake Champlain at Fort Montgomery , N. Y,, 1S75 to 1898. DEEP WATERWAYS 323 Means, 1 year. SSSSSSSSSS53S S g £ !3 SSgg33$S8 Decem¬ ber. ^asssggsssssBgsasseagssss gsssgsgsssggssgssgsggssss K, Novem¬ ber. ..gag»5SSSSSgSSS8SSg:#SS23S te. October. k. 2 . ■Sfe P *0 <2 .§5!Ss33»3Ig§?3e?}?S5SaSS5sSt5S—5 333$3333333333S$S3SS333sa? fc, < ..sssssgggsssss&sgs&gggggs jggSsgSSggSgggggSSgSSSSgg k, July. .3S£3SS3E$iSS:g5ggS33$3$3S5 SSsSSSSSSSSSSsSSSSSSSSSgSS k © 0 S .££S£gS~£2?2gS£c3R3£H?S3s=;S ig£gSg£§ggggagg££gggg£gSS K, May. .3SSSSS3g5SSSSc£S??g9??Sgg3g 8 3 S' g 3 S g ?: S 5. 3 ?: g s 5 g S 3 g 3 3 3 ® 3 g k, ^ — April. .8SS3SSSgSSSSSggSSgSS?SS3S 8 g 3 g g g g a g g s g 3SS3£33gggS3 3 K, - 1 A £ X . S—g3SS3SSS33gg3S33g3g3Sg t3gS**ggg33£3SS*gggggSgS3 Febru¬ ary. ^.SSSS88SSg83gS36g2888®5gS3S k January. ^3S3S8$aSS3S3S3SSg5i83aggg r $33S3S;S3SS ssgssgsssssssss K. Year. • • i » 1 i • • * 1 i • « i • A t • • 1 « * • 1 imimmiimmimMiii Monthly mean discharge of Lake Champlain, 1S75 to ISOS, 324 DEEP WATERWAYS. O 3 s o x 1 k i s 1 I p a © Mean monthly storage and supply to Lake Champlain , 1S75 to 1S9S. DEEP WATERWAYS. 325 I ■si 3 I ii ri + a a <8|3 <§ §?f& o^co ill ?5aS |1 as Si 11 av; of >» a a •4 6 £ 131 5 Cl 2 ■*#* * *& ii CO IS £ CH ■ ?! 33 S3 : a a I mmmmmmmnmmm g t si § ?i s s * 3 3 i s s a a g s a a ® s a @ s? s a T a «J N 3- CO 30 to —' r i I. 3 '"- H Of C2 Of rt*' ~ X* —* Of —* CC - s; aessasssssssfesgsssisgggssg •5 ANrtjiHHdoc'OTNNHrijiHdrtN-iarijirio « fe, I a p efi •k §s a » 3 Lai?.? r- r> € «f»a'ss353a*a ! | ■ ail5Sl5S23 = 23 v,»ScMS^a S a. — a. 11 CO Cl = 3sS*2S 1*0 50 1-1 i.— C t' s a o CC ^ . co oi ^ -+ —i oi t- 1 — oi t- 6 § of of 1 cot-* mi- i* a 30 ! Rise. ic to ?3?3 SSSSS38 ^a=aaa=^a_^ ^ w w i -1 Cr-^rHCH as §1 5 5 a SiggsS 3iWWzra P< P< P cz: 11 L*S ss iU a ro% = a 55 g aa 53 a a l c ;S £? 5ia oo 1ISiSI i 1i111$i1 x §111I1I§ DEEP WATERWAYS. 326 The watershed of the lake has an area of 7,750 square miles, on which the average precipitation is about 33 inches per year. Cubic feet per second. Mean discharge of Richelieu River.... 13,700 Evaporation from lake in S months = 1.25 feet.. ..... 724 Total....... 13,424 One foot storage in lake = 386. Average rainfall on lake = 386 X 2.75...... 1,061 Average run off from watershed..... 12,363 Average run off = 65 per cent of the precipitation. The discharge for an elevation of 101 feet is about 33,000 cubic feet per second, or equivalent to the maximum supply to the lake. The lake has not exceeded this elevation since 1875 except when affected by wind, and the mean monthly supply has exceeded the correspond¬ ing discharge only once (April, 1896). The rapid rise of April, 1896, is thought to have been abnormal and due to ice in the Richelieu River diminishing the flow and thereby indicating a greater supply than actually existed. During low-water years the lake level falls to an elevation of about 94 feet above tide water in September and October, with a discharge of about 5,000 cubic feet per second; and since the power companies on the Richelieu River are entitled to the full value of the natural low-water discharge, any diminution produced either by regulation or use of feed water for the Hudson River Canal must be made good by an equivalent supply from the St. Lawrence River through the canal from Lake St. Francis to Lake Champlain. This supply will not exceed 1,500 cubic feet per second, or about one-fourth of a foot per second velocity in the canal. In order to maintain the low-water level of Lake Champlain at or above elevation of 100 feet above tide water, regulating works will have to be constructed near the foot of the lake capable of passing the maximum supply to the lake at times of high water and of main¬ taining the discharge at about 5,000 cubic feet per second during the usual low-water period in the fall of the year. Two different projects have been proposed for this purpose, viz: First. To construct a dam 5,000 feet long between Stony Point and Windmill Point, with a lock and regulating sluices at Stony Point, about 1.5 miles south of Rouse Point, the crest of the dam to be at an elevation of 99.5 feet, and the sluices of sufficient capacity to pass 5,000 cubic feet per second when the lake level is at an elevation of 100 feet above tide water; or, Second. To construct a dam with crest above the high-water stage of the lake, and sluices at Stony Point of such dimensions that when all are open the entire volume of maximum supply may be passed. DEEP WATERWAYS. 327 With the first of these plans, the discharge when the lake is at a stage of 100 feet will he approximately 5,000 x 2.8 x (0.5) * 5,000 cubic feet per second; and for a stage of 101.5 feet above tide water the discharge will be 5,000 x 2.7 x (2) 5=38,000 cubic feet per sec¬ ond, or practically the maximum supply to the lake during the greatest rise of the lake since 1871. The change of the lake level at the regulating works from wind effect and the consequent fluctuations of the discharge through the Richelieu River will be a serious objection to the use of a long sub¬ merged dam for the purpose of controlling the lake level. If the dam be built up to the elevation of high-water stage, and a sufficient num¬ ber of sluices constructed to pass the discharge at times of maximum supply, the flow can be controlled at all times and the reservoir capacity of the lake, due to change of level, utilized so as to maintain a greater low-water discharge than under natural conditions. The elevation of the lake at Fort Montgomery for a discharge equiva¬ lent to the maximum supply to the lake is about 101 feet above tide water; and as the head required to produce and maintain flow through the sluices will be about 0.75 foot, the high-water stage of the lake will be between 101.5 and 102 feet above tide water, or a range of the lake level of less than 2 feet, when not affected by the wind. At the site of the proposed dam rock crops out at the water surface at Stony Point, and at 500 feet from the shore is 20 feet below the pro¬ posed plane for regulated stage of the lake, making a favorable loca¬ tion for a lock and for foundation for piers and sills of sluice gates. If these sluices be made 20 feet deep, with an aggregate width of 300 feet, a head of about 0.75 foot will be required to pass the dis¬ charge at times of maximum supply to the lake. Such a system of sluices, with a high dam for the control of the lake level, will be more expensive than a long submerged weir with a single sluice, but the more complete control of the discharge which can be obtained by the use of the former is of sufficient importance to warrant its adoption and construction. A lock 300 feet long, GO feet wide, 14 feet deep on the miter sills, with lift varying from 1 to 7 feet, depending upon the stage of water, has been estimated for in connection with the regulat¬ ing dam and sluices. The maximum volume of water needed for feeding the Hudson River division of the Champlain route will not exceed 1,200 cubic feet per second, and for a large portion of the season will be much less. At times of high water in the Hudson River the water in the river at Fort Edward will be higher than in Lake Champlain at Whitehall, and, if desirable, a portion of the Hudson River flood may be diverted into Lake Champlain through the canal. Guard locks have been esti¬ mated for on this section of the waterway, so that the flow in either direction may be controlled as desired. 328 DEEP WATERWAYS. The details of estimate for regulating works and lock are as follows: Excavation in lake .... §89,223 Embankment . 12,188 Timber crib dam .......... 390,434 Sluice gates, piers, and operating plant... 122.300 Lock ._ ... .....I. 276.099 890.244 Engineering, superintendence, and contingencies, 10 per cent. 89,024 Total... 979.268 Respectfully submitted. Geo. Y. Wisner. The Board of Engineers on Deep Waterways. Appendix No. 9.—Part 1. INSTRUCTIONS FROM SECRETARY OF WAR FOR THE GUIDANCE OF THE BOARD OF ENGINEERS ON DEEP WATRWAYS. Orders.] War Department, Washington , October 20, 1S97. The orders of August 31, 1897, promulgating instructions for the guidance of the Deep Waterways Board, designated and appointed by the President under the provisions of the sundry civil appropria¬ tion act approved June 4, 1897, are hereby amended and enlarged so as to read as follows: Par. 1. The said Board is authorized to rent such necessary office rooms (except in the city of Washington) and, when the exigencies of the service will not admit of advertisement and contract, to purchase in open market such materials, including instruments, books, maps, field outfits, provisions, and other supplies of any kind, as in its judg¬ ment are deemed necessary for the prosecution of its work, and to employ such assistance as it may deem essential, and to pay such reasonable compensation therefor as it may deem proper, the expenses for all purposes not to exceed the amount appropriated, viz, 8150,000. Par. 2. The secretary to the Board, James II. Kidd, is hereby des¬ ignate] as special disbursing agent to disburse the appropriation in addition to his duties as secretary, lie will make requests on the Secretary of War for funds as they may be needed from time to time, said requests to be approved by the president of the commission, and in forwarding the same will name some United States depositary in which he wishes the funds placed, and state the address to which he desires the notification to be sent to him from the Treasury Depart¬ ment. 1 Ie will account tor all expenditures, as well as for all property, under the orders and supervision of the Board, and will give bond in the sum of $25,000 for the faithful execution of these duties. An DEEP WATERWAYS. 329 itemized account of all disbursements, with proper vouchers, will be transmitted monthly to the War Department for approval and trans¬ mitted to the Auditor for settlement. Par. 3. The Board shall authorize and supervise all disbursements, and the power to approve the accounts of the disbursing officer, so far as relates to the necessity or expediency of the expenditure, and the prices paid, is hereby delegated to the president of the Board, and such approval by him shall be final in the War Department. The approval of the Board by the president shall be indorsed on the account current. Par. 4. Journeys may be made by any member of the Board when¬ ever necessary for the public service, subject to subsequent approval by the Board. Journeys to be performed by the employees of the Board maybe ordered by any member of the Board. When the emer¬ gency requiring the journey does not permit of obtaining an order previous to making the journey, the employee must certify upon the expense voucher “that urgent public duty required the journey to be performed without previous orders.” All journeys^ must be finally approved by the Board, and the vouchers must be accompanied by that approval. Through tickets must be obtained for all travel when the journeys are to be continuous, and transportation requests used when¬ ever practicable. Par. 5. All vouchers in payment for articles or materials shall have attached to them the original receipted bill or bills for the articles or materials, said Dills to be furnished by the person, firm, or corpora¬ tion supplying them. Par. 6. The following reports and returns shall be made on the forms of the Corps of Engineers, United States Army, or other form authorized by the War Department, and transmitted by the disburs¬ ing officer to the Secretary of War within ten days after the expira¬ tion of the month or quarter to which they relate, except Form 13, which will be rendered within twenty days after the expiration of the quarter, Form 17, which will be rendered at the close of business hours of each week, and Form 14, which will be rendered at the close of the month, viz: (1) Weekly Money Statement, Form 17. (2) Monthly Money Statement, Form 14. (3) Monthly Account Current, in duplicate, Form 3. (4) Monthly Abstract of Disbursements, in duplicate (Forms 4, 5, or 5 a ), with proper vouchers (Forms 0, 7, 8, 8 a , 9, 9 a , 10, 11, 12, 22, 23). (5) Quarterly Return of Public Property Received and Expended, Forms 13 1 , 13 2 , 13 3 , and 13 a . Par 7. All officers, agents, or other persons who are charged with the safe-keeping, transfer, or disbursement of the public moneys shall keep an accurate entry of each payment or transfer, and shall 330 DEEP WATERWAYS. render distinct accounts of the application thereof according to the appropriation under which the moneys may have been- advanced to them. Every officer or agent who, having received public money which lie is not authorized to retain as salary, pay, or emolument, fails to render his accounts for the same, shall be deemed guilty of embezzlement, and shall be fined in a sum equal to the amount of the money embezzled, and shall be imprisoned not less than six months nor more than ten years. (R. S., 3623, 3643, 5491.) Par. 8. An error made in an account must be corrected in the next account current of the disbursing officer after he is informed of the error, and reference will be made therein to the particular voucher in which the error occurred. Par. 9. Funds received from overpayments previously made will be entered on the account current in the proper column. The entries should show by whom and to whom the overpayments were made and on what account, and refer to the voucher and abstract. Par. 10. Whenever money is refunded to the Treasury, the name of the person refunding and the purpose for which it was done must be stated. Par. 11. The disbursing officer will promptly, at the close of busi¬ ness at the end of each month, and also on Saturday of each week, when that month does not end on Saturday, transmit to the War Department a statement showing explicitly where his funds are depos¬ ited. He will include in the sums claimed to be on deposit only such funds as have been officially credited to him, and of which credit he has been duly informed by the depositary. All public funds on deposit must be to the official credit of the disbursing officer. Par. 12. Disbursing officers have no authority to insure public money or property. They are not authorized to settle with attorneys of claimants, heirs, executors, or administrators, except by instruc¬ tions from the War Department upon accounts duly audited and cer¬ tified by the proper accounting officers of the Treasury. (Sec. 3477, Rev. Stat.) Par. 13. All payments by the disbursing officer must be made in checks upon his deposit in a public depositary, in lawful coin, or in United States notes. Payments, whenever possible, shall be made within the quarter in which the liability was incurred. Par. 14. When funds are available, the disbursing officer shall pay cash and not open an account. Par. 15. When the disbursing officer draws checks in payment of accounts, on funds placed to his credit with an assistant treasurer or other depositary of tHe United States, he will note upon the receipt taken for such payment the number, date, and amount of the check given in payment and designate the assistant treasurer or depositary upon whom it is drawn; and when an account is paid in part by cur¬ rency, the amount of the same will be stated. The same rule will be observed in regard to invoices of funds transferred. Mutilated checks DEEP WATERWAYS. 331 shall at once be forwarded to the depositary to which they pertain, and a record made on the stubs of the check book of the date of t rans¬ mission. Par. 16. In all cases of contracts for the performance of any service or the delivery of articles of any description for the use of the United States, payment shall not exceed the value of the service rendered or of the articles delivered previous to the payment. Par. 17. Persons employed in the service of the Board may be allowed actual expenses for travel on duty under orders. For all persons except the members of the Board and its secretary, and mili¬ tary and naval officers, these expenses will be limited as follows: (1) Cost of transportation (including parlor-car fare) over the shortest usually traveled route. (2) Cost of transfers to and from railroad stations, not exceeding 50 cents for each transfer. (3) Cost of one double berth in a sleeping car, or customary state¬ room accomodation on boats and steamers when extra charge is made therefor. (4) Cost of meals, not exceeding 13 per day while en route, when meals are not included in the transportation fare paid, and not exceed¬ ing $3 per day for meals and lodgings during necessary delay en route. (5) Cost of meals and lodgings, not exceeding 13 per day, while on duty at places designated in the orders for the performance of tem¬ porary duty. When such expenses are less than 825 under any given order, the certificate of the person performing the journey, as set forth in Form 9, will be sufficient; when the expenses are 825 or over, the affidavit of the person will be required, as follows: I.-, solemnly swear that the above account is correct and just, and that I necessarily performed the above journey on public business, at the time specified, with all practicable dispatch, by the shortest usually traveled route, in the customary reasonable manner, and under the order hereto annexed; that I did not travel on any conveyance belonging to or chartered by the United States: that the above expenses were actually and necessarily incurred and paid by me in per¬ forming said journey. Sworn to before me and subscribed in my presence this-day of-. The legal fee for the same may be included in the account. In all cases there should be added the certificate of a member of the Board approving the account and certifying that the travel was performed under the orders specified in the voucher. Par. 18. No member or employee of the Board shall accept vol¬ untary service for the Government, or employ personal service in excess of that authorized by law, except in cases of sudden emergency, involving the loss of human life or the destruction of property. 332 DEEP WATERWAYS. Par. ID. The time of all employees noted or kept as hours shall be reduced to “days of eight hours each,” and the rate of pay shall be noted and designated as the rate “per day of eight hours each,” and so shown upon tlie vouchers or pay rolls, as the case may be. Time employed should not be shown upon the pay rolls or vouchers in “hours,” and the rate of pay should not be designated as “per hour.” When pay rolls or vouchers show persons to have been employed and paid by the day, there should be added to the certificate 1 he state¬ ment that the days for which payment has been made were “days of eight hours each.” Par. 20. When it is intended to have any work done, service pro¬ cured, or purchase made, by contract or written agreement, all the papers in the case must previously be submitted to the Board for its approval. Par. 21. Where il is clearly for the benefit of the United States that a contract should be extended, power is hereby delegated to the Board to grant such extension, but extensions of contracts or written agree¬ ments are not in any case to be made until the approval of the Board has been obtained. Par. 22. Vouchers for the disbursement of money will correctly specify the quantity and price of each article bought, the name and business place of the person from whom it is procured, and the date and manner of purchase. When the vouchers are for services ren¬ dered, they will state the nature and period of service, rate of pay per day or month, etc. Par. 23. Original vouchers must accompany the accounts. Copies can not be admitted unless accompanied by satisfactory evidence of the loss or destruction of the originals, or that their retention is indis¬ pensable to the performance of duty. Par. 24. When originals can not be furnished, copies duly certified as true by a disinterested officer of the Government maybe accepted. Par. 25. When vouchers are not sent with the account to which they belong, an explanation must be made as to why they were not pro¬ duced with and included in the proper account. Par. 26. All employees of the Board are forbidden to give or take re¬ ceipts in blank for public money or property; in all cases the voucher will be made out in full and the exact amount of money or quantity of property, in words, will be written out in the receipt before it is signed. When vouchers are sent by mail for signature, the date in the receipt will be left blank, and the check in payment will not be drawn until the vouchers are received back properly signed, when the date of the check will be added to the receipt. Par. 27. When a signature is not written by the hand of the party it must be witnessed. Par. 28. When an account is presented in person by an individual who is not known to the disbursing officer, the latter will require such DEEP WATERWAYS. 333 evidence of identity as will secure the Government as well as himself against loss. Par. 29. Upon all vouchers for purchases or for services other than personal, the method of purchase, or of procuring the service, must be shown, and if “by written agreement,” a copy of the agreement must accompany the voucher. Par. 30. Property returns will be rendered quarterly. After a com¬ plete return has been furnished, if there have been but few changes during a subsequent quarter, it will be sufficient for three quarters to state these changes. Par. 31. At the time when vouchers are submitted with the accounts, if any of the articles purchased as noted thereon have been “expended and applied to the purpose for which purchased,” a certificate to that effect may be made by the disbursing officer on the voucher, and when so made the article referred to need not be taken up on the property returns. Par. 32. The certificate of an authorized agent of the Board of the delivery of supplies or materials, or of the performance of service, shall constitute authority to the disbursing officer for payment there¬ for; but should the facts not be as set forth in said certificate, the amount shall be charged to the certifying agent and be credited to the disbursing officer in the settlement of his accounts. Par. 33. No member or employee of the Board shall be pecuniarily interested, either directly or indirectly, in the employment of any person, or the purchase or supply of any article or materials for the service of the Board. Par. 34. All persons having charge of property pertaining to the Board shall be held responsible for its preservation and safe-keeping, and shall give receipts to the disbursing officer for the same. They are required to take all proper measures to protect it from loss, damage, or waste; and when such property is not satisfactorily accounted for, the person to whom it was intrusted shall be charged with its value. Par. 35. In all cases not covered by these regulations, the rules for money and property accountability established in the United States Army Regulations of 1895, so far as applicable to the service of the Board, shall be observed. Par. 36. The instructions to United States disbursing officers contained in circular from the Treasury Department dated August 14, 1897, and published in General Orders, No. 53, War Department, Adjutant-General’s Office, August 25, 1897, must be strictly complied with. R. A. Alger, Secretary of War. 334 DEEP WATERWAYS. % Appendix No. 9.—Part 2. INSTRUCTIONS TO FIELD PARTIES. The following instructions are for the general guidance of field par¬ ties, and will be supplemented from time to time by such special instructions from the Board as may be found necessary as the work progresses. The immediate object of the work is to obtain data for the develop¬ ment of plans, estimation of cost, and the construction of maps and profiles of the proposed ship canal from the Great Lakes to the Atlantic. Considerable discretion will be allowed chiefs of parties as to meth¬ ods by which information sought shall be obtained and the distribu¬ tion of work among members of the party, but the records of all sur¬ veys and examinations must be so full and complete that they may be easily and correctly interpreted by computers in the office of the Board without assistance from the engineers who make the observations. The work must lie such that when reduced and plotted a paper location may be made of the proposed ship canal, a correct estimate made of the amount and kind of material which will need to be exea- uatecl, and the size and character of all necessary structures projected. ORGANIZATION. The personnel of the parties herein outlined is based on normal conditions of country, a full-sized field party consisting of a base¬ line party, level party, sounding party (where needed), two or three boring parties, three stadia parties, and draftsman and computers sufficient to keep up reduction and plotting of field notes as close as practicable with the field work. All parties are to work with the general purpose in view that the combined results of the different divisions should be sufficient to furnish the necessary data on which to base designs, plans, and estimates of proposed ship canal route, and in order that the best progress for amount of work done shall be obtained it is essential that the parties should not duplicate their own work or that of others, except where necessary for correcting errors. Each of the stadia parties should consist of 1 instrument man, 1 recorder, 2 rodmen, and 1 stadia man for open country, and such axmen as may be needed to cut lines where topography of wooded country is to be determined. In organizing new parties, recorders and rodsmen, where capable, should always be promoted in preference to appointing new men. METHODS OF WORK. Reconnaissance .—A careful reconnoissance should 1>e made of the proposed line of survey by the assistant engineer in charge in advance of all field work, to determine tlie limits of survey and the probable DEEP WATERWAYS. 335 location of right of way along which it will be necessary to determine contours carefully, and make borings sufficient to establish the eleva¬ tion of rock surface, where it exists within limits of probable depth of cut. Topography .—The topography will be based on continuous transit and level lines between terminals of any section of the proposed canal, and be determined by stadia readings from the stations and bench marks of said lines to all characteristic points. Topography will be taken with the view of correctly plotting con¬ tours for 2-foot intervals along right of ways, but outside of such limit slight irregularities of the surface are of no importance, and only elevations of characteristic points should be taken by the stadia men, and sketches made between. Base line and level observations, where possible, should be kept in the advance of topographical work, and the elevations and location of transit stations, with azimuth of line, furnished to the stadia parties. Transit tines .—The distances between transit points of main line should be measured with a steel tape of which the length has been determined by actual measurement of a base line of known length, or by comparison with some standard measure already determined, and such distances checked by intersections on known points or by stadia readings of the topographical parties. The true azimuth of the survey line should be determined at inter¬ vals of about 5 miles along the route, by observation on elongation of Polaris, for the purpose of correcting the transit line for errors of observation and for change in azimuth due to convergence of merid¬ ians. To determine the azimuth of a line from observations of Polaris at elongation, a set of readings should be made between the star and reference mark, and the telescope should then be reversed and a second set made. The difference between the observed azimuth at any station and that brought forward from previous station by transit line is always equal to the algebraic sum of the errors of observations and the change due to the convergence of meridians. The latter of these two correc¬ tions should be distributed back through the transit stations propor¬ tionately to the respective westings, and the former proportionately to the number of transit stations between azimuth stations. In order, therefore, that all corrections be properly applied, there should be an azimuth determination of transit line in advance of all work plotted on office map. Permanent marks as reference points will be established by the transit and level parties at frequent intervals along the line, such that any paper location made upon the maps may be easily and cor¬ rectly staked out on the ground surveyed. Levels .—The elevations of all bench marks of level lines will be based, wherever possible, on the elevation of bench marks established by precise level parties of the United States Lake Survey, and, unless I 336 DEEP WATERWAYS. checked by readings on points of known elevations, should be deter¬ mined by duplicate lines run in opposite directions. Duplicate level lines must agree within the amount shown by the formula 1( |(| y distance between bench marks in miles. Frequent readings should be made for connecting with the water surfaces, wherever the lines run are in the immediate vicinity of lakes, rivers, or canals. Methods of stadia work .—Stadia parties should be furnished with corrected azimuths of transit line, and all stadia circuits should be carried forward from such initial directions and checked on other lines of known azimuth wherever possible. Where this can not be done, the magnetic azimuth of lines should be read occasionally, and readings made to prominent points on the line of survey, the location of which may be independently determined. In stadia circuits, stakes need be set only at instrument stations, and at points where side lines are to be started for closures made with other circuits. Where areas of timber land and orchards interfere with lines of sights, contours may be developed sufficiently well by determining cross profiles at intervals by stadia readings along compass line. Xo matter which member of the party may make the observations, it should be distinctly understood that he should make the complete readings, both as to direction and distance, and no other person be allowed near the instrument. The instrument man and recorder should make the observations, keep the records, and make sketches, except where the conditions are such that only two rod men are needed, in which case the extra rod- man may be detailed to sketch or make reconnoissance for advance work. In addition to readings to develop contours, all buildings, fences, roads, creeks, culverts, bridges, timber lands, etc., should be located, and where structures are on line of probable final location, the dimensions and character should be noted. 'Highways, fences, rail¬ roads, streams, and embankments should be located by readings to points where directions change, and a sketch made showing connec¬ tion between consecutive points. Where rivers are to be sounded, all range stakes should be set in advance by either the transit or stadia field parties. The width, depth, and cross-section of streams for high, low, and medium stages should be determined wherever possible. The limit of work for survey should be determined by reducing side shots in the field, and not by running special circuits along lines of elevation required. 'Fiie instrument man should have a system of signals for his party, such that he may direct a rodman in any direction, and to make any kind of location desired. In all cases the work should be so planned that stations need not be DEEP WATERWAYS. 337 reoccupied, except to. start newlines or to correct errors, and that work done by other divisions of the main party will not be duplicated. Where possible to do so, the stadia party should leave stakes for boring party to connect with when holes are drilled, and thus save labor of locating holes after work is done. Stadia parties in the field need do only sufficient plotting to insure that records are free from error and that sufficient locations have been made to develop the topography. Borings .—Borings will be made at such intervals and depths along the lines that the location, amount, and kind of material to be exca¬ vated may be accurately estimated for a ship canal of 30 feet depth located within the limits of strip surveyed. Reduction of notes .—The field party should keep their notes re¬ duced up to date and ready for checking by the computers in office of engineer in charge. All notes, records, and sketches should be so clear and complete that no misinterpretation can occur. PLOTTING OF FIELD NOTES. In the office of assistant engineer in charge of party there should be a draftsman and sufficient computers to reduce and plot the field notes as fast as sent in by the stadia parties. The computers should be men capable of reducing field notes and assisting the draftsman in plotting preliminary lines and circuits on maps. Maps should be on a scale of 1: 2,500 or 1: 5,000, depending on the flatness of the country, and should be drawn on mounted paper 28 by 40.5 inches in size, with working limits of 25 by 37 inches, and laid out so as to have a wide margin at left-hand end of completed map. All lines, fences, roads, etc., should be plotted with india ink and all contours with burnt sienna. Transit lines and important stadia circuits should be plotted by coordinates, and side lines and readings by direction and distance. For small vertical angles no reduction for distance is necessary for side shots, and for main circuits the correction may be taken direct from a table prepared for that purpose. Transit line distances should be corrected for tape error and tem¬ perature before plotting. Only such lettering need be put on maps as necessary for a correct interpretation, and no azimuth of courses except line indicating the true meridian through some azimuth or transit station. All work to be plain and accurate, and of a style to be rapidly done. FINISHING FIELD SHEETS AND REDUCTION. The finishing of field maps not provided for in these instructions and the reduction for publication will be done at the office in Detroit. All maps for publication are to be reduced to a scale of 1: 20,000 by means of a pantograph, such reduced sheets to show the amount of H. Doc. 149-22 338 DEEP WATERWAYS. detail to be put on tracings, except that swamp and timber areas will be indicated with topographical symbols, and where slopes are steep hacliures will be used instead of contours. Where wood and swamp areas are large, only the limits need be indicated by symbols. The line of proposed waterway will be indicated by two parallel lines spaced 250 feet apart, except in open rivers and lakes, where actual width estimated will be shown. Location and depth of borings will be indicated by showing eleva¬ tion of bottom of hole, and if rock be found, such number will be followed by the letter R. The projected line will be marked every 1,000 feet and at every mile on the map, and a profile of the line marked to correspond with that on map will be placed at bottom of the sheet. REPORTS. A weekly report should be made at the end of each calendar week, giving briefly the amount of work done and any other information of interest or importance. A monthly report should be made promptly at the close of each month, giving a detailed statement of work done, condition of work or works in progress at that date, and other important information or suggestions. Appendix No. 10. NIAGARA ROUTE AND CHAMPLAIN ROUTE, HUDSON RIVER DIVISION—FLOOD MEASUREMENTS MOHAWK RIVER—LEVELS FROM HUDSON RIVER TO LAKE ONTARIO. Detroit, Mich., January 15 , 1900 . Gentlemen : I have the honor to report herewith on the surveys and examinations intrusted to me in connection with the proposed deep waterway from the lakes to the Atlantic. The report is arranged under the following headings: 1. A waterway from Lake Erie to Lake Ontario of 21 feet and 30 feet depths. 2. A waterway from Lake Champlain to the Hudson River at Troy of 21 feet and 30 feet depths. 3. Flood-discharge measurements of the Upper Mohawk River and other streams. 4. Report on results of two lines of levels run from the Greenbush bench mark to Lake Ontario. The above work, excepting item 4, was executed by one organiza¬ tion, made up in Chicago between August 17 and September 12, 1897, with such additions, from time to time, as the necessities of the work required. After purchasing and testing instruments, the party left DEEP WATERWAYS. 339 Chicago September 12, 1897, and arrived at Tonawanda, N. Y., the next day, and immediately began the surveys for the Tonawamla- Olcott route. The order in which the other work was done is fully explained in the reports covering it. I wish to acknowledge receipt of valuable information relating to the hydrography of the Hudson River, from the Glens Falls Paper Company, of Fort Edward, N. V.; the Duncan Company, of Mechan- icville, N. Y.; the Hudson River Power Transmission Company, of Meehanicville, N. Y., an. Nearly all of these bor¬ ings are plotted on the published maps and are designated thus ©. The figure preceding this mark indicates the elevation of the bottom of the hole, and, when followed by the letter (R), indicates that rock was struck at that elevation, the object being to determine the eleva¬ tion of the bed rock, within the limits of any probable excavation, and the character of the material overlying it. Samples were taken of the various kinds of material penetrated, preserved in bottles and properly labeled for identification. The apparatus used in making the borings is shown in fig. 1. A 2-inch rope passed from the drum over a sheave at the top of the der¬ rick and was attached to the swivel in the top of the l^-inch drill rods, which worked inside of 24-inch flush-joint casing. Both the casing and the rods are those used in ordinary diamond-drill work. With a hand pump water was forced through the hollow drill rods and passed out through four small holes in the cross bit attached to the lower end, and then up to the surface of the ground, through the annular space between the rods and casing, bringing with it all loosened material. The material was loosened, depending on its char¬ acter, by churning the rods or by turning them with pipe tongs. The casing was worked down by turning with pipe tongs. Bowlders were broken up with dynamite when encountered. It was necessary to haul water in tanks with teams for nearly all the holes, which added materially to the cost of the work. The force for making borings consisted of 1 superintendent of bor¬ ings and 3 boring crews, each composed of 1 foreman, 3 laborers, and 1 teamster, with team to haul water and move the machines from hole to hole. 342 DEEP WATERWAYS. The surveys were begun September 13, 1897, and completed about March 1, 1898, covering a period of five and one-half months. Owing to delays in getting the apparatus for making borings, this part of the work was not begun until November 1, 1897, and completed April 16, 1898, covering a period of five and one-half months. Using the base line for the standard, and as correct, the average error of closure in horizontal distance for stadia circuits was 1 in 1,250, and the average error of closure in elevation was 0.177 foot. . The survey, including the location of all borings, was plotted on the standard size sheets, 28 inches by 40.5 inches, to a scale of 1: 5000, except in the vicinity of the gulf on the Tonawanda-Olcott route, where a scale of 1: 2500 was used for three sheets. These maps were entirely completed by this party for the survey at Buffalo, for the Lasalle-Lewiston route, and about one-third of the Tonawanda-Olcott route, the other two-thirds of this route being mapped in the main office at Detroit. The total length of line surveyed was 39 miles, as measured along the center line of the located canal, being divided as follows: Miles. Buffalo, regulating works to foot of Squaw Island___ 2.7 Tonawanda-Olcott route, including harber at Olcott_ __ _ 25.7 Lasalle-Lewiston route_ ______ 10.6 Total___ _____39.0 The total cost of borings, which includes plant, repairs, labor, and all other expenses connected therewith, except surveys for location of holes, was $6,582.65, or $168.78 per mile. Four hundred and four borings were put down on the two lines, with a total of 9,624 linear feet, making a cost of $0,684 per linear foot. The plant for both bor¬ ings and surveys was considered sunk when the work was completed, and its entire cost proportioned against the routes on which it was used. The salary of the assistant engineer was charged to surveys, and no part of if to borings, the object being to have the borings represent actual cost for such work, leaving out the cost of general supervision and the survey for locating them. The cost of surveys, which includes instruments, tools, labor, and all other expenses connected with the field survey, mapping, and final plans and estimates, was $17,049.30, or $437.16 per mile, making a total cost of the borings, surveys, and estimates $605.94 per mile. This cost includes making the final maps as published, but does not include plotting that part of the Tonawanda-Olcott route done in the Detroit office. It should be stated that this work was begun with a party energetic though inexperienced in this class of work, and that most of the field work was done during the winter when the weather and the short days were against rapid progress. DEEP WATERWAYS. 343 LASALLE-LEWISTON ROUTE. LOCATION AND CHARACTER OF MATERIAL TO BE EXCAVATED. After the contour maps were completed the location of the canal was laid down upon them by the Board, it being the same for the 21 and 30 foot channels. For the estimate for regulated lake surface it was assumed that the works for regulating the level of Lake Erie would be built at the foot of the lake, as shown on plates 84 and 14, and discussed in the special report of the Board on the Regulation of the Level of Lake Erie, Appendix No. 0. The channel 600 feet wide originates at deep water in the lake and passes to the right of the regulating works through the Erie Basin to the head of Black Rock Harbor, where it narrows up to the regular canal prism and continues to the foot of Squaw Island, where it widens to (300 feet. At the head of the island lock No. 1, with a lift at low water of 8 feet, is located. From the foot of the island the 600-foot channel continues down the Niagara River, passing to the right of Strawberry Island and Grand Island to Tonawanda (at which point the Tonawanda-Olcott route leaves the Niagara River), and thence to the head of Cayuga Island, where the channel contracts to the regular canal prism on leaving the river, and continues in a north¬ westerly direction through the village of Lasalle and to above the escarpment at Lewiston. It then descends, by a double flight of six locks and a double flight, of two locks, to the Niagara River. There is a level 1,730 feet long between the two flights of locks. The total lift of the two flights at Lewiston at low water is 318.8 feet. From Lewiston to Lake Ontario the river is from 40 to 60 feet deep, and does not require any improvement for a channel of either 21 or 30 feet depth. In the lake, off the mouth of the river, there is a shoal over which vessels drawing 21 feet can pass, but to make a channel 30 feet deep and 1,000 feet wide would require the excavation of about 654,000 cubic yards of sand and gravel. MATERIAL. The material to be excavated on this route is earth, sand, gravel, clay, liardpan, limestone, and shale, but the classification for the pur¬ poses of making estimates will consist of solid rock, which includes both the limestone and shale; liardpan and earth, which includes silt, loam, clay, sand, gravel, and all other material which can be easily excavated with steam shovels. Extending from shore to shore and from lock No. 1 to deep water in Lake Erie there is a ridge of hard limestone, known as corniferous limestone, almost entirely bare except in Black Rock I larbor, where it is covered with sand, gravel, and clay. The rock disappears below grade just below lock No. 1 and shows up again opposite Strawberry 344 DEEP WATERWAYS. Island, for a distance of about 1 mile, and again opposite Frog Island, for a distance of 500 feet. At this point it disappears below grade again, and then shows up opposite Tonawanda Island. From Tona- wanda to the Niagara River at Lewiston the rock is to be found at or above the grade of the channel. The corniferous limestone runs out soon after leaving Lake Erie. From this point the rock outcrop is composed of Salina shales to the vicinity of Lasalle, and the Niagara limestone is found from there to the escarpment. To determine the character of the rock, live diamond-drill borings were put down at various points north of Lasalle. Boring No. 1 was about one-half mile north of Lasalle and 400 feet to the right of sta¬ tion 910. Light-gray Niagara limestone was struck at elevation 550 feet above sea level and the boring penetrated it 20 feet to elevation 524. It is a hard rock in thin broken strata, but near the bottom a stratum 28 inches thick was struck. The grade of the channel from Lasalle to lock No. 2 is 533.5 feet. Boring No. 2 is located near Gill Creek and about 740 feet to the right of station 1090. Rock was struck at elevation 579 and pen¬ etrated 56 feet to elevation 523. It was limestone in thin broken layers and very dark in color for 34 feet, but the other 22 feet was lighter in color and quite solid. Boring No. 3 is just above the escarpment and 700 feet to the right of station 1220 of the channel. It is also east of lock No. 2, and just south of Lewiston. The stratification was: 595 to 557. Hard light gray Niagara limestone, quite solid and grad¬ ually merging into shale at the bottom. 557 to 491. Niagara shale, firm but easily broken into short pieces. The last 20 feet contains many fossil shells. At the bottom it merges gradually into Clinton limestone. 491 to 460. Hard Clinton limestone of blue-gray color. 460 to 417. Sandstone intermixed with thin layers of red shale. It varies in color from gray, mottled, and red, and is known as Medina sandstone. Boring No. 4 is located on the side of the escarpment, and is intended as a continuation of boring No. 3. It is about 1,200 feet to the right of station 1240, and opposite lock No. 5. The stratification was: 389 to 367. Very hard light gray sandstone, known as quartzose sand¬ stone. 367 to 330. Firm, red shale mixed with bands of green shale. The line of demarcation between the sandstone and red shale is well defined. Boring No. 5 is north of Lewiston, and about 2,800 feet to the right of station 1300, where the channel enters the Niagara River. The strati¬ fication is: 335 t o 239. Soft, red shale, mixed with bands of green shale broken up. Below elevation 264 there are occasional bands of hard shale. 239 to 209. Firm, red shale in thin horizontal layers. DEEP WATERWAYS. 845 All the rock penetrated by the above five borings is in flat layers, with a dip to the south of about 36 feet to the mile. Practically all the rock to be excavated between Tonawanda and lock No. 2 at Lewiston is limestone, but the excavation for the two flights of locks will include all the different rocks found in borings Nos. 3,4, and 5. The general character of these rocks is pretty well known and need not be further described or considered except to determine the form of the side slopes in the deep cut at the escarpment, and to decide as to their suitable¬ ness for foundations. As to the latter point, it may be said that all of the rocks will afford good foundations for locks at the elevation required. The limestones and sandstones are solid and firm and show little or no signs of dis¬ integration when exposed to the action of the elements for centuries, but all of the shales when so exposed disintegrate to a greater or less extent. According to geologists the gorge of the Niagara River at Lewiston has been in existence from sixty to one hundred centuries and it has not widened to any great extent during that time, although the exposed faces of the shales show marked signs of disintegration. Nearly vertical faces of Niagara shale are exposed at “The Gulf,” west of Lockport and near diamond drill borings No. 3, which show decided indications of disintegration under t he action of the elements, but it has stood for centuries without receding more than a few feet. The Erie Canal at Lockport is cut through a ridge of Niagara shale, and the northwest bank of this cut stands nearly vertical and shows but little signs of disintegration, though it has been exposed to the elements over fifty years. During the winters of 1896-97 and 1897-98 the Erie Canal was deepened above Lockport, much of the excavated material being Niagara shale, which, after being broken up and exposed to the action of the weather in the spoil banks, disintegrated very rapidly. This same shale was solid and firm where it formed the walls of the canal prism below the ordinary water level. The shale excavated from the wheel pits at Niagara Falls disintegrated very rapidly when exposed to the action of the weather. The conclusion reached, after examining the shales found above the escarpment on both routes, is that they are perfectly stable when under water and disintegrate slowly when left in the natural bed, with faces at right angles to it, exposed to the action of the elements, but when broken up and placed in spoil banks they rapidly deteriorate. The shale found below the escarpment at Lewiston is of a red color and soft near the top, but becomes much harder near the bottom of the proposed excavation. It crumbles rapidly when broken up and exposed to the elements and finally becomes a red clay. It does not disintegrate under water and affords good foundations. Taking up the material overlying the bed rock and within the limits of the excavation for a channel of 30 feet depth, it is found to be silt, sand, gravel, bowlders, and clay, and may be classified under the one I)KEP WATERWAYS. 346 general head of drift. At Squaw Island it is generally sand and gravel, with a few feet of silt on top, and near the head of the island is found to be gravel and bowlders cemented together, forming a hard mass. From Squaw Island to the head of Cayuga Island, where the channel leaves t he river, sand and gravel are found, which can easily be dredged. From this point to the escarpment a firm yellow clay is found on top, a soft, reddish clay almost entirely free from grit and varying in thickness from 2 to 10 feet is found under this from the river to about 1 mile north of Lasalle. Under this soft clay and extending north to Lewiston is found a mixture of red clay and gravel, which in some places is a perfect hardpan and in other places it is not much harder than a good firm clay. Below the escarpment the drift is sand, gravel, clay, and bowlders—the bowlders being generally limestone, angular in shape, and are probably fragments broken off from the cliff above. As a rule a layer of sand or gravel from 6 to 12 inches thick is found immediately above the rock and below the hardpan or clay from Lasalle to Lewiston. GRADES. In connection with the discharge measurements of the Niagara River at Buffalo, which were taken from October, 1897, to December, 1897, by E. E. Haskell, United States assistant engineer, several gauges where established from Lake Erie to the foot of Squaw Island and read to measure the slope of the river at the several points. To deter¬ mine the slopes farther down the river, I established gauges at Ger¬ mania Park dock opposite Strawberry Island, Rattlesnake Island, the bridge across the river to Tonawanda Island, near the head of Cayuga Island, and Schlosser’s dock, and had them read several times daily from October 23 to December 16. At times of high or low water, caused by strong winds on Lake Erie, each one of these gauges was read every ten minutes during the day. In this way the low water slope of the river under present conditions was determined from point to point. This slope is shown for a low stage on plate 90. For an elevation of the lake 571.3, the corresponding elevation of the surface of the water in the river at the various points given would be as follows: Foot of Squaw Island._.... . 566.4 Germania Park. . .566 Rattlesnake Island..... 565.5 Head of Tonawanda Island..... 565.13 Head of Cayuga Island............ 563. 63 Schlossers Dock........ 563.08 If the lake be taken as 570.5, then the head of Tonawanda Island would be 564.6 and the head of Cayuga Island 563.17. For the purposes of making the estimates, the elevation at Tona¬ wanda was taken as 565 and at Cayuga Island as 563.5, which, under present conditions, corresponds with elevation 571 of Lake Erie. But DEEP WATERWAYS. 347 it is certain that the slope, especially from Tonawanda to Lasalle, will be reduced when the channel 600 feet wide and 30 feet deep is con¬ structed. How much this reduction will be cannot be predetermined on present data, but the water level at Cayuga Island can be regu¬ lated within the limits required by depositing the necessary amount of the excavated rock in the river below the head of the island, thereby reducing the cross section of the river to the required area and shape. The elevations of the water surface from deep water in the lake to the head of lock No. 1, on which estimates are based, are for the regulated level 574.5 and for standard low water 571.4. The lock will have a lift at low water, after regulating works are built, of 8 feet, making the grade immediately below 566.5, which corresponds to elevation 571 of the unregulated lake. The slope is considered as uniform from this point to Tonawanda and to Cayuga Island. From Cayuga Island to lock No. 2 the water surface is 563.5 and level throughout; from Tonawanda to lock No. 2 on the Tonawanda-Olcott route the water surface is 565 and level throughout. The levels below locks No. 2 on each route depend on the lift of the locks, which will be further considered under the head of locks. For the 21-foot channel and 30-foot channel the elevation of the bottom of the channel is 21 feet and 30 feet, respectively, below the low-water surface, following the slope in the river and being level through the canal sections. Retaining walls and slope walls are provided in accordance with the standard plans and designs, except for the retaining wall between the canal and river from the regulating works to lock No. 1, where it is made of concrete masonry, with a top width of 12 feet and the base not less than two-thirds the height. The locks are of the standard design adopted by the Board, with such variations as are necessary to lit local conditions at each place. Their size for the 30-foot channel is 740 feet by 80 feet for single locks, but for double locks one is 60 feet wide and the other 80 feet wide, both being 740 feet long. For the 21-foot channel all locks are 600 feet by 60 feet. The filling and emptying culverts are located in the lock walls. Miter gates are used in all cases. The details of the lock construction are given in Appendix No. 1. Lock No. 1, at Buffalo, is a double lock of 8 feet lift and founded on solid rock. Its various parts conform to the standard design, except that the wall on the river side is made thicker than the one on the land side on account of having water on both sides of the wall. Water power for driving the operating machinery can be developed above the lock. Above lock No. 2, at Lewiston, the elevation of low water is 563.5, as heretofore explained. Below lock No. 9 the standard low water of the Niagara River is 244.7, making a total fall of 318.8 feet to be overcome by locks. 348 DEEP WATERWAYS. From a casual inspection of the profile at this point it is evident that single locks with a basin between would involve an unnecessary amount of excavation. It is also evident that a single flight of locks would seriously delay the traffic when ships were going in opposite directions. To provide the greatest capacity at the minimum cost a double flight of locks should be constructed. The arrangement finally decided upon is shown on plate 15, and the design is shown on plate 69. For the 30-foot channel estimates are made on a flight of 6 double locks of 40 feet lift each and a flight of 2 double locks with a lift of 39.4 feet each; also for a flight of 8 double locks of 29 feet each and a flight of 3 double locks with lifts of 28.93 feet each. These esti¬ mates show that the locks of 40 feet lift can be built $4,600,000 cheaper than those of 30 feet lift. In both cases there is a level, with connect¬ ing side basin, to reduce the fluctuations of the water surface when putting in or taking out a lockful of water between the two flights of locks. The channel throughout the length of this level and also for 2,500 feet above the head of lock No. 2 is excavated 2 feet below grade, so as to insure a full depth of channel when the locks are being filled. For the 21-foot channel estimates were made only for locks of 40 feet lift, as it is quite evident that those of 30 feet lift would cost more and would not be so efficient. Water power for driving the operating machinery and pumping out the locks may be developed cheaply by making an open cut from the channel at a point above lock No. 2 to the top of the bluff near the Suspension Bridge, and then carrying the water down the side of the bluff to the wheel pit situated between the foot of the bluff and the Niagara River. The tail race can be made by tunneling from the bottom of the wheel pit to the river. This arrangement would give a head of 300 feet. It can also be developed by carrying the water in pipes along the west wall of the locks from the head of lock No. 2 to opposite lock No. 7, where a wheel pit could be made and the tail water pass out through the tunnel to the river. By this arrangement a head of 270 feet can be obtained. The power may then be distributed by elec¬ tricity or by compressed air to the points required for the two flights of locks. The power will be needed for lighting, operating the gates and valves, and pumping out lock No. 9. It is evident that locks Nos. 2, 3, 4, 5, and 6 can each be drained into the next lower lock, and a tunnel from the bottom of lock No. 7 to the river will drain this lock and also provide the tailrace for the water used in developing the power. To provide against unequal leakage in the several locks of the flight a pipe 34 feet in diameter is carried in the middle wall from the head to the foot of the flight. Valves are provided to supply water to DEEP WATERWAYS. 349 each lock of the flight. The level basin between the two flights is provided with a spillway to carry off surplus water which is not needed in locking through the second flight. It is evident that lock No. 8 can be drained into lock No. 0, but on account of the river being higher than the bottom of the lock, it will be necessary to provide a pumping plant to empty lock No. 0. The Niagara River from Lewiston to its mouth affords a magnificent harbor 7 miles long and from 40 to 60 feet deep, which already exists and will cost nothing to maintain. STREAMS CROSSED. The channel crosses Cayuga Creek, Gill Creek, and Fish Creek, each of which can be taken into the canal. There is very little or no flow in Cayuga Creek at low stages. Its depth is from 6 to 8 feet and is due to the backwater from Niagara River. The present water surface is therefore about the same as the proposed water surface in the canal. Starting at a point 700 feet to the right of station 885, the prism of the creek can be enlarged and made to enter the canal at station 885 with a width of 300 feet and 10 feet deep, which would give a low-water cross section of 3,000 square feet and bring the flood waters into the channel at a velocity of about one- half foot per second. This would not erode the banks nor interfere with navigation. It is also proposed to put a sill across the creek 3 feet above its bed, so that any sand and silt washed down by the stream would be deposited before entering the canal. Gill Creek at station 1092 is a small stream, with no flow except at times of rains or melting snows. Its bed is composed of earth and is about 24 feet above the water surface of the canal, but solid rock is found about 5 feet above this water surface. It is proposed to exca¬ vate the earth 100 feet wide back about 200 feet and then build a dam or retaining wall up to the bed of the stream. The waters will pour over this wall onto the rock below and then into the channel. Fish Creek is crossed at station 1214. It has no flow except at times of rain or melting snow, and the flood discharge is small. It is pro¬ posed to divert it so that it will enter the canal opposite station 1210. This diversion will be in solid rock, with a channel 100 feet wide and 5 feet below the water surface in the canal for a distance of 50 feet back from the channel line. Back of this point the grade of the diversion will be such as to carry the water from the present bed of the creek. Bridges and road changes required are as follows: At station 127 the channel crosses the Grand Trunk Railroad, which will require a single-track swing bridge. At Lasalle, from station 860 to station 875, it will be necessary to build a combined swing bridge to accommodate the highway travel, aud a double track for the Buffalo and Niagara Falls Electric 350 DEEP WATERWAYS. Railway and 0.5 mile of new road; a double-track swing bridge for the New York Central Railroad and 1.8 miles of new roadbed, and a single-track bridge for the Erie Railroad and 2.1 miles of road. It is also necessary to build 1.5 miles of new highway to the right of the channel. The highway travel can be accommodated by a ferry at station 910. At station 1040 a double-track swing bridge should be built for the New York Central Railroad, and also to accommodate the highway travel. It will also be necessary to move the junction of the Niagara Construction Company’s Railroad farther west, involving the build¬ ing of 1.3 miles of mew roadbed. .A highway bridge must be provided near station 1188. At Lewiston a crossing must be provided for the New York Central and the Rome, Watertown and Ogdensburg railroads, which may be done by building a fixed bridge across the channel at Lock No. 4, near station 1235, and bringing both roads to it on top of the bluff, instead of following down the gorge of the Niagara River, as at present. After crossing the bridge both roads would drop to the level of the lower plane by an eas 3 T descent along the side of the escarpment and to the present junction of the two roads. Lewiston could then be reached bj r a separate line to the east of Lock No. 9. This would require the building of 5.9 miles of new double-track and 3 miles of single-track road. A highway crossing in the village of Lewiston must be provided which can also be used for the crossing of the electric railway. This can be accomplished by building a bridge over Lock No. 9 and con¬ necting the railways with it on both sides, involving the building of 1.3 miles of single-track electric railway. A tunnel under the canal must also be provided at Buffalo for carrying the water supply from the in-take pipes to the waterworks pumps. Estimates are based on the following widths of channel: From deep water in Lake Erie to Black Rock Harbor, 600 feet wide; from Black Rock Harbor to the foot of Squaw Island, canal section, in the Niagara River, 600 feet wide; from Lasalle to the river at Lewiston, standard canal section. Table No. 1, following, shows the existing railways and highways that cross the proposed center line; No. 2, the location, length, and cost of the proposed bridges required on this route; No. 3, the location, cost, etc., of locks; No. 4, the detailed estimates for 30-foot channel, and No. 5 the detailed estimates for the 21-foot channel. DEEP WATERWAYS 351 Table No. 1.— Existing crossings — Lasalle-Lewiston route. . Location. Present Place. Station. grade. R/ 0 imn ks. Railway. International bridge. 128 590.0 Single-track railroad between bridge over Niag¬ ara River and the one over Black Rock Harbor. Lasalle... 852 +50 573.0 Buffalo and Niagara Falls Electric R. R., double track. Do... 853 +50 576.0 New York Central and Hudson River R R., double track, Buffalo and Niagara Falls branch. Do.... 854+50 576.0 New York, Lake Erie and Western R. R., single track, Buffalo and Niagara Falls branch. 1040 604.0 New York Central and Hudson River R. R., double track, Niagara Falls and Lockport branch. Lewiston. 1236 548.0 Rome, Watertown and Ogdensburg R. R., single track. Do. 1245 370.0 New York Central and Hudson River R. R., Niagara Falls and Lewiston branch, single track. Do.. 1258 345.0 Do. Highways. Lasalle. 852 +50 860 573.0 574.0 River road. 864 574.0 Not much used. 909 573.0 Mile-line road. 991 597.0 Packard road. 1019,1023,1027 1031,1034,1035 \ f. Streets laid out, but in no way improved. 1048 606.0 Lockport road. 1052,1057 1061,1065 }.- Streets plotted, but not improved. 1095 603.0 Witmere road. 1188 616.0 Reservation road. 1225 593.0 River road. Lewiston. 1277 346.0 Center street, single-track electric railroad, Lewiston to Youngston. Do. 1284 360.0 Second street or River road; streets are also crossed at 1258,1264,1271, and 1290, but they are only slightly improved. Table No. 2. — Location, cost, etc., of proposed bridges — Lasalle-Lewiston Route. Location. Intei-national bridge. Lasalle. Do Do Do Do ... Do ... Lewiston Do . Lasalle . Do . Total. Sta¬ tion. 128 861 868 874 1040 1188 1235 1292 .890 1235 Kind of bridge. Railway Highway and electric rail¬ way. Railway. .do. Highway and railway. Highway. Railway. Highway and electric rail¬ way. Bridges not over canal. Highway ... Railway .... Num¬ ber of tracks. 1 Swing or fixed. 1 Swing. _do.. _do.. _do.. —do.. -do.. Fixed . Swing. Fixed . ...do._ Thirty-foot channel. Num-_ her of spans. I Total length. Esti¬ mated cost. 1 53< ^ $146,686 5171 1 550 155,230 530 1 550 225,607 530 1 5374 128,230 5174 3 711 303,628 687 1 580 68,546 556 1 353 101,572 333 1 2404 47,434 197 1 100 8,504 100 1 30 7,095 30 1,192,5112 Twenty-one foot channel. Total length. Esti¬ mated cost. $129,650 130,206 204,575 153,398 a 276,033 63,478 92,366 44,098 8,504 7,095 1,109,403 a Double-deck, draw span, and two 63-foot girders. Feet clear opening. Note.—H ighway bridges. 22 Single-track railway bridges. 14 Double-track railway bridges . 26 Double-track double-deck bridges. 29 352 DEEP WATERWAYS Table No. 3.— Lasalle-Lewiston route. LOCKS. Location. Length of level above. No. Lift. Kind. Elevation stand¬ ard, low water. Place. Station. Single or double. Individual or in flight. Above lock. Below lock. Buffalo .... Lewiston, upper flight Do _ _ 88+07 1200+47 Miles. 21.3 1 o 3 4 5 6 7 8 9 Feet. 8 40 40 40 40 40 40 39.4 39.4 Double. — do... _do_ Individual... Flight. .do. 574.5 563.5 523.5 483.5 443.5 403.5 363.5 323.5 284.1 566.5 523.5 483.5 443.5 403.5 363.5 323.5 284.1 244.7 Do . _do_ .do. Do . _do... _do.. Do . _do_ .... .do.. Do 1257 + 47 1284+47 1291+87 _do_ _ . do_ Lewiston, lower flight Do .. 0.4 _do... _do.. .do. .do... COST.d Location. 30-foot channel. 21-foot channel. Operating machin¬ ery. Buffalo..... 6 SL 720,358 12,296,609 4.420,037 i>$l, 135,413 8,206,614 2,952.169 $100,000 | 500,1)00 Lewiston, upper flight______... Lewiston, lower flight____ Total______ 18.437.004 600,000 12.294.196 600,000 600,000 Operating machinery____ Total......... 19,037,004 12,894,196 a The cost is that of the structure complete except the excavation. b For regulated lake level. Table No. 4.— Estimate for 30-foot channel. NIAGARA RIVER (FROM STATION — TO STATION 8291. Quantity. Cost per unit. Total. With regulating works. Earth...cubic yards.. Rock, dry. ..do — Rock, wet (in Lake Erie).. .do_ Rock, wet (in Niagara River)....do_ Hardpan..... do_ Regulating works....... 13,909.221 90.014 1,687,650 9,128,800 4,200 $0.15 1.00 3.00 2.00 .30 $1,951,383 90,014 5,062,950 18,257,600 1,260 796.92)1 1,720,358 100,000 Lock No. 1... Operating machinery. Total...... 27.980,488 Without regulating works. Earth ..cubic yards.. Hardpan .. do_ Rock, dry... do_ Rock, wet (in Lake Erie).do_ Rock, wet (in Niagara River)....do_ Lock No. 1...... 13,045,421 4,200 90,014 2,227,900 9,424,300 . 15 .30 1.00 3.00 2.00 1,956,813 1,260 90,014 6,683,700 18,848,600 1,726,838 100,000 Operating machinery... Total.... 29,407,225 Quantities common to both plans. Retaining wall ... . cubic yards.. Slope wall.square yards.. Back fill. cubic yards.. Timber cribs, pine...feet B. M.. Hemlock. do ... Stone fill.cubic yards.. Bridges.... number 224,148 22,957 223,702 756,860 4,012,480 48,000 436,330 58,860 1 46 1,000 4.00 1.10 .25 a 30.00 «23.00 a 50.00 .03 .60 896,592 25,25)1 55,926 22,706 92.287 2.400 13,090 35,316 146,686 184,000 30,000 Right of way...acres.. Tunnel waterworks.linear feet.. Total. 4,000.00 30.00 1,504,256 a Per 1,000 feet. DEEP WATERWAYS 353 Table No. 4 .—Estimate for 30-foot channel —Continued. SUMMARY FOR NIAGARA RIVER. With regulating works: Excavation, etc ...... $27,980,488 Retaining wall, etc....... 1,5 4,256 Total.-...- 29,484,744 Without regulating works: Excavation, etc... 29,407.225 Retaining walls, etc.-. 1,504,256 Total........ 30,911,481 CANAL SECTION (STATION 829 TO STATION 1313). Quantity. Excavation. Cost per unit. Total. Lasalle to foot of the first flight of locks (829 to 1262 + 21): Earth...cubic yards.. Hardpan ....do.. Rock. ....do_ Foot of first to foot of second flight of locks (1262 + 21 to 1296 +61): Earth.cubic yards.. Rock .....do_ Foot of second flight of locks to 30-foot depth, Niagara River (1296 + 61 to 1313): Earth. cubic yards.. Rock, dry.do_ Rock, wet...do- 6,707,150 1,094,400 20.091,400 2,030,477 1,997,466 170,378 112,496 190,296 $0.20 $1,341,+30 .311 328,320 . 65 13.059,410 .20 406,095 . 65 1,298,353 .20 34,076 .65 13.122 1.75 333,018 Walls , locks, etc. Retaining wall.. Slope wall. Back fl 1.... Timber cribs: Pine... Hemlock .. Oak.. Iron. Stone filling.... Locks 2 to 9, inclusive..... Operating machinery... Bridges. Steam ferries.. Railroad changes. Entrances of streams. Right of way: Village property.. Farm property. Total. ..cubic yards., square yards.. ..cubic yards.. 105.881 4.00 29,713 1.10 827.794 .25 423.524 32,684 206.949 ...feet B. M_. .do 1, ..do .. .pounds.. cubic yards.. number.. _do — 170,960 1:58,080 7. 440 119,435 a 30.00 a 23.00 a 50.00 .03 15,900 .60 9 1 5,129 26,176 372 3,583 9,540 16,716,646 500,000 1,045.846 20,000 363,424 20,255 acres. .do. .. 856 2,055 728,800 272,250 37,249,002 a Per 1,000 feet. MOUTH OF NIAGARA RIVER (STATION 1655 TO STATION 1703). Excavation. Earth....cubic yards.. $98,111 SUMMARY. With regu¬ lating works. Without regulating works. Niagara River. $29,484,744 37,249,002 98, 111 66.831,857 $30,911,481 37,249,002 98,111 68.258,594 Can al section...... Mouth of river..... Total...... H. Doc. 149 ■23 354 DEEP WATERWAYS Table No. 5.— Estimate for 21-foot channel. NIAGARA RIVER (FROM STATION- 38 TO STATION 820). With regulating works. Quantity. Cost per unit. Total. Earth . Rock, dry... Rock, wet (in Lake Erie)- Rock, wet (in Niagara River) Regulating works. Lock No. 1____ Operating machinery. Total__ Without regulating works. cubic yards .do_ .. ..do_ .do_ 6,809,904 13,219 419,850 2,561,800 SO. 15 1.00 3.00 2.00 $1,021,486 13.219 1,259,550 5,123,600 796,923 1,135,413 100,000 9,4.50,191 Earth. Rock, dry._.... Rock, wet (in Lake Erie). Rock, wet (in Niagara River) Lock No. 1-.. Operating machinery. Total. Quantities common to both plans. cubic yards ..do.... .do_ .do_ 6,846,154 13,219 718.100 2,843,950 .15 1.00 3.00 2.00 1.026,923 13,219 2,154,300 5,687,900 1,141,893 100,000 10,124.235 Retaining wall. Slope wall. Back fill.. Timber cribs, pine .. Hemlock.. Oak... Iron..... Stone fill.... Bridge.. Right of way__ Tunnel waterworks Total. ..cubic yards., square yards ..cubic yards.. .feet B. M._ .do_ .do_ .pounds.. ..cubic yards.. .number.. _acres.. _linear feet.. 224,148 4.00 23,887 1.10 213,830 .25 756,860 a 30.00 2,693,240 a 23.00 48,000 a 50.00 311,340 .03 41,010 .60 1 46 1,000 30.00 896.592 26,276 53,458 22,706 61,945 2.400 0,340 24.606 129,650 184,000 30,000 1,440,973 a Per 1.000 feet. SUMMARY FOR NIAGARA RIVER. With regulating works: Excavation, etc .... $9,450,191 Retaining walls, etc... 1,440,973 Total....... 10,891,164 Without regulating works: Excavation, etc. 10.124,235 Retaining walls, etc... 1,440,973 Total... 11,565,208 CANAL SECTION (STATION 829 TO STATION 1310). Excavation. Lasalle to foot of the first flight of locks (829 to 1255 + 01): Earth.cubic yards.. Hardpan ...do_ Rock ..... do_ Foot of first to foot of second flight of locks (1255 + 01 to 1296 + 61): Earth.cubic yards.. Rock.do_ Foot of second flight of locks to water 21 feet deep, in lower river (1296 + 61 to 1310): Earth .. .cubic yards . Rock, dry ..do_ Rock, wet.do ... Quantity. Cost per unit. Total. 5,670,161 $0.20 $1,134,032 824,650 .30 247,395 16,089.387 .65 10, 458,102 1,934,954 .20 386,991 1,181,361 .65 767,885 103,478 .20 20,696 107,900 . 65 70,135 64,229 1.75 112,401 DEEP WATERWAYS. 355 Table No. 5 .—Estimate for 21-foot channel —Continued. CANAL SECTION (STATION 829 TO STATION 1310)—Continued. Quantity. Cost per unit. Total. Walls , locks , etc. Retaining wall.-.. Slope walls..-... Back fill.... Crib walls: Pine...... Hemlock . _ . .. Oak.... Iron..... Stone fill.... Locks 2 to 9. inclusive _ _ . .cubic yards.. _square yards.. .cubic yards.. ..do_ .do ... ..cubic yards.. 40,624 45.540 372,275 470,710 4,606,915 59,724 431,586 63,306 $4.00 1.10 .25 a 30.00 u23.00 a50.00 .03 .60 $162,496 50,094 93,069 14,121 105,959 2,988 12,948 37.984 11,158,783 Oneratiner machinerv _ _ _ ... ___ __ 500,000 979,753 Bi-idees _ _ _ . .number 9 Steam ferry .... 1 20,000 Railroad changes.._... ... 363,424 20,255 Entrances ot' streams... Risrht of wav_ _ _ _ _ Village property._...... .acres.. 856 728,800 Farm property. .do .. 2,055 272,250 Total ... _ 27,720,559 a Per 1,000 feet. SUMMARY. With regulating works. Without regulating works. Niagara River ... . $10,891,164 27,720,559 $11,565,208 27. 720, .559 Canal section . ... Mouth of river...... 0 0 Total........ 38,611,723 39,285.767 TONAWANDA-OLCOTT ROUTE. LOCATION. As before stated, the location of this route is identical with that of the Lasalle-Lewiston route from Lake Erie to station 541, opposite Tonawanda. At this point it swings to the right, passes just above the head of Tonawanda Island, and leaves the river near the bridge, connecting the mainland with the island, and continues through the city of North Tonawanda and to Olcott, as shown on plates 16 and 17. Another location was considered, leaving the river near the foot of the island, swinging to the right, and joining the adopted location in the vicinity of Sawyers Creek, but this was abandoned for the rea¬ son that it gives bad alignment and leaves the river at nearly right angles to the general direction of the current. The latter condition would make it very difficult, if not impossible under some conditions, to manage ships when passing from the river section to the canal sec¬ tion, and vice versa. The location adopted gives an easy curve for leaving the river and a good alignment thereafter. The cost of the two locations is about the same. A different location was also con¬ sidered from Sawyers Creek to below the escarpment west of Lock- port. This would leave the adopted location in the vicinity of Saw- 356 DEEP WATERWAYS. yers Creek, swing to the right and cross the Erie Railroad about oppo¬ site station 000, and continue south of the railroad to about opposite station 1200, where it would swing to the left to the head of the gulf and thence by a tangent to the main line at station 1375. This would require two extra railroad crossings, but the most serious objection is the sharp curve necessary to reach the head of the gulf. The align¬ ment is bad. A preliminary study was made of a location entering the gorge of Eighteen Mile Creek in the vicinity of the village of Newfane, and continuing down the creek to Olcott, but the difficulties of construct¬ ing a channel on this location were such as to make it more expen¬ sive than a line entirely independent of the gorge. Estimates in detail were not, therefore, made on this line. From station 1245 to station 1658 an alternate line is estimated, which is shown on the maps as “B line.” The difference in cost of the two lines from station 1245 to station 1358, common points on both, is 1398,000 for the 30-foot channel and $543,800 for the 21-foot channel in favor of the “A” or adopted line. The alignment of the adopted line is better and it would be much more satisfactory to navigate. The material to be excavated is similar to that found on the Lasalle- Lewiston route. Limestone rock appears above the grade of the 30-foot channel between stations 602 and 607, and then dips below grade and reappears at station 715 and continues above grade to station 775, where it dips below grade and reappears at station 870, and then continues above the grade to station 1345 at the second escarpment. From the foot of lock No. 5 to Lake Ontario the rock is the so-called red shale. It is in fiat layers from 1 to 3 inches thick, though at rare intervals layers of 4 to 6 inches are found. For a depth varying in different locali¬ ties from 10 to 30 feet these layers are separated one from another by a thin layer of reddish mineral dirt, and they are broken into small pieces, which gives it the appearance of being soft shale, but in fact each individual piece is quite hard. On breaking a solid piece at right angles to its bed it is seen that the interior portion is a dark-gray color and the outer portions are of a reddish color. In these respects it differs from the red shale found at Lewiston. DIAMOND-DRILL BORINGS. To determine the character of the rock nine diamond-drill borings were put down on this line. The location of each is shown on plates 16 and 17, and they are described in the report of R. C. Smith, Appendix No. 19. Boring No. 1 is located at Sawyers Creek, 250 feet to the left of station 766. Rock was struck at elevation 545 and penetrated 16 feet lo elevation 529, which is 6 feet below the grade of the channel. DEEP WATERWAYS. 357 The rock its a very hard light-gray limestone, much broken up and in layers one-quarter inch to 3d inches thick. Boring No. 2 is just east of Pendleton Center and 900 feet to the right of station 1010. Limestone rock was struck at elevation 566 and was penetrated 89 feet to elevation 477. The first 50 feet was hard Niagara limestone, in thin layers near the top and thicker layers tow r ard the bottom, one being 22 inches thick. The next 39 feet was a very hard limestone conglomerate of dark color. Boring No. 3 is near the gulf west of Lockport and 1,400 feet to the right of station 1266. The stratification is: 588 to 582. Soft limestone, full of shells. 582 to 572. Mixture of limestone and sandstone in hard, thin layers. 572 to 502. Firm Niagara shale. 502 to 490. Limestone. 490 to 465. Limestone and shale, with few layers of green shale at the bottom. 465 to 425. Medina sandstone mixed with red and green shale. 425 to 407. Very hard Medina sandstone mixed with hard red shale. 407 to 399. Hard Medina sandstone. 399 to 388. Sand shale mixed with gray sandstone, shale softer. By comparing this with borings Nos. 3 and 4, at Lewiston, it will be seen that the stratification is nearly the same in both cases and that the different materials were found at about the same elevation. Borings Nos. 4 and 5 were put down through Niagara limestone to determine, in connection with boring No. 3, the elevation and dip of the shale, but no attempt was made to penetrate the shale. No. 4 is located 250 feet to the right of station 1238. The elevation of shale is 558. No. 5 is 750 feet to left of station 1277. Elevation of shale, 580. These give a dip of 5.6 feet per 1,000 feet, and if the same rate of inclination continues to the south, the shale would dip below the bottom of the channel near station 1200. Boring No. 6 is located 600 feet to the left of station 1316, at which station the grade of canal is 504 feet for 21-foot channel. Rock is found at the surface. 523 to 501. Clinton limestone. The last 15 feet contains layers of green shale. 501 to 486. Hard green, red, and gray sandstones, mixed with red shale and Clinton shale. 486 to 432. Medina sandstone, gray and red; the last 20 feet pink in color; all very hard. 432 to 400. Gray sandstone and shale mixed. 400 to 393. Firm red and green shales. 393 to 385. Soft red shale. 385 to 332. Firm red shale. Boring No. 7 is located 1,250 feet to the left of station 1400. The grade of 21-foot channel is 344 at this station. Soft red shale was 358 DEEP WATERWAYS. struck at elevation 353 and continued to 310; from this elevation to 207 firm red shale was found. Boring No. 8 is located 600 feet to the left of station 1581. The elevation of grade of 21-foot channel is 344. Soft red shale was struck at elevation 345, changing to firm red shale at 328 and con¬ tinuing to elevation 176. Boring No. 9 is located at the shore line of Lake Ontario near Olcott, and 150 feet to the left of station 1893. The grade at station 1893 is 223.5 for 21-foot channel. The stratification is as follows: 251. to 242. Sand and gravel. 242 to 223. Soft red shale. 223 to 206. Firm red shale. 206 to 182. Firm red shale, mixed with soft Medina sandstone. It will be seen from these borings that all of the rock to be excavated above the second escarpment is either limestone or firm shale and below this point it is either soft or firm red shale. The general char¬ acteristics of these several rocks, except the red shale, are the same as those for the Lasalle-Lewiston route and need not be further con¬ sidered here excexA to state that the shales south of station 1350 will cost about as much to excavate as the limestone, and from this station to Lake Ontario the top 15 feet will be much easier to excavate. An examination of the shale found along the Erie Canal near Lock- port shows it to be a well-defined shale, much softer and more broken than the limestone, but to the southwest it gradually becomes harder and in general appearance more like the limestone. At the point where it dips below the grade of the Erie Canal it is difficult to dis¬ tinguish between the two from external appearances. The material overlying the bed rock is sand and gravel from station 541, in the Niagara River, to station 600. From station 600 to station 1100, near Ilodgeville, the top 6 to 15 feet is a firm yellow clay mixed with varying quantities of sand. Below this is found a soft red clay varying in thickness from 5 to 20 feet. From station 600 to station 940 it is generally from 15 to 20 feet, and from station 940 to station 1100 it is from 5 to 10 feet thick. It is a pure clay, almost entirely free from grit. Under this soft clay and overlying the bed rock is a mixture of red clay and gravel varying in hardness from a good hard clay to hard-pan or bowlder clay. From station 1100 to the gorge of Eight¬ een-Mile Creek it is a drift composed of clay mixed with sand and gravel, and may be classed as a firm clay. In the gorge of Eighteen- Mile Creek to Olcott it is a very soft silt and sand. GRADES AND LOCKS. As already explained, the water surface at the point where the channel leaves the Niagara River is taken at elevation 565 and con¬ tinues level to lock No. 2. The slope required to carry the necessary water for lockage and power development is very slight. Lock No. 2 has a lift of 40 feet, so that the elevation of the water surface below DEEP WATERWAYS. 359 it is 525 and continues level to the head of the flight of three double locks, Nos. 3, 4, and 5, which have a lift of 40 feet each, making the elevation of the water surface below the flight 405, which continues level to lock No. 0, with a lift of 40 feet, making the level of the water surface below 3G5 feet, which continues level to lock No. 7, with a lift of 304 feet, making the water surface below 334.5, which continues level to the head of the flight of three double locks, Nos. 8, 9, and 10, with a lift of 30 feet each, making the water surface below lock No. 10 244.5, which is the level of low water in Lake Ontario. The grade of the 21 and 30 foot channels is 21 and 30 feet, respectively, below the water surface. The location and lift of locks are the same for both channels. As before stated, the lock chambers for the 30-foot channel are 740 by 80 feet for single locks, but for double locks one is 740 by 80 feet, and the other 740 by 60 feet. Regulation of water surface in pools between locks JVos. 2 and S and 5 and 6 .—Locks Nos. 2 to 6, inclusive, all have the same lift. Locks Nos. 2 and 6 are single locks 80 feet wide, while Nos. 3, 4, and 5 constitute a double flight. The flight on the right is 80 feet wide, while that on the left is but GO feet wide. In the consideration of the locks, it is assumed that vessels will ordinarily use the right-hand flight of locks; that is, down-bound vessels will use the 80-foot locks and up-bound vessels the GO-foot locks. Vessels passing down consecutively will leave no change in the water levels in the two pools between the locks after they have passed out. Vessels locking consecutively upward will lower the level of the lower pool and raise that of the upper pool. Vessels alternating will raise the lower pool and lower the upper one. These contingencies, as well as any deficiency in the long level between locks Nos. G and 7, have been provided for by the construction of open channels around lock No. 2 and also the flight, and by making use of the bed of Sherman Creek, at lock No. G. The channel at lock No. 2 is controlled by sluice gates so that the required amount of water may be supplied the basin below. The channel connecting the upper and lower pools is provided with a small weir and sluice gate. The weir serves to pass any excess of water in the upper pool and the sluice gate to supply deficiencies in the lower pool. The lower pool is enlarged on the left by the construction of an embankment around the low ground, no excavation being required, and is regulated by a small weir discharging into the old bed of Sherman Creek. Sherman Creek crosses the level between locks Nos. G and 7 just below lock No. 6 by means of a discharge weir over the right side of the canal. This weir also serves to regulate the water in this level. POWER. Electric power for operating the valves, gates, pumps, etc., can be developed by taking water from the level above lock No. 2 and lead¬ ing it through a tunnel and pipes to the power house, about 1,200 feet 360 DEEP WATERWAYS. to the right of station 1200, using the gulf and Eighteen-Mile Creek as a tail race. A head of 100 feet can be obtained. From this central station the power can be distributed to each of the locks on this route, excepting lock No. 1, at Buffalo, where it would be cheaper to develop power near the site of the lock. “n” LINE. The “B” line leaves the main line at station 1245, swings to the right, and enters the gulf about station 1281, where a flight of two double locks with a lift of 40 feet each is to be built, making the ele¬ vation of the level below 485. It then follows the gulf to station 1308 and swings to the left to sta¬ tion 1300, where a flight of three double locks with 40 feet lift each is put in, making the level of the water below 305, which is carried to station 1630, where a single lock of 304 feet lift is put in, making the level below 334.5, the same as below lock No. 7 on the main line. It then joins the main line at station 1658, and from this point to Lake Ontario the two lines are identical. A masonry dam 70 feet high across the gulf would be required to the right of station 1327. The west branch of Eighteen-Mile Creek would be taken into the channel at station 1400 and pass out over a spillway at station 1490. Six highway bridges and 2 miles of new highway would be necessary. Power for operating valves, gates, pumps, etc., can be developed by taking water from the channel about station 1335 and carrying it by canal and pipes to the power house in the gulf, which would be used as a tail race. A net head of 110 feet can be had. However, this would require a by-pass about 1,000 feet long to carry the water to the left of lock No. 2 from the upper level to the gulf opposite sta¬ tion 1290. A net head of 70 feet could also be utilized at this point, making a total net head available of 180 feet if two separate plants are installed. The power thus developed can be distributed electric¬ ally to all of the locks on the route except No. 1, at Buffalo. Estimates are based on the following widths for both the 21 and 30 foot channels: In Niagara River, 600 feet wide. From the Niagara River to the gorge of Eighteen-Mile Creek, at the foot of lock No. 10, standard canal sections. Foot of lock No. 10 to Lake Ontario, 400 feet wide. From the shore of the lake to deep water, 600 feet wide. HARBOR. / The harbor at Olcott is shallow, with a shale rock bottom, which must be excavated to the required depth and a breakwater built to protect it from the waves of the lake. The discussion and estimate of the breakwater are given in Appendix No. 3. The design of the channel from the foot of lock No. 10 to the lake contemplates an DEEP WATERWAYS. 361 inner harbor 400 feet wide and about a mile long, so that it will not be necessary to make any further excavation in the lake than the 600-foot channel until the volume of traffic requires it. It will be necessary to make provision for taking care of the flood waters of the streams on this route. Tonawanda Creek can enter the Niagara River as at present with¬ out any additional work. Sawyers Creek, at station 767, has a flood discharge of about 1,200 cubic feet per second and practically no flow during dry weather. The bed of the stream is at about elevation 570, or 5 feet above the low water in the canal. The bottom of the canal at this point is in solid rock and retaining walls are built on each side. It is proposed to fill the bed of the creek on the right to the level or above the top of the retaining wall; on the left, for a length of 200 feet, the wall will be built to elevation 555, giving a sectional area of opening at low water of 2,000 square feet, through which the water can pass into the canal. This area will be larger at high water. A basin 200 feet wide and 600 feet long will be excavated to elevation 552, in which the sediment will be deposited before entering the canal. At the upper end of this basin a dam will be built to the elevation of the present bed of the stream. Bull Creek is crossed at station 841, where the bottom of t he canal is in earth. It can be taken into the canal through a basin 100 feet wide and 10 feet deep excavated back 500 feet, with a dam for holding back the earth. The bottom and sides of the basin to above the high- water line should be paved near the canal. The method of taking care of the flood waters of the small creek crossed just below Lock No. 6 has already been described. The next and last stream is Eighteen-Mile Creek, which 1ms practi¬ cally no discharge at low water under normal conditions, but under existing conditions its low-water flow depends on the amount of water being drawn from the Erie Canal at Lockport. No measurements could be obtained of its flood discharge. The back water from Lake Ontario extends to the railroad bridge, some 3,000 feet above where the canal enters the creek, and if the dirt overlying the rock is exca¬ vated back for a distance of 1,000 feet it will give ample room for sedi¬ mentation before the waters enter the canal. In the event the canal should be built on this route, a number of changes of both railroads and highways will be necessary, and bridges must be built to cross the canal. These changes and structures would be subject to agreement between the United States and the several parties interested, and the final disposition might be materially changed from any project now submitted. This statement applies with equal force to all the routes considered. However, for the pur¬ pose of estimating the cost of the work, the following project is suggested. DEEP WATERWAYS. 362 A rearrange men of the railroads should be made in the vicinity of Tonawanda. The yards of the New York Central and the Erie rail¬ roads should be moved entirely north of the channel, and the align¬ ment of the Lockport branch of the New York Central should be changed to leave the main line about one-half mile north of station 610, and then continue in a straight line to a junction with the pres¬ ent location near Sawyers Creek. The Lockport branch of the Erie Railroad, which has recently been converted into an electric road, should leave the main tracks about 1,000 feet south of station 617 and follow the present line of the New York Central road to opposite sta¬ tion 660, where it would join the existing line of the Erie. This arrangement would locate the Lockport branch of the Erie road entirely to the right and that of the New York Central entirely to the left of the canal. Taking up the bridges and road changes in regular order, beginning at the Niagara River, we have the following: 1. Move bridge at station 600 about 1,000 feet downstream. 2. Build new highway bridge at station 607. 3. Build new double-track bridge at station 613 for the New York Central Railroad. (Bridges 2 and 3 may be combined into one, but the cost would be about the same in either case.) 4. Build single-track swing bridge at station 619 for Erie Railroad. o. Build combined highway and single-track electric railway bridge at station 625. 6. Build combined highway and double-track electric railway bridge at station 641. 7. Build highway bridge at station 670. 8. Build highway bridge at Sawyers Creek, station 765, and change Shawnee road to the west of New York Central Railroad so as to reach this bridge. 9. Build highway bridge at station 849 and change both the Bear Ridge and Town Line roads to cross this bridge. 10. Build highway bridge at station 1026 and change the Pendleton, Sulphur Springs, and Hodgeville roads to cross this bridge. 11. Build combined highway and double-track bridge for New York Central Railroad at station 1232, and change both highway and rail¬ road to cross at this point, and build new highway to left of channel from station 1232 to station 1302. 12. Build highway bridge over Lock No. 2, at station 1302, and change Gulf road on east side to connect with it. 13. Build highway bridge at Lock No. 6, station 1394, and change road to cross it. 14. Put in ferry for highway crossing at station 1454. 15. Build highway bridge over Lock No. 7 and change road to cross it. 16. Build highway bridge at station 1700. 17. The canal crosses the Rome, Watertown and Ogdensburg Rail¬ road above the head of Lock No. 8, station 1808. The elevation of the DEEP WATERWAYS. 363 low water on this level is 334.5 and the high water about 336. If the top of the center pier is made 3 feet above high water and the top of the rail 14 feet above the top of the pier, we would have 336+3 + 14 = 353 for the grade of the railroad, if crossing at this point. The eleva¬ tion of the present grade of the road is about 318. It would, there¬ fore, be necessary to raise the roadbed 35 feet at the canal crossing and extend the filling each way some 4,000 feet, and rebuild the bridge across Eighteen-Mile Creek. As an alternate plan, the alignment of the road can be changed to cross below lock No. 10, station 1840, with a fixed span giving 90 feet clear headway. Estimates are made on this plan for a fixed bridge for both railroad and highway, together with the necessary changes of roads. 18. Provision should be made for a highway crossing at Olcott, either by steam ferry or a drawbridge. Estimates are based on the latter plan. In addition to the above items the pipe system of the Tonawanda waterworks should be rearranged and the electric transmission cable running from Niagara Falls to Buffalo should be taken under the canal by tunnel. In case the “B” line should be adopted, similar changes of high¬ ways and railroads would be necessary. For the 21-foot channel the location of canal, lift and location of locks, location and character of structures, etc., are the same as for the 30-foot channel. Estimates of quantities and cost of all work, including right of way, are given in the following tables: Table No. 6. —Existing crossings — Tonawandci-Olcott route. Location. Present Remarks. Place. Station. grade. Railway. International bridge. 128 590.0 Single-track railroad. Tonawanda.. tan 576.0 Island street. Tonawanda. Single-track railroad and highway bridge. Draw span, 2 openings, 80 feet in clear each. Do.. 611-613 575.0 New York Central and Hudson River R. R., Buffalo and Niagara Falls Branch, 9 tracks—2 main tracks and 7 sidings. Do.. 617 576.0 Erie R. R., Buffalo and Niagara Falls Branch, 7 tracks—1 main track and 6 sidings. Do.. 627 575.0 Single-track electric railway. Do. 641 579.0 Buffalo and Niagara Falls double-track electric rail¬ road. Do.. 644 580.0 Erie R. R., Lockport Branch, single track. Do.. 716 580.0 New York Central and Hudson River R. R. Single track, Lockport Branch. Do.. 1238 610.8 New York Central and Hudson River R. R. Double track, Lockport and Niagara Falls Branch. Newfane station_ 1808 318.0 Rome, Watertown and Ogdensburg R. R. Single track. Highway. North Tonawanda 601 577.0 Swing bridge across branch of Niagara River carry¬ ing single-track siding. Do. 607 577.0 Main street. Important. Do__ 627 575.0 Vandervort street, single-track electric railroad. Do.. 641 579.0 Paynes avenue, double track electric railroad. Be¬ sides these streets, Shenk street at 619, Oliver street at 623, Robinson street at 631, Keil street at 636, Mil¬ ler street at 648, are crossed in North Tonawanda. 364 DEEP WATERWAYS. Table No. 6 —Existing crossings — Tonawanda-Olcott route — Continued. Location. Place. Station. Present grade. Remarks. Highway —Continued. Nortli Tonawanda .. 663 575.0 Nash road. Important. 689 575.0 Martinsville road. Unimportant. 764 East avenue. Unimportant. 766 578.0 Creek road. Important. Shawnee road. Not very much used. 786 576.0 832 579.0 Town line road. Not very much used. 869 578.0 County road. Unimportant. 966 590.0 Cross road. Considerably used. 1018 590.0 Pendleton road. Important. 1035 591.0 Sulphur Springs road. Considerably used. 1093 589.0 Hodgeville road. Not much used. 1149 596.0 Buttermilk lane. Not much used. 1181 608.0 Hinman road. Important. 1236 610.0 Lockport road. Important. 1276 615.0 Pekin road. Considerably used. 1293 573.0 Crapsey road. Not important. 1310 530. 0 Gulf road. Not important. 1357 425.0 Eldridge road. Not important. 1389 390.0 Stone road. Important. 1454 370.0 Turnpike road. Not verv much used. 1535 360.0 Swamp road. Unimportant. 1579 358.0 Bennett road. Unimportant. 1661 346. 0 Unimportant. 1700 345. 0 Ide road. Considerable travel. 1782 325.0 Not very much used. 1825 318.0 West Creek road. Not very much used. 1887 253.0 Lake road. Fixed span bridge across Eighteen-mile Creek. Table No. 7. — Location, cost, etc., of proposed bridges — Tonawanda-Olcott route. Location. Sta¬ tion. Kind of bridge. Si a 5 a>.M ^ s a 2 £ ° Swing or fixed. Num¬ ber of spans. Thirty-foot channel. Twenty-one-foot channel. Total length. Estimat¬ ed cost. Total length. Estimat¬ ed cost. Feet. Feet. International 128 Railway .. 1 Swing.. 1 537.1 $146,686 5174 8129,650 bridge. North Tona- 607 Highway. _do... 1 545 100,246 525 92.226 wanda. Do . 613 Railway.. 9 _do_ 1 551) 231,063 530 203,427 Do_ 619 _do . 1 _do... 1 5374 143,016 51/4 127;412 Do. 625 Highway_ _do... 1 545 105,430 525 89,604 Do. 641 Highway and •> ....do... 1 550 158,466 530 139,938 electric rail- wav. Do .. 670 Highway. _do ... 1 545 105,430 525 89,604 Sawyers 7t55 .do. _do.. 1 545 73,700 525 88,664 Creek. Do. 849 .do. . .do... 1 545 105,430 525 94,547 Do ... 1026 .do. _do... 1 567 547 Do... 1181 ._. do. ...do_ 1 600 74,392 579 69,806 Do. 1232 Railway and 2 _do... 1 738 309;774 718 a 294,069 highway. Do. 1302 Highway .. ... do.. 1 235 19,986 195 Do. 1394 ._ do. ... do.. 1 235 19,986 195 Do.. 1614 _do.. _do ... 1 235 19,986 195 Do. 1700 .do.. do.. 1 547 Do. 1840 Railway and 1 Fixed .. o 558 156,996 558 b 156,996 highway. Olcott. _ 1888 Hii?hwav Swing.. ‘> 858 109,638 858 c 109,638 Bridges not over canal. Tonawanda... 607 Highway. _do. 3 4574 33,091 33,091 Sawyers _do. Fixed 1 80' 3,726 3,726 Creek. Total.... 2,052,190 1,899,240 a Double deck, draw span, two 63-foot girders. c Two draw spans. Note.—H ighway bridges. Single-track bridges. Dou ble-traek bridges. Double-track double-deck bridges b Width C. to C. trusses, 40 feet. Feet clear opening. . 22 . 14 26 29 DEEP WATERWAYS 365 Table No. 8 . — Tonawanda-Olcott route. LOCKS. Location. Length of level Num¬ ber. Lift. Kind. Elevation standard low water. Place. Station. above. Single or double. Individual or in flight. Above lock. Below lock. Miles. Buffalo. 88+97 1 8.0 571.3 565.0 566.3 525.0 Lockport.. 1299 +97 22 9 ‘> 10.0 Single -. _ _do .. Lockport flight_ Do.. 1338 +97 .7 3 1 10.0 io.o Double... . do Plight. . do 525.0 185 0 485.0 445 o Do. 5 io.o ... do.. . do 115 0 405 In 9.2 to 10. inclusive . . -- do — . .pounds.. cubic yards. . 4,330,980 14,683,520 218,160 1,563,154 239,450 a |30.00 a 23. 00 a 50.00 .03 .60 §129,929 33 r. 721 10,908 46,895 143,670 10,465.188 600,U00 Odpth timr machinerv. ... _ _ __. _ _ - _ Kride*es .. n umbel - . 19 1,769.590 199,640 68,9+3 20,000 39,585 1,246,960 823,875 Railroad changes.... . .. TYivftrsinn of st.rp.ams . _ _ . _ _ _ _ _ number I Right of way: . acres.. 190 Farm property . . ... ....do _ 6,249 Total_ _ _ 36,850,840 a Per 1,000 feet. LAKE ONTARIO (STA. 1893 + 50 TO 1915). Excavation. Earth....cubic yards.. Rock . -..do- 8,550 101,750 $0 15 1.75 §1,283 178,063 296,334 475,680 SUMMARY. With reg¬ ulating works. Without regulating works. § 6 ,794,794 36,850,840 475, 680 §7,468,838 36,850,840 475,680 Canal section______.._ Total cost of route_ ______ 44,121,314 44,795,358 WATERWAY FROM LAKE CHAMPLAIN TO THE HUDSON RIVER AT TROY, DESIGNATED AS THE HUDSON RIVER DIVISION OF THE CHAMPLAIN ROUTE. After the surveys were completed, and while the borings were being finished on the Niagara routes, the survey party was engaged in measuring the flood discharge of the Upper Mohawk River and other streams, which will be reported upon in detail in Part III. After the flood measurements were completed the party proceeded to Troy, N. Y., on April 15, 1.898. The borings were completed on the Niagara routes April 20, and the boring parties arrived at Troy on the 22d. Both the survey and boring parties immediately began work on the Hudson River division of the Champlain route. This division begins at deep water in Lake Champlain opposite Port Henry and follows southerly up the lake to Whitehall, where it cuts across country southwesterly to the Hudson River near Fort Edward, following generally the valleys of Wood Creek and Bond DEEP WATERWAYS. 369 Creek. From Fort Edward to the State dam at Troy, which is the southern end of this division, the channel follows generally the bed of the Hudson River. Below the Troy dam and to deep water at Germantown the surveys and borings were made by II. F. Dose, assistant engineer, and are designated as tin* Hudson River survey. Lake Champlain marks the path of ancient glaciers which passed from the valley of the St. Lawrence to the valley of the Hudson. The foothills of the Adirondack Mountains, broken and steep, form its west shore, and the broken and rocky foothills of Green Mountains form its east shore, making the lake a deep, narrow trough cut through these mountains. The valley of Wood Creek and Bond Creek marks the old glacial path from Lake Champlain at Whitehall to the Hudson River at Fort Edward. It varies in width from a few hundred feet to 1 mile and has broken, rocky banks on each side. From Fort Edward to Troy the valley of the Hudson River is generally from one-quarter mile to 1 mile wide, with high, broken banks on the east and west, while the bed of the river is from 400 to 1,200 feet wide. The location follows the lowest ground from Lake Champlain to Troy and is generally parallel to the Champlain Canal, the several sections of which were opened from 1819 to 182-3. SURVEYS. The surveys and examinations on this route were made in accord¬ ance with the instructions of your honorable Board to the assistant engineers, Appendix No. 9, and differed from the surveys of the Niagara route only in that soundings and borings were made in the river and lake and the notes were plotted in the field office as the work progressed. In 1897 I). J. Howell, assistant engineer, began the surveys of the Mohawk River and, in connection with them, measured a base line up the Hudson River from the Troy dam to the Waterford bridge, and also made a shore-line survey of the river and established bench mark No. 7 at Waterford, elevation 30.62, which was used as the starting point for the levels on this line. The base line was started from sta¬ tion 24-G + 77.46 of Mr. Howell’s Hudson River base. Between the Troy dam and Waterford the surveys consisted of soundings and borings in the river. All the topography on the left bank and about half of it on the right bank, also the base line and levels, were done by Mr. Howell. The levels between Mr. Howell’s bench mark No. 1 and the Greenbush bench mark were run by II. F. Dose, assistant engineer. All elevations are referred to the Green- bush bench mark, elevation 14.73 above mean tide at New York. The survey was begun at Troy the latter part of April, 1898, and completed to Port Henry in January, 1899, and covered a length, H. Doc. 149-24 370 * DEEP WATERWAYS. measured aiong the center line of the located canal, of 97.5 miles, divided as follows: Miles. Hudson River from Troy to Fort Edward.. 38.1 Fort Edward to Whitehall___ _ -........ 23.4 Lake Champlain from Whitehall to Port Henry..36.0 Total....-... 97.5 The nature and extent of the work required a larger force than had been employed on the Niagara routes. It was increased to meet the demands and the following organization effected: 1. A base-line party, composed of six men, which ran the base line and the duplicate line of levels and computed the coordinates of each transit station, after first adjusting the line between azimuth points. A copy of the coordinates and bench marks, together with sketches showing their location, was turned into the field office for distribu¬ tion to the various stadia parties according to their location. When the base-line party got further ahead of the other work than was desired, it would take up stadia work. The levels were run with Buff & Berger wye levels and New York rods. 2. Two stadia parties, composed of six men each, who made the sur¬ veys, computed the coordinates of all stadia stations, reduced the elevations of all stadia shots, and plotted on a protractor sheet enough of the runs of each day to ascertain if the circuits closed within the required limits. Each party was given a portion of the line varying, according to the conditions of work, from 2 to 6 miles long, and were furnished with a copy of the base-line notes and the elevations of all bench marks. 3. A sounding party, composed of six men, who made soundings in the Hudson River from Troy to Fort Edward. The sounding ranges were put in about 300 feet apart, and soundings were taken 25 feet apart on these ranges. A copy of the base-line notes and locations and elevations of bench marks were furnished to this party, who staked out all ranges and determined the elevation of the water surface at each range and reduced the soundings to elevations. After complet¬ ing the soundings to Fort Edward it took up stadia work, thus giving three stadia parties in addition to the assistance from the base¬ line party. 4. An office force for plotting and inking the topographic maps, varying from three to fifteen draftsmen and assistants, according to the conditions of the work. The maps are of the standard size, 28 inches by 40.5 inches, and are plotted to a scale of one in five thou¬ sand throughout the entire length of the line. These were matched for the purpose of locating the canal. In addition to these, maps were plotted on a larger scale for localities where the surface was ver* T irregular. Briefly, the method of mapping was to arrange the direction of the sheets so as to take in the greatest possible length of the line. Co- DEEP WATERWAYS. 371 ordinate lines 2,000 feet apart were then plotted, and the base-line stations and stadia stations were plotted by latitudes and departures. The stadia shots were plotted by azimuth and distance—one man called off and another plotted. The plotting was checked by the men reversing positions and repeating the operation. Each sheet was entirely completed by one draftsman. Two-foot contours are devel¬ oped on each map. 5. A superintendent of borings had immediate charge of all bor¬ ings. From Troy to Fort Edward one party, consisting of a foreman, three laborers, and a teamster with team, made the borings on land, and one party, consisting of a foreman and three laborers, made the borings in the river. After reaching Fort Edward the river party was changed to a land party, thus giving two land parties for the work between Fort Edward and Whitehall. For the land work the same plant was used as on the Niagara routes, and that for the river work only varied in having a catamaran from which to put down the holes. It was better adapted to the work than a scow, in that it could be taken apart and carried around the several dams in the river. The surveys from Troy to Whitehall were made wide enough to cover any probable location of the canal. From Whitehall to Port Henry the base line and levels were run along the west shore of Lake Champlain and terminated on the “North base” of the Crown Point base line of the Coast Survey triangulation system 1872, lat. 44° 01' 25.58"; long. 73° 25' 49.55". The stadia survey extended from the shore of the lake to high ground, which was generally only a few hundred feet, and up the several streams to above elevation 100. The surveys for this part of the work were completed October 21, 1898, and the borings to Whitehall October 17, 1898. From Whitehall to Port Henry, a distance of 36 miles, the channel is located in Lake Champlain, which varies in width from a few hundred feet to over 6,000 feet. To make and locate soundings and borings in open water would require a large force and be very expen¬ sive, so it was decided to wait until the lake froze over and do this work on the ice, which was begun December 14, 1898, and completed January 21, 1899. The center line of the proposed location was run with transit, and soundings were made to the banks in the narrow parts and 400 feet right and left in the wide parts of the lake. In the open lake the sounding ranges were 200 feet apart, and the soundings were spaced 50 feet apart on these ranges. Between Whitehall and Putnam station the ranges were 100 feet apart and the soundings every 25 feet on these ranges. BORINGS. On account of cold weather the machines used for making the borings could not be operated without protection for the pumps and water swivels. This was afforded by building a small shanty on 372 DEEP WATERWAYS. runners and moving it from hole to hole with a team. It was pro¬ vided with a stove and trapdoors in the floor and roof, through which the drill rods could be passed. The men worked inside. Each of the three boring parties was provided with one of these shanties, and they were admirably adapted to the work for which they were used. There was no difficulty in working when the temperature was 30° below zero, while in the open air the pumps would freeze up when it was colder than 20° above zero. RESULTS OF STADIA WORK. Taking the base line as standard and correct, and comparing the stadia surveys with it, the following results were obtained: Total number of stadia circuits run...... 290 Tot:) 1 length, in feet .. ....... 2,038,370 Mean length of circuits, in feet . ...... 7,944 Mean error in latitude per circuit, in feet.. . 3. 73 Mean error in departure per circuit, in feet. 2.79 Mean error of closure per circuit.... 1 in 1,373 Mean error of elevation per circuit, foot..... 0.167 COST OF WORK. / For the purpose of considering the cost of the work the line is divided into two parts, one taking in Lake Champlain and the other that part of the line between Whitehall and Troy. The cost of the survey includes all labor, instruments, supplies, etc., connected with the field and office work for the survey, mapping, plans, and estimates and reduction of published charts. The cost of borings includes labor, plant, and all other expenses connected therewith, except the cost of surveys for locating them and a portion of the assistant engineer’s salary for general supervision. For both surveys and borings all plant was considered as sunk when the work was completed and its cost proportioned to the various routes on which it was used. Lake Champlain. Length of line, in miles..... Cost of surveys.... Cost of borings.... Linear feet of borings.... Cost per foot... Cost borings per mile.. Cost surveys per mile.... 36 $5,636.00 $2,268.00 20,169 $0.1124 $63.00 $156.44 Whitehall to Troy dam. Length of line, in miles... 61.5 Cost of surveys.$21,106 Cost of borings .... $4, 905 Linear feet of borings.. 37,822 Cost per foot .. .. $0.1297 Cost of borings per mile. $79.76 Cost of surveys per mile.$343.19 DEEP WATERWAYS. 373 The organization was somewhat broken and the cost of the work increased by the transfer of men to other parties and filling their places with new men. It is quite evident that the work could be done more cheaply by experienced men than by those new to the work. After completing all field work, the boring parties, axmen, etc., were discharged and a force of ten engineers was retained to complete the maps and make final estimates. They reported to the Detroit office January 31, 1899, for this work. LOCATION. The location of the channel was laid down on the topographic maps by your honorable Board. For the purpose of discussing the plans, character of excavation, etc., the line may be divided as follows: 1. Port Henry to Whitehall, which includes all work in that part of Lake Champlain. 2. Whitehall to Fort Edward, which includes all work across the divide between Lake Champlain and the Hudson River. 3. Fort Edward to the State dam at Troy, which includes all work in the Hudson River. PORT HENRY TO WHITEHALL. In designing the channel and determining the grades of same through Lake Champlain, it is assumed that the low-water level will be regu¬ lated as indicated in Appendix No. 8, at elevation 100, and the high water will never be more than 102 feet above mean tide at New York. The grade of the 30-foot channel would then be at elevation 70, and • of the 21-foot, channel at elevation 79, and both level throughout. To near Putnam station, station 1257, the proposed width is 600 feet, except for a short distance at Chipmans Point, where it narrows up to the present width between the rock bluffs on each side of the lake. From Putnam station to station 186-1 the width is 150 feet, except at the “Narrows,” where it is about 250 feet between the high rock bluffs on each side. From station 1861 to Whitehall, station 1957, it is 300 feet wide, except through the rock cut at the “ Elbow,” where it is of the standard canal section. The above widths and loca¬ tion apply to both the 30-foot and the 21-foot channels. The material to be excavated in the lake is silt, sand, and rock and is classified in the estimates as earth and rock. Between Port Henry and a point about 2 miles south of Putnam station the earth is mostly sand, mixed with more or less silt and mud, but from this point to Whitehall it is a soft silt and mud, mixed with a smallamount of sand, all of which has been washed into the lake by the floods of South Bay, Wood Creek, East Bay, and other small streams. At low water the lake is a narrow stream winding through a low, flat marsh about 1,600 feet wide, with an average elevation of 95 feet; at high water the entire marsh is flooded. If the lake is regulated at elevation 100, this marsh would then be under 5 feet of water. It 374 DEEP WATERWAYS. would not support a liigli spoil bank, but it would probably be safe to assume that a bank 5 to 6 feet above low water would be stable. All of the earth to be excavated in the lake can be easily handled with a hydraulic dredge; and if this method is adopted for doing the work, ample spoil area requiring a maximum lift of less than 10 feet above water surface is available. The earth slopes are 1 on 3 for all excavation in Lake Champlain, while they are estimated as 1 on 2 on all other parts of the line. Rock is found at. the following points: At station 570, Larabees Point, Chipmans Point, and station 1404, opposite Cold Spring, small quantities of rock are found above the grade of the 30-foot channel, but it is all below the grade of the 21- foot channel. The rock at Larabees Point is shale or slate, and at the other places it is quartzite. About 1 mile north of Whitehall is a point of quartzite rock projecting into the lake. It is locally known as the “Elbow,” and has an average elevation where the channel cuts through it of about 135, and the length of the cut is about 800 feet. A part of this rock is classified as wet excavation, but most of it is dry excavation. Rock is also encountered in the harbor at Whitehall. WHITEHALL TO FORT EDWARD. As before stated, it is assumed that the level of Lake Champlain will be regulated at elevation 100. Low water in the Hudson River at Fort Edward, where the canal enters, is at elevation 117.6 under present conditions; but if the river be deepened to 30 feet, this low water would be only slightly higher than the Fort Miller dam (115.1), which is 7.1 miles downstream. Two and seven-tenths miles below Fort Miller is the Northumberland dam, with a crest elevation of 102.5. The low-water level of the pool below this dam is about 85. The divide between Lake Champlain and the Hudson River may be crossed in either of two ways: 1. By locking up at Whitehall to the level of the Hudson River and then making the grade of the channel level to Fort Miller. 2. By making a through cut on the level of the lake channel to the Northumberland dam. The first plan would involve supplying water to the high level for lockage at Whitehall and at Fort Miller. This can be done by con¬ structing a system of reservoirs on the head waters of the Hudson. It would also involve the construction of two additional locks. On the other hand there would be a large saving in the quantity of excava¬ tion. In case this plan be adopted, it would be best to raise the crest of the dam at Fort Miller to an elevation of 118, which would reduce the excavation and not flood the valley of the Hudson at high water. The second plan would involve a greater amount of excavation for the channel, but would save the expense of providing a water supply and the cost of two locks. In this case the water for lockage would DEEP WATERWAYS. 375 be brought from the St. Lawrence River through Lake Champlain and down to the Hudson River on a continuous down grade, requiring but little expense over the cost of constructing the canal. The crest of the dam at Northumberland would be cut down to the level of Lake Champlain—elevation 100. Ships would be saved the time of locking at Whitehall and at Fort Miller. A preliminary estimate indicates that the cost of the two plans is about the same; and since the second plan gives a more certain and unlimited water supply, as well as a channel that can be more quickly navigated, it was decided to base the estimates on the low-level cut. The alignment is shown on plates 51 and 52 and is the same for both the 30-foot and 21-foot channels. Rock appears above grade at White¬ hall for a distance of about 1,500 feet, dips below grade at station 1972, and appears again at station 2290, 6.3 miles south of Whitehall, and continues above grade to station 2523, just south of Fort Ann. All of this rock is a hard quartzite, irregularly stratified, and dips to the east. It contains many vertical cracks or seams and is hard to drill, but breaks well when blasted. Between Fort Ann and Fort Edward rock is found above grade in several places, and is a hard shale or slate when covered with earth or water, but breaks up into flakes and splinters when excavated and exposed to the air and eventually becomes a clay soil; but when exposed to the air in its natural bed, it decidedly, though slowly, disintegrates. It breaks up well by blasting and is easily drilled. The material overlying the bed rock is generally soft clay mixed with sand, but pockets of pure sand are found just north of Fort Ann and between Whitehall and Comstocks. No bowlders are found, and but very little gravel mixed with the earth. In fact, it is good material to excavate with a hydraulic dredge, though the lift may be too great in the deepest cuttings for this method. The streams crossed are Wood Cieek, Granville River, Halfway Creek, and Bond Creek. On leaving Lake Champlain, the channel enters the valley of Wood Creek, which is a sluggish stream winding through a low valley, and follows it with frequent crossings for 16.5 miles to station 2830. One-half mile north of this point it should be taken into the canal. Its elevation is 138 feet, or 38 feet above the proposed water level. The Granville River is crossed 1.7 miles south of Whitehall and is 7 feet above the canal. Halfway Creek is crossed just south of Fort Ann, halfway between Whitehall and Fort Edward, and is 28 feet above the canal. Separate gaugings of these streams could not be had, but they all enter the lake through one channel and have a combined low-water flow of 300 cubic feet per second and a high-water flow of 7,000 cubic feet per second. These streams can be taken into the canal l> 3 r letting them down over a dam to the level of the canal and constructing a basin of sufficient cross section to let them enter at a low velocity. 376 DEEP WATERWAYS. By cutting a diversion channel about 1,200 feet long Wood Creek can be taken into the canal on a solid rock foundation at station 2855. It will be seen on plate 52 that the Champlain Canal enters Wood Creek at Fort Ann and the two streams then become one and the same to station 2260, a distance of 5.2 miles. They pass through the nar¬ row gorge about 4 miles, from station 2300 to station 2516, and it is evident that the canal can not be maintained while the new channel is being excavated if the work is done in the dry. In fact, the only way to do this part of the work in the dry is to construct a dam across the head of the gorge and excavate a channel along the line of the canal to carry the waters of Halfway Creek and Wood Creek to the Hudson River at Fort Edward. In case this part of the work is exca¬ vated with hydraulic dredges, these streams would furnish a water supply for pumping. There are also several small streams crossed on this reach of the canal, but they can all be cared for at small expense and need not be considered in detail. It will be necessary to change the alignment of the Delaware and Hudson Canal Company Railroad at the following points: 184“. 2127. 2337. 2471). 2525. 2040. From station— Total . To sta¬ tion— Dis¬ tance. Feet. 11*11 6,400 2191) 6,300 23)32 2,500 2507 3. 700 2557 3, 200 2730 9,000 a 31,100 a Equals 5.7 miles. I Near station 3110 it is proposed to build a guard lock and by-pass to serve the double purpose of regulating the low-water flow from the lake to the Hudson River and the high-water flow from the river to the lake. The lock could be located at any point near the Hudson, but the site is selected on account of affording rock foundation. FORT EDWARD TO STATE DAM AT TROY. This part of the canal is located generally in the bed of the Hudson River, but makes such cut-offs at the bends as are necessary to give good alignment and may more properly be called canalized river. It involves hydraulic problems somewhat different from those met with on the other parts of the lines, in that navigation must be main¬ tained during the times of extreme low and extreme high water, the range of which is great. For the purpose of showing the conditions which govern the high and low water in the Hudson River a tabulated statement is given of the dams existing across the river, together with the distance above the State dam at Troy, the average elevation of crest, and, as near as can be ascertained, the elevation of high and low water at several points. DEEP WATERWAYS 377 Table No. 11. fl - £ .d a © _be O be -P -4—' o3 rfl ri i-< © cS a ^ • ®-s * > O EH © 'd < 4-1 0 A © © © ©<*_, beo ci be 3 < 4-1 c ition c water ition ( water Remarks. > o be fl © © > © -p c2 > © > © bed fl c8 c3 < W W « K Miles. Feet. State dam at Troy 0.0 1,100 13.5 Mar. 1,1896 23.4 14.2 9.2 High water caused by ice gorge; dam submerged. Do . 1861 28.4 High water caused bv ice gorge. Infor- mation from McEl- roy’s notes. 3 miles above Troy Do . 3.0 1861 30.0 Do. 3.0 1894 29.6 15.4 14.2 High water caused by ice gorge. Au- thority, Waterford bridge tender. 4.9 miles above 4.9 1894 33.5 16.8 16.7 High water caused bv ice gorge. Au- Troy. 9.0 1869 45.1 31.6 13.5 thority- Frank W. Van Fleck. Authority, Mr. Has- broock. Mark on 9 miles above Troy willow tree. Hudson River 9.4 708 48.0 Apr. 19,1896 54.9 48.9 6.0 High water com- Power Trans¬ mission Co. ’s puted. Dam built 1897-98. Assumed dam. part of water is passing through sluices. The Duncan Co.’s dam. Just below Still- 11.5 794 64.5 .do. 1850 72.7 83.9 65.3 76.8 7.4 7.1 Authority, The Dun¬ can Co.’s gauge readings. Caused by ice gorge. Authority, McEl- water dam. roy’s notes. Stillwater dam_ 13.9 820 83.6 Apr. 19.1896 91.5 84.0 7.5 High water com- puted. Angular alignment and ir- regular crest. 0.7 miles above 14.6 .do. 93.2 84.1 9.1 Authority. Win. To- Stillwater dam. ban. Stillwater, N. 0.15 miles above 14.05 1844 92.6 1 . Caused by ice gorge. Authority. McEl Stillwater dam. roy's notes. 2.7 miles above 16.6 Apr. 19.1896 93.5 84.1 9.4 Caused by ice gorge. Authority. W. N. Stillwater dam. Hill.BemisHeights. N. Y. 4.4 miles above 18.3 .do. 95.3 84.2 11.1 Caused by ice gorge. Authority, Mrs. Jo- Stillwater dam. seph Holmes, Bemis Heights, N. Y. 6.1 miles above 20.0 Apr. 26,1895 96.2 84.2 12.0 Caused by ice gorge. Authority.(4 A.En- Stillwater dam. 10.4 miles above 24.3 1896 98.3 84.4 13.9 sign, C. E. Mark, and photograph. Nail in tree by. Stillwater dam. Schuyl erville Bridge. 26.3 1896 99.2 84.6 14.6 Authorities, Abra¬ ham De Riddle.east end bridge: F. B. Pannock, west end bridge: also, photo¬ graph. All agree. State dam at 28.1 810 102.5 1896 110.3 103.8 7.0 . High water com- N ortbumberland puted. Irregular crest and angle in dam. Do. 28.1 111.2 Authority, McEl- roy’s notes. 0.4 miles above 28.5 Apr. 19,1896 112.2 103.5 8.7 Authority, Warner, C. E.; also, photo- State dam at Northumber¬ land. graph by him. Do. 28. 5 111.4 Authority, McEl- roy’s notes. 378 DEEP WATERWAYS. Table No. 11—Continued. Location. Just above Fort Miller dam. east side. Fort Miller dam.. 1.6 miles above Fort Miller dam. Do. 2 miles above Fort Miller dam. 3.3 miles above Fort Miller dam. 5.5 miles above Fort Miller dam. Do 7.3 miles above Fort Miller dam. East end Fort Ed ward railroad bridge. At Glens Falls Paper mills. Glens Falls Pa¬ per Co.'s dam at Fort Edward. p EH < Miles. 31.0 31.0 33.6 32.6 33.0 34.3 36.5 36.5 38.3 38.5 38.9 39.0 Length above dam. Average elevation of crest. Feet. 710 115.1 . 588 140.3 Date of high water. Elevation of high water. Elevation of low water. Range between high and low water. 122.6 1896 122 .5 115.9 6.6 1896 127.0 116.3 10.7 1869 129.0 116.3 12.7 131.2 116.5 14.7 1896 127.9 116.5 11.4 1896 132.0 116.9 15.1 1869 135.0 116.9 18.1 1896 132.7 117.6 15.1 134.8 1896 132.9 120.0 12.9 Apr. 19,1896 148.56 140.7 7.86 Remarks. Authority, McEl- roy’s notes. High water com¬ puted. Authority. Marlow Dickinson. Ap¬ proximate mark. ‘ Do. Authority, McEl- roy’s notes. Authority. Seth W. Bristol. Authority, Geo. P. Cook. Correct within a tenth or two. Authority. Geo. P. Cook. Definite mark. Name not taken. Authority, McEl- roy’s notes. Authority, superin¬ tendent of paper mills. Definite. High water from pa- per company’s gauge readings, Jan.l, 1896, to date Note 1.— The distances are measured along the center line of the proposed channel from the Troy dam to Fort Edward, and not along the river channel as it now exists. Note 2.—All elevations are corrected to refer to Greenbush bench mark, elevation 14.73 above mean tide at New York. It is evident that tlie dams control the elevation of the high water at the points where they are located, and the elevation at other points is the slope in the river from these points to the dam below, plus the depths over the dams, so that the range between the low and the high water is greater at all other points than at the dams. Since the control of the Hood waters is an important factor in designing the channel, it might be well to inquire as to the period that such data cover. Samuel McElroy, civil engineer, made a survey of the Hudson River from Fort Edward to Troy, in 1866, for the State of New York. In his report, dated January 1, 1807, he gave the elevation of the high and low waters at several points, and states that they are the ‘"greatest known in one hundred years.” II owever, many of them must be eliminated on account of the changes made in the river since that time. New dams have been built and the crests of the old ones have been changed, so that the elevation of the water then and now would not in many places be the same for a given volume of flow. The high waters noted by him at DEEP WATERWAYS. 379 Scliuylerville, Crockers Reefs, and Fort Edward would probably be about the same now for the same volume of flow. Since this survey was made two notable high waters have occurred—April, 1869, and April 19, 1896—the former being from 1^- to 3 feet higher than the latter. The high-water marks given by Mr. McElroy are lower than those of 1869 and higher than those of 1896. It is probably safe to say that these records cover a period of at least sixty years. The high waters in the vicinity of Troy have been caused by ice gorges below the State dam, and do not, therefore, represent normal conditions. These floods rise rapidly in two to three days, and recede in about the same time. The normal low-water flow and low-water level are difficult to deter¬ mine under present conditions. Water wheels are installed at the dams with a much greater capacity than the low-water flow of the river, and as a result the water in the pools above the dams is fre¬ quently drawn down from 1 to 2 feet below the crest of the dam. If the wheels stop at such times, there will be no flow below the dam until the pool above is filled to a level above the crest, thus giving a period of no flow in the river at that point, and if tlie wheels start up again the flow will be greater than the normal supply to the river. Under these conditions the low-water level may vary 2 feet or more while the normal supply remains constant. Taking the list of dams given in the above table and also the several dams above Fort Edward, it will be seen how difficult it would be to select a time for making measurements when the river was discharging its normal low-water volume at any given point. However, it is not important for the purposes of this report to know the exact low-water flow except that it shows there will be little or no slope in the proposed channel at low-water stages. There would be a very flat slope when the discharge is double the extreme low-water flow. This condition fixes the grade of the proposed channel as level from one dam to another to give the required low-water depth throughout. For the flood stages of the river it is assumed that the mean veloc¬ ity should not exceed 4 feet per second to afford safe navigation and protection to the banks of the channel. It is also desirable to limit at all points the fluctuations between low and high waters to a mini¬ mum. Starting with a low-water level at Northumberland and enter¬ ing the river below the Troy dam at tide level, there is a difference of level of 100 feet to.be overcome by locks and dams. If the present dams can be maintained at their present elevations, it is evident that the industries depending on the power at the dam sites will be disturbed the least possible amount. Fortunately the locations and elevations of the dams are such that this general plan can be followed except for the one at Fort Miller. This must be entirely taken out in order to carry the level of Lake Champlain to Northumberland, where the first dam and lock would 380 DEEP WATERWAYS. be put in after leaving the lake. At each of the dams below this point it is proposed to have a lock and to maintain dams at about the pres¬ ent elevation. It is desirable to build dams with as great a length of crest as possible, so as to limit the range between low and high waters in the levels above them. The cross section of the proposed waterway must be of a size to not only afford good navigation, but to carry the flood waters with a mean velocity not greater than 4 feet per second, or 2.73 miles per hour. According to Table No. 11, the flood of 1869 is the greatest recorded, and is such as may not occur oftener than once or twice in a century, and it would be better to let the current exceed this velocity when it occurs than to enlarge the channel. The flood of 1896 is such as may occur, say, once in fifteen years, and the channel should be designed to take care of it without damage to the banks or to navigation. The Duncan Company has gauge readings of the depth of water over the crest of its dam from 1886 to date, and the Glens Falls Paper Company has readings from January, 1896, to date. They show that the ordinary spring flood is about 50 per cent of the maximum floods; also that the maximum floods do not remain above the ordinary for a period of more than from four to six days and are not at the extreme stage longer than from one to two days. The above data give the height of the water at various points, but it is also important to know the volume of flow in order to determine the height to which it would rise on the new dams, the slope in the channel, and the velocity of the current. For this purpose the flood of April, 1896, will be taken as one which the channel should carry with a mean velocity not exceeding 4 feet per second, and for any flood greater than this the velocity would be allowed to increase. It would obviously be better to repair any damage done and to suffer any delay to navigation than to make the expenditure for construct¬ ing a channel of the required larger dimensions, especially as the only damage that may be done will be to delay navigation for from two to four days. The evidence points to the conclusion that there has occurred only one flood in the past one hundred years as large as that of 1869. From the gauge readings on the dams at Fort Edward and at Mechanicville the volume of flow may be computed for April 19, 1896. The crest of the former dam is 588 feet long and the depth of water over it was 8.16 feet, while the crest of the latter dam is 794 feet long and the water over it was 8.33 feet. The usual formula for computing the flow over dams or weirs is that- deduced from the Lowell experiments by James B. Francis, and is— Q = C L li § in which, Q = discharge in cubic feet per second. L = length of dam in feet. h = depth of water over dam. C = a coefficient. DEEP WATERWAYS. 881 The value of C varies through a wide range for different depths of water and for different cross sections of the dam or weir used. Most experiments have been with small volumes of water and with a depth over the weir of less than 2 feet. The most complete experiments of this character were made by the eminent French engineer, M. Bazin. These experiments were extended with large volumes of water and depth over the crest as great as 6 feet, under the direction of your honorable Board at the Cornell Hydraulic Laboratory, and are fully reported on in Appendix No. 16. While none of the forms used have exactly the same cross section as the Fort Edward dam, Cornell experi¬ ment No. 6 comes the nearest to it and gives 3.63 as the value of C for a depth of 6 feet. Taking into account the roughness of the structure, it would be conservative to take C = 3.5 for a depth of 8.16 feet on the Fort Edward dam, which would give a discharge of 47,700 cubic feet per second on April 19, 1896. The section of dam used in Cornell experiment No. 18 is the same as the Duncan Company’s dam at Mechanicville, except the latter has an apron on the downstream side, which would have the effect, if any, of increasing the value of C. In this experiment 3.44 was obtained as the value of C for a depth of water of 6 feet. Using this coefficient for a depth of 8.33 feet on the Duncan dam, we get a discharge of 65,800 cubic feet per second for April 19, 1896. Having fixed the volume of flow at these two points, it then becomes necessary to determine, as near as may be, the flow at the proposed location of other dams, to compute the flood heights over their crests, and the slopes and velocities in the new channel when constructed. Between Fort Edward and Northumberland, Moses Kill and Snook Kill enter the Hudson, and the additional watershed will furnish also a small amount of water. The flow at Northumberland is taken as 49,000 cubic feet per second. Batten Kill, Fish Creek, and other small streams are assumed to increase this to 52,000 at Stillwater. Just below Stillwater the Iloosick River enters and increases the flow to 65,800 at the Duncan dam. The distribution of this increased flow is probably not correct, but it is near enough to the truth for the purpose of this report. No other important streams enter the Hudson until the Mohawk is reached just above Troy. In fixing the size of the cross sections it was assumed that 17,000 cubic feet per second would be turned toward Lake Champlain through the by-pass at the Fort Edward guard lock. In this way the volume to be provided for just below Fort Edward is reduced to 32,000 cubic feet per second, and at the other points mentioned by a like amount. It would probably not be necessary to pass the full 17,000 cubic feet per second toward Lake Champlain for more than one day, and three days would likely be the longest period that it would be necessary to divert any flow in that direction during any one flood. Table No. 12 shows the location, elevation, and length of crest of the proposed dams, together with the regulated high-water discharge and the eleva- 382 DEEP WATERWAYS. tions of high water above and below the dams as they will occur after the channel is constructed. The new dams are to be of the ogee sec¬ tion, and 4 feet is taken as the value of C when the depths are 6 for computing the discharge over them. Bazin’s experiment No. 193 gives C 4.55 for a head of 1.26 feet on this form of dam. Cornell experiment No. 19 gives C = 3.675 for a head of 6 feet, but this is not the ogee form. For computing the flood depths over crest, C= 4.01 is used. In computing the slopes and velocities in the channel, the formula V = G \]rs is used. Cisderived from Kutter’s formula by assigning a value of 0.30 to n. Dam No. 1 . — Located at Northumberland. The old dam is to be taken out and replaced by a concrete dam of the ogee section. Dam No. 2 .— Located at Stillwater, and is to be of concrete and ogee section. Dam No. 3 .— The Duncan Company’s dam at Mechanicville. It is to be lengthened from 794 feet to 860 feet, but otherwise unchanged. Dam No. If .— The Hudson River Power Transmission Company’s dam, 2 miles below Mechanicville. It is of the ogee section, 708 feet long, and built in 1897. It is to remain unchanged. Under the flood heads about 14,000 cubic foot-seconds flow through the 16 sluices and the water wheels, leaving 51,800 cubic foot-seconds to flow over the crest of the dam. Dam No. 5 .— Located 1 mile above Waterford, and is to be a new dam of the ogee section. Dam No. 6. — The State dam at Troy, which has an elevation of 13.5, but the low-water level is raised to 15 by the use of flash boards. No changes are required in it. Table No. 12.— Dams. CHAMPLAIN ROUTE, HUDSON RIVER DIVISION. [Thirty-foot channel.] Dam No. Loca¬ tion sta¬ tion. Crest. Elevation of water sur¬ faces. Estimated discharge, Length. Eleva- Above dam. Below dam. cubic feet per sec- Remarks. tion. High. Low. High. Low. ond. 1 . 3720 800 100 104.6 100 90.4 85 32,000 O 4473 900 85 89.6 85 71.6 65 35,000 3_ 4.582 860 65 71.5 65 53.6 48 48,800 4.... 4710 708 48 53.3 48 35.9 30 34,800 14,000 through sluices and 5. 4920 1,000 30 35.3 30 a23.3 15 48,800 turbines. 6 . 5208 1,100 15 22.5 15 0 High water estimated. a The high-water elevations given at dams Nos. 2 and 5 are below the locks, instead of immedi¬ ately below the dam. Note.— Elevation high-water station 3197=106.9. The observed high waters which have occurred at Troy have been caused by ice gorges below the dam, and the high waters which have occurred independent of ice gorges are affected by the backwater, so that no reliable data exist from which to compute the flood discharge DEEP WATERWAYS. 383 of the river at this point. If the channel is deepened, below the dam to 30 feet and 300 feet wide, it is not probable that these gorges will occur, and, in addition to this, the height of the backwater will be reduced. The flood of 1896 gave a depth of 9.9 feet over the crest of the dam. It is probable that this would not be greater than 9 feet under the new conditions, giving an elevation of 22.5 feet when the dashboards are down. This is the elevation assumed in the above table. Locks are located near each of the above dams. They are all single locks of the standard design, with rock foundations. The locations of the locks and also of the dams are the same for both 21-foot and 30-foot channels. Continuing the numbering as brought from Lake Ontario down the St. Lawrence River and to Lake Champlain, the first one on this division would be No. 6. The numbers continue succes¬ sively downstream. Table No. 13 gives a statement of the locks to be built. Table No. 13 .—Locks. No. Station.a Length of level above, in miles. Elevati( water s Above. >n, low- urface. Below. Lift. Remarks. 6 .. 3112 127.4 100 100 Guard lock at Fort Edward. 7... 3727 11.5 100 85 i5 At Northumberland. 8 .. 4488 14.4 85 65 20 At Stillwater. 9. . 4585 1.75 65 48 17 Just above Mechanicville. 10 . 4722 2.43 48 30 18 2 miles below Mechanicville. 11 . 4999 3.72 30 15 15 1 mile above Waterford. 12 . 5215 4.09 15 0 b 15 At Troy dam. COST, c Lock No. Thirty-foot jTwenty-one- channel. foot channel. Operating machinery. 6 . $777,799 1,1X12,988 1,054,289 1,024,458 1,041,229 976, 674 1,001,658 $453,116 632,554 660,829 046.329 656.329 617,229 630,629 $100,000 100,000 100,000 100,000 100,000 100,000 100,000 7.......... 8 . 9 ... . 10 .::;.:::..;::;:::::;;:.;:;;:_::::::. ii. 12 . Total. 6,879,095 700,000 4,297,015 700,000 700,000 Operating machinery______ Total... 7,579,095 4,997,015 a The stationing given above is at the middle of the lock. Water power for driving the oper¬ ating machinery can be developed at the site of each dam. b The lift of this lock was made 15 feet to provide for a minimum low water, c The cost is for the lock structure complete, except excavation. The grades of the 21-foot and 30-foot channels are, respectively, 21 feet and 30 feet below the water surface given in the above table. ALTERNATE PLAN. While the tabulated estimates in this report are based on fixed dams with height of crest the same as that of the proposed low-water level, a part of the line could be treated in a different way at a great 384 DEEP WATERWAYS. saving in cost of construction. The elevation of the Troy dam at present is 13.5, and the low-water level is raised to 15 by the use of dashboards. By substituting for them a movable form of dam this could be increased to 18 feet without injury to any existing industries along the river. At times of high water the movable part of the dam could be lowered so as to keep the flood stage at or below the present high- water level. This would require 3 feet less cutting back to lock No. 11. Similarly, dam No. 5 could be raised 4 feet, dam No. 4, 4 feet, and dam No. 3, 5 feet without decreasing the head now existing at each of them. In fact, the available head would be greater than at present. In addition to this, the fixed crest could be placed lower than at present, and thus decrease the higli-water level. Dam No. 2 could not be raised without decreasing the head at dam No. 1. Dam No. 1 could not be raised, for the reason that its level is fixed with reference to the level of Lake Champlain; and the guard locks at Fort Edward could not be used to hold the waters back, since the supply is needed from Lake Champlain for lockage. However, it may appear cheaper to raise dam No. 2 and pay the damage due to the decreased head at dam No. 1. To raise dams Nos. 3, 4, 5, and 6 as indicated above would make a saving for the 30-foot channel of approximately $5,230,000 in exca¬ vation and lock and retaining-wall construction. From this must be deducted the increased cost of constructing the dams, $40,000, making a net saving of $5,190,000. For the 21-foot channel the net saving would be $4,774,000, less $40,000, equals $4,734,000. These changes would also change the lifts of locks Nos. 8, 9, 10, 11, and 12 for both the 30-foot and the 21-foot channels, shown in the following table: Table No. 14. — Locks for alternate plan. Lock No. Station loca¬ tion. Elevation of low water Lift of lock. Remarks. Above. Below. 8 . 4486 85 70 15 At Stillwater. 4585 70 52 18 Just above Mechanicville. ID. 4722 52 34 18 2 miles below Mechanicville. 11 . 4999 54 18 16 1 mile above Waterford. 12 ... 5215 18 00 18 At Troy dam. If dam No. 2 be raised 3 feet, the lift of lock No. 7 would be changed from 15 to 12 feet and lock No. 8 from 15 to 18 feet. The net saving, exclusive of damage to water power, would be $6,837,000, minus the increase in cost of dams, for the 30-foot channel, and $6,072,000, minus the increase in cost of dams, for the 21-foot channel. This alternate plan seems worthy of consideration. DEEP WATERWAYS. 385 STABILITY OF THE CHANNEL. The upper Hudson does not carry a large amount of sediment, and there is but little evidence of caving banks or “silting up.” A com¬ parison of the profiles made from the survey of Mr. McElroy in 1866 with those made from the survey in 1898 (thirty-two years later) shows that there have been only slight changes in the bed of the stream in that time, although there have occurred in this interval the extreme high waters of 1869 and 1896. Since the high-water slope and the velocity of the currents will be less in the new channel than in the present river, we may safely conclude that the channel will be stable and require but little excavation to maintain the navigable depth. The material to be excavated is classified under the heads of rock and earth, there being no hardpan on this line. The rock is a slate or shale, in irregular layers which dip sharply to the east, firm and hard when in its original bed, but disintegrates rapidly when broken up and exposed to the air. It is more easily drilled than limestone, and breaks up readily under the blast. The earth from Northumber¬ land to Stillwater, a distance of 14 miles, is an alluvial deposit com¬ posed principally of soft clay and sand, while that on the rest of the line is somewhat harder, but composed of the same materials. All of it can be easily excavated with hydraulic dredges. - For the entire length of the line from Port Henry to Troy there is no hardpan and very little gravel or bowlders, making the earth of a class that is easy to excavate. The estimate of quantities of excavation is based on the standard sections of channel adopted by your honorable Board, except for Lake Champlain, where the side slopes are made 1 on 3 instead of 1 on 2, as in other cases, on account of the soft material through which the cut is made. As already stated, the widths of channel adopted are: From Port Henry to Putnam station, 600 feet; from Putnam station to 1 mile north of Whitehall, 450 feet; from 1 mile north of Whitehall to Whitehall, 300 feet, except in the cut through the “Elbow,” where it is of standard canal section; in Whitehall from station 1957to station 1971 is 200 feet wide; from Whitehall to tlie Hudson River at Fort Edward, standard canal section; from Fort Edward to Northumber¬ land the 30-foot channel is of the standard canal section, but the 21-foot channel is made 300 feet wide to carry the flood waters of the river; from Northumberland to Troy both channels are made 300 feet wide. BRIDGES. Tables Nos. 15 and 16 give certain data in regard to the bridges now existing across Lake Champlain and the Hudson River; also the railway and highway crossings which are encountered between White¬ hall and Fort Edward. H. Doc. 149-25 DEEP WATERWAYS 386 Table No. 15.— Existing bridges, Champlain route, Hudson River division. Location. CM 0 «w O rP +-> . hft © rP . H-5 3 798 31 37.4 Wooden arched truss and elec- bridge, built in trie rail- 1803. Toll bridge road. owned by private Lansingburg . 5116 .do. Swing.. 2 5 822 37 39.3 Draw span 129 feet long C. to C. end pins. 3 through Warren trusses. Toll bridge owned by private com- pany. DEEP WATERWAYS. 387 Table No. 16. —Existing highway crossings, Champlain route, Hudson River division. Station. Place. Remarks. 2237. Not much traveled. Of considerable importance. Do. Important crossing. Not very much used. Important. Not much used. Do. Do. Important. Do. Considerably used. Do. Important. 2302. 2345. Comstocks P. O. 2467. Flat Rock__ 2553. 2640. Fort Ann....... 2761. Smiths Basin 2872. 2907. 2940. 3002... Dunhams Basin.. 3137. Fort Edward...... 3166_ 3179. _do... .do... 3187. .do.. Some of these highway crossings may be consolidated without mate¬ rial injury to the traveling public, which would be through an agree¬ ment with the parties interested. At the present time an electric railway is being built along the valley of the Hudson River from Stillwater to Northumberland, where it crosses the river below the State dam. This railway can be taken across the canal on the bridge to be built at station 3756. For the purpose of determining the num¬ ber and character of bridges which must be built or reconstructed it is assumed that they will be needed at the locations given in Table No. 17. Table No. 17.— Location, cost, etc., of proposed bridges, Hudson River division, , Champlain route. Sta¬ tion. Location. Kind of bridge. Num¬ ber of tracks. Swing or fixed. Num¬ ber of spans. Addison June- Railway_ 1 Swing.. _do_ 4 1968 tion. Whitehall. Highway. i 1987 .do. _do. _do. i 1994 _do .. Railway_ 1 _do_ i 2299 Comstocks_ Highway. _do... i 2467 Flat Rock. _do.. _do.. i 2553 Fort Ann. .do. ... do.. _ i 2646 .do. ... do.. i 2762 Smiths Basin. .do. _do.. i 2857 .do. _do... i 3(KX> Dunhams Ba- .do_ _do_ i 3112 sin. Fort Edward . .do. _do... i 3180 .do. __ do. _do.. i 3757 Clarks Mills .. Highway and electric rail- 1 _do... i Stillwater . way. Highway. .. .do. i 4554 Mechanic ville Railway_ 2 i 4618 .do. Highway .... ....do i 4721 _~do . . _do... i 5051 Waterford.... Highway and electric rail- 2 —do... i 5116 Troy (Twelth street). Clarks Mills .. way. .do. 2 _do.. *> 3757 Highway and electric rail- i Fixed .. 3 way.a Total cost. Thirty-foot channel. Twenty-one-foot channel. Total length. Esti¬ mated cost. Total length. Esti¬ mated cost. 9771 $174,676 957* $174,676 475 50,002 475 50,000 545 105,430 525 89,604 537) 161,810 517* 133,496 545 73,338 525 71.894 603 75, 456 579 70,218 545 112,882 525 96,186 545 88,264 525 91,912 545 118,708 525 102,144 565 80,387 541 75,665 545 123,976 525 109,914 255 18,633 255 18,633 545 116,850 525 98,608 565 71,533 541 66,504 565 68,769 541 64,800 550 203,383 5: (0 191,093 545 84,774 525 83,304 235 19.986 195 16,650 550 128,208 530 120,774 650 145,960 630 138,536 450 19,330 450 19,330 2,042,355 1,881,051 a Bridge not over canal. Feet clear opening. Highway bridges. . 22 Single track railway bridges... 14 Double-track railway bridges ..... 26 388 DEEP WATERWAYS. RIGHT OF WAY. Without specifying in detail, the right of way estimated is intended to be sufficient for the prism of the channel and for spoiling the mate¬ rial excavated by any method that is suitable for the work. Through towns it is narrowed up to small limits, but in the farming districts a generous width is provided. The estimates of quantities and cost of all work, including right of way, are given in Tables Nos. 18 and 19. Table No. 18.— Champlain route, Hudson River division. [Estimate, 30-foot channel.] Quantity. Cost per unit. Total. Lake Champlain (station 0 to station 1957). Excavation: Earth, wet. Rock, quartzite, dry Rock, quartzite, wet Rock, shale, wet. Right of way: Village property .... Farm property. Railroad changes.. Bridge. cubic yards.. ..do_ ..do_ .do_ 47,717,339 409,004 45,983 26,100 acres., .do_ number.. 120 192 1 Total. Whitehall to Fort Edward (station 1957 to station 3197). $ 0.12 .75 2.50 2.00 $5,726,081 306, 798 114,958 52,200 33,000 5,050 61,500 174,676 6,474,263 Excavation: Earth, dry. Rock, dry, quartzite Rock, dry, shale. Retaining walls, etc.: Retaining wall. Slope wall. Back fill. Timber cribs: Pine... Hemlock. Oak.. Iron... Stone fill. Right of way: Village property ... Farm property. Railroad changes. Bridges. Entrance of streams_ Gates for by-pass.. Lock No. 6... Operating machinery.. Total. ..cubic yards. .do_ .. do_ .do ... square yards.. ..cubic yards.. .feet B. M-. .do_ .do_ ..._pounds.. ..cubic yards.. .acres.. ..do_ number.. Hudson River (station 3197 to station 5235). 79,839,083 6,569,002 2,179,083 93,865 506,290 353,970 .15 .75 .60 4.00 1.10 .25 932,865 3,973,940 53,280 423,582 65,620 o 30.00 a 23.00 a 50.00 .03 .60 11,975.862 4,920,752 1,307,450 375,460 556,919 88,493 27,986 91,401 2,664 12,707 39,372 477 5,594 12 782, .500 559,400 63,000 1,125,736 55,010 52,115 777,799 100,000 22,920,626 Excavation: Earth, dry.. Earth, wet. Rock, dry. Rock, wet. Walls, etc: Retaining walls Slope walls. Back fill. Crib walls: Pine. Hemlock. Oak. Iron... Stone fill. ..cubic yards.. .do_ .do_ .do_ .do_ square yards.. ..cubic yards.. .feet B. M.. .do_ .do_ .pounds.. ..cubic yards.. 14,027,058 25,355,910 7,197,489 19,350,931 159,619 380,155 659,9:34 .15 2,104,059 .20 5.071,182 .60 4,318,493 2.00 38.701,862 4.50 718,286 1.45 551,225 .25 164,984 6,564,270 9,440,730 270,640 1,373,766 202,008 a 30.00 a 23.00 a 50.00 .03 .60 196,928 217,137 13,532 41.213 121,205 a Per 1,000 feet. DEEP WATERWAYS 889 Table No. 18. —Champlain route, Hudson River division — Continued. Quantity, Total. Hudson River (station 3197 to station 5235) —Continued. Right of way: Village property. . . . acres 185 5,209 8 4 6 $1,397,700 636.700 741,943 139,568 6,101,296 600,000 118,186 20,000 FarnT property...do .. Bridges ..number.. Dams Nos. 6, 7.8,10.. Locks .. . .. ... Operating machinery. . Entrance of Hudson ... ...... ... Steam ferry ...... Total...... 61,975,499 SUMMARY. Lake Champlain. Whitehall to Fort Edward .. Hudson River. $6,474,263 22,920,626 61,675,466 Total 91,370,388 Table No. 19. —Champlain route, Hudson River division. [Estimate, 21-foot channel.] Lake Champlain (station 0 to station 1957). Quantity. Cost per unit. Total. Excavation: Earth, wet. Rock, quartzite, dry. Rock, quartzite, wet Right of way: Village property. Farm property .. Railroad changes. Bridge. cubic yards . .do_ .. do_ acres., .do_ 18,999,386 334, 716 11,195 120 192 number.. 1 Total. Whitehall to Fort Edward (station 1957 to station 3197). $0.12 | $2,279,926 .75 | 251,037 2.50 I 27,988 33,000 5,050 61,500 174,676 2,833,177 Excavation: Earth, dry.. Rock, dry, quartzite Rock, dry, shale. Walls, etc.: Retaining wall. Slope wall. Back fill..... Timber crib: Pine. Hemlock.. Oak. Iron... Stone fill... Right of way: Village property.... Farm property. Railroad changes. Bridges... Entrance of streams_ Gates for by-pass. Lock No. 6.. Operating machinery... cubic yards.. .do_ .do_ 68,780,484 4,743,531 1,220,480 .15 .75 .60 .do_ square yards.. ..cubic yards.. 62,102 4.00 546,849 1.10 221,877 .25 ...feet B. M.. .do_ .do_ _pounds.. cubic yards.. 943,425 3, (X10,880 53,280 337,557 50,515 a 30.00 a 23.00 a 50.00 .03 .60 .acres. ..do... .number. 477 5,594 12 10,317,073 3,557,648 732,288 248,408 601,534 55,469 28,303 69,020 2,664 10,127 30,309 782,500 559,400 63,000 1,008.276 55,010 43.385 453,116 100,000 Total. Hudson River (station 3197 to station 5235). 18,717,530 Excavation: Earth, dry Earth, wet Rock, dry . Rock, wet . cubic yards.. .do_ .do_ .do_ 10,722,835 18,023,672 4,650,489 12,344,104 .15 .20 .60 2.00 1,608,425 3,604,734 2,790,293 24,688,208 a Per 1,000 feet. 390 DEEP WATERWAYS. Table No. 19 .—Champlain route , Hudson River division —Continued. Quantity. Cost per unit. Total. Hudson River (station .; 19? to station 52S5)—Continued. Walls etc.: Retaining: walls ... Slope walls-— Back fill., Crib walls: Pine. Hemlock. Oak... Iron.. Stone fill. Right of way: Village property . Farm property . -. Bridges. Dams.. Locks... Operating machinery Entrance of Hudson . Steam ferry. Total.. ..cubic yards.. square yards.. ..cubic yards.. .feet B. M.. .do ... .do ... ..pounds. . ..cubic yards . .acres.. ..do_ 61.H0 468,555 367, 788 4.50 1.45 .25 6,463,670 9,116,940 266,320 1,332,656 195,036 a 30.00 a 23.00 a 50.00 .03 .60 185 5.209 8 . 4 6 275,130 ‘>70,405 91,947 193,910 209.690 13,316 39,930 117,022 1,397,700 636,700 700,991 139,568 3,843.899 600.000 118,186 20,000 41,769,104 a Per 1,000 feet. SUMMARY. Lake Champlain. ...$2,833,177 Whitehall to Fort Edward........18,717,530 Hudson River........ 41,769,104 Total... 63,319,811 FLOOD-DISCHARGE MEASUREMENTS OF THE UPPER MOHAWK AND OTHER STREAMS. In accordance with your instructions, on March 1, 2, and 3, 1808, I made a reconnoissanee of the Mohawk River and tributaries from Rome to Little Falls, and of Fish Creek from Lake Oneida to Taberg, with the view of determining at what points approximate Hood meas¬ urements could be easily and cheaply made during the spring of 1898. The time allowed to prepare for and take the observations precluded the use of current meters and similar methods for determining the velocity of the current. As a result of this preliminary study it was decided to take the measurements at points and by methods as follows: 1. The Mohawk River at Little Falls. A masonry dam, locally known as the “Middle dam,” exists across the river. It is about 10 feet high, and all the water not used by the mills passes over its crest. By measuring the depth of water flowing over it and adding to this the amount of water used by the mills the total flow could be com¬ puted at this point. 2. The West Canada Creek at Herkimer. About 2 miles above Herkimer a timber and rock fill dam exists across the stream, and it was decided to use the depth of flow over it for computing the dis¬ charge of the stream, but later it was found that the dam farther up the stream, at Middleville, was better adapted for the purpose and it was used instead of the one at Herkimer. DEEP WATERWAYS. 391 3. The Sauquoit Creek, which empties into the Mohawk at New Hartford about 2 miles above Utica. At this point the measurements were made over the dam, as at Little Falls. 4. The Oriskany Creek, which empties into the Mohawk at Oriskany. A dam was used at this place also. 5. Nine-Mile Creek, which empties into the Mohawk opposite and above Oriskany. A dam was used at this place. 6. The Mohawk River at Rome. Near the city is a masonry dam, but at times of high water it is drowned out and evidently could not be utilized. About 2 miles above Rome, at Ridge Mills, a suitable dam was found over which the greater part of the Hood waters passed, and the rest flowed through two openings under highway bridges. This point was selected for making the measurements of flow in the Mohawk at Rome. 7. Fish Creek and Wood Creek, which empty into Lake Oneida at Sylvan Beach. No dams exist across either of these streams over which the flow could be measured. Wood Creek is a sluggish stream, winding through a low, flat swamp, and at no place are the banks high enough to hold the waters within the prism of the stream, and no measurements of it were undertaken. Just below the junction of the east branch and the west branch of Fish Creek is the only place between this point and Lake Oneida where the ordinary floods are confined within the banks of the stream, and the extreme floods spread over the bottom lands here. This is about 1 mile below the town of Taberg, and about 9 miles above the outlet of the stream. This point was selected for making the measurements, and the velocity of the current was determined by the use of rod floats. The various members of the party arrived at Rome from the Niagara routes on March 8, 9, and 10, and were placed as follows: Little Falls, George F. Anderson; West Canada Creek, P. If. Ash- mead and Curtis Hill; Sauquoit Creek, ,T. W. Jenkins; Oriskany Creek, H. 11. Lotter; Mohawk River at Ridge Mills, H. F. Dose and M. W. Tenny; Nine-Mile Creek, R. McD. Geraty; Fish Creek, II. C. Goodrich and A. A. Conger. The warm weather existing at this time caused the snow to melt rapidly, with the result that the maximum flood stage in the streams was reached March 12. Rains on that night prolonged but did not increase the flood period to the 14th, which gave ample time to gage the depth of water flowing over the dams. The surveys of the various dams were made after the waters receded. The floods of the Mohawk, West Canada Creek, Nine-Mile Creek, and Fish Creek were the high¬ est which had occurred for several years, but were probably consider- ably less than the maximum which has occurred; while the flow in Oriskany Creek and in the Sauquoit Creek was much below the ordi¬ nary spring floods. The middle dam at Little Falls was built in 1892, and this flood was the greatest which has occurred since the dam was built. 392 DEEP WATERWAYS. The discharge of the various streams was not finally worked up, but all notes were turned over to the Water Supply Division, and the results are included in the report of Mr. George W. Rafter, Appendix No. It*. REPORT ON RESULTS OF TWO LINES OF LEVELS RUN FROM THE GREENBUSH BENCH MARK (OPPOSITE ALBANY, N. Y.) TO LAKE ONTARIO. These two lines of levels were run by six different parties and the results of the work are set forth in this report. The first line considered starts at the Lake Survey bench mark at Greenbusli, N. Y., with elevation 14.73 feet 1 above mean tide at New York, and runs up the Hudson River to the mouth of the Mohawk River and follows the valley of the Mohawk to Rome. From Rome it runs to Lake Oneida and down the valleys of the Oneida and Oswego rivers to the Lake Survey bench mark A at Oswego, N. Y. The other line runs from the Greenbusli benchmark, elevation 14.73, up the Hudson River to Fort Edward, and then crosses the divide to Whitehall, at the head of Lake Champlain, and then cont inues down the shore of the lake to the Crown Point light-house. From this point to Cooperville, near Rouse Point, the levels are transferred by water levels. From Cooperville to Cape Vincent, at the foot of Lake Ontario, the levels were run with wye levels and precise levels. The two lines were connected through Lake Ontario by water levels, using Oswego and Cape Vincent as the points for making gauge readings. OSWEGO-MOHAWK ROUTE. All levels on this line were run with Buff & Berger wye levels in opposite directions, and the means of the two runs were used in deter¬ mining the elevations of the bench marks. The line was run by three separate parties and the work of each will lie considered in regular order, beginning at Greenbusli and ending at Oswego, on Lake Ontario. First. Iu August, 1898, levels were run, under the direction of H. F. Dose, assistant engineer, from bench mark 1, at Troy, N. Y., elevation 21.85 feet, to the bench mark on gristmill at Greenbusli, making its elevation 14.50 feet, or 0.23 foot lower than the adopted elevation of 14.73 feet for this bench mark. Therefore we have 21.85 + 0.23 = 22.08, which is the elevation of bench mark 1, at Troy, corresponding to elevation 14.73 of the Greenbusli bench mark. Second. In October, 1897, levels were run, under the direction of D. .T. Howell, assistant engineer, from United States Lake Survey bench mark 6, elevation 49.69, on lock No. 4 of Erie Canal, to bench mark 1, 1 See pages 70 and 71, Report of United States Deep Waterways Commission, 1896. DEEP WATERWAYS. 393 on Congress Street Bridge, at Troy. The elevation of bench mark 1 thus established was 21.85. But the elevation of this bench mark, as determined by Mr. Dose, starting with elevation of 14.73 of the Greenbusli bench mark, is 22.084, a diffeience of 0.23 foot. Since the levels were not run connecting bench mark 1 with the Greenbusli bench mark until August, 1898, Mr. Howell continued his levels from bench mark 1 (using elevation 21.85) up the Mohawk Valley to bench mark 88, at Herkimer, N. Y., making its elevation 389.18 feet. There¬ fore we have 389.18+0.23 = 389.41 feet for the elevation of bench mark 88, corresponding to elevation 14.73 of the Greenbusli bench mark. This line of levels was run in connection with the topographic survey, and was completed in October, 1898. Third. A line of levels was run in connection with the topographic survey, under the direction of A. ,T. Himes, assistant engineer, from United States Lake Survey bench mark A, at Oswego, to bench mark 88, at Herkimer. It was begun in October, 1897, and completed in July, 1898. Starting with elevation 251.9G of bench mark A, an ele¬ vation of 388.52 was obtained for bench mark 88. But, as stated above, the elevation of bench mark 88, referred to 14.73 of the Greenbusli bench mark, is 389.408 feet, a difference of 0.89 foot. Therefore we have 251.90 + 0.89 = 252.85 for the elevation of bench mark A at Oswego. Using elevation 251.96 for bench mark A, United States Assistant Engineer Churchill has determined the elevation of the zero of the United States gauge at Oswego as 244.18 feet. Therefore we have 244.18+0.889 = 245.07 for the elevation of the zero of this gauge re¬ ferred to 14.73 of the Greenbusli bench mark. Readings of the water surface were taken at 7 a. m., 12 noon, and 6 p. m. on this gauge from July 27, 1898, to August 6, 1898, both inclusive. There were no great fluctuations during the period, the extreme at Oswego being from 1.47 to 1.60 feet. A mean of all of these readings gives the water surface as 1.53 feet above the zero of the gauge, or elevation 246.60 feet. CHAMPLAIN ROUTE. The levels on this route were run by several different parties using different methods, and will be considered in regular order from Troy, N. Y., to Lake Ontario. First. Wye levels from Troy, N. Y., to Crown Point light-house. Second. Water levels from Crown Point light-house through Lake Champlain to Kings Bay, at Cooperville, N. Y. Third. Wye levels from Rouse Point bench mark via Cooperville to Hogansburg, on the St. Lawrence River. Fourth. Precise levels from Hogansburg to Cape Vincent, at the foot of Lake Ontario. 394 DEEP WATERWAYS. TROY TO CROWN POINT LIGHT-HOUSE. This line was run in connection with the topographic survey, under the direction of C. L. Harrison, assistant engineer, with Buff & Berger wye level in opposite directions, and the means of the two runs were used in determining the elevations of the bench marks. It was begun in May, 1898, and completed in September, 1898. Starting at Troy from bench mark 1, elevation 21.85, as noted above, and running via Fort Edward and Whitehall to Crown Point light-house, bench mark 231 was established, with elevation 130.585. Therefore we have 130.585 + 0.23 = 130.815 for the elevation of bench mark 231, corresponding to 14.73 of the Greenbush bench mark. Con¬ nection was also made with United States Coast Survey bench mark 30, on lock No. 23 of Champlain Canal, at Whitehall, making its ele¬ vation 104.55 feet, and with United States Coast Survey bench mark 39, at Putnam Station, making its elevation 107.615. WATER LEVELS THROUGH LAKE CHAMPLAIN. The levels were transferred by water through Lake Champlain from Crown Point light-house to Kings Bay at Cooperville in January, 1899, at a time when the entire lake was frozen over from Whitehall to Rouse Point, excepting about 10 miles of the “broad lake,” near Burlington, Vt. Crown Point is at the north end of the narrow and shallow part of the lake, which extends southward to Whitehall. Kings Bay is 0 miles south of Fort Montgomery and near the north end of the wide and deep part of the lake, which extends southward to Crown Point. These two points were selected for making the transfer, as they were not so liable to be affected by slopes due to flow of water through the lake and the piling up of water due to wind action as points situated in the narrow part of the lake. The transfer of the levels was made in the following manner: On the dock at Crown Point light-house and on the bridge pier at Cooperville index points were established. By means of float gauges the distance of the water surface below the index points was measured at ten-minute intervals, beginning at 8.40 p. m., January 13, and ending at 8 p. m., January 15, making a continuous set of readings, covering a period of nearly forty-eight hours. The extreme fluctua¬ tion of the water surface during this time was 0.22 foot at Crown Point gauge and 0.21 foot at Cooperville gauge. The index point of gauge at Crown Point was connected by spirit levels with the bench mark on light-house, and the index point of gauge at Cooperville was connected with bench mark 1 on bridge pier. Aneroid barometers were read hourly at each station, but these readings were not used in working up the final results, because the correction due to the difference of atmospheric pressure at the two points was less than the probable error in the barometers themselves. From these observations we have: DEEP WATERWAYS. 395 Bench mark on Crown Point light house.. .. 130.815 Index point Crown Point gauge. ........ 98.761 Mean distance water surface below index point. 3.128 Elevation water surface at Crown Point... 95.633 Elevation water surface at Cooperville .... 95.633 Elevation index point above mean water surface.. 1.675 Elevation index point of Cooperville gauge.... 97.308 Elevation bench mark 1 above index point... 5.730 Elevation bench mark 1 at Cooperville. 103.038 If the barometer correction, 0.045 foot, be applied, this elevation would become 103.083, but it is neglected for the reasons given above. In August, 1808, a duplicate line of wye levels was run, under the direction of Frank P. Davis, assistant engineer, from the Rouse Point bench mark, using elevation 110.06, as established by the Coast Sur¬ vey in 1882, to Cooperville, and establishing bench mark 1, with ele¬ vation 104.147, which is 1.109 feet higher than obtained above. We have, therefore, 110.06—1.109=108.951 for the new elevation of the Rouse Point bench mark. The elevation 110.06 of the Rouse Point bench mark was transferred in 1882 by water levels through Lake Champlain from Putnam station to Rouse Point. Since the new determination of this is 1.109 feet lower, it might be well to compare the gauge readings taken at Fort Mont¬ gomery and at other points on the lake. From January 1 to January 15 the gauge readings were not taken at Fort Montgomery, but gauges were read at Whitehall and at Putnam station at the same time read¬ ings were taken at Crown Point and Cooperville—January 14 and 15. But on the 15th the flood waters (caused by the rains on the 14th) from Wood Creek and the Granville River increased the slope in the lake from Whitehall to Putnam station about 0.3 foot. Taking the readings on the 14th, when there was very little water flowing into the lake at Whitehall, we have the following results: Elevation water surface at— Feet. Whitehall. 96.01 Putnam station. 95.90 Crown Point... .. 95,68 showing a slope of 0.11 foot from Whitehall to Putnam station and 0.22 foot from Putnam station to Crown Point. The gauge at White¬ hall was set from bench mark 36, elevation 104.55, and the gauge at Putnam was set from bench mark 39, elevation 107.615. The eleva¬ tions of these benches as determined by the Coast Survey are: No. 36, 104.71 feet, and No. 39, 108.46 feet. If these elevations had been used in setting the gauges, then we would have had: Water surface at— Whitehall. Putnam station Feet. 96.17 96.745 DEEP WATERWAYS. 396 making a fall in the water surface from Putnam station to Whitehall of 0.575 foot. This is impossible, as the direction of the flow is from Whitehall toward Putnam station. The difference, then, of 1.109 feet between this (108.951) and the former (110.06) determination of the Rouse Point bench mark seems to be distributed as follows: Difference in levels— Feet. Greenbusli to Whitehall...... 0.160 Whitehall to Putnam station........685 Slope in lake from Putnam to Crown Point.. . 220 Transferring levels across lake.......014 Total..._.-..... 1.109 During the progress of the surveys gauges were established and read at Crown Point village, about 6 miles south of Crown Point light¬ house, and at Whitehall for short periods. In May, 1899, Frank P. Davis, assistant engineer, ran a checked line of levels from Rouse Point bench mark, 108.951, to the United States engineers’ gauge at Fort Montgomery, determining the elevation of its zero as 93.50 feet. Reducing the readings of this gauge to ele¬ vations for periods at which we have readings at other points in the lake, we have: Date. Mean water surface. Differ¬ ence. From— To— Crown Point village. White¬ hall. Fort Mont¬ gomery. Sept. 13, 1898. Dec. 3, 1898 . Oct. 20, 1898 . 94.77 94.68 95.34 95.68 0.09 .34 .42 Dec. 21, 1898 95.68 96.10 Jau. 16. 1899 .. Jan. 27,1899. The readings September 13 to October 20 were taken when the lake was open and affected more or less by the winds, those in December when the lake was partially frozen over, and those in January when the lake was almost entirely frozen over. The slope observed Janu¬ ary 14,1899, at low water, when the lake was frozen over from White¬ hall to Crown Point, was 0.33 foot. If this be applied, we have 0.09 foot for the slope from Crown Point light-house to Fort Montgomery. These gauge readings indicate that there can be no substantial error in the levels from Whitehall to the Rouse Point bench mark. If the old elevations of the bench marks at Whitehall and Rouse Point were used to determine the water surface at these points, there would be a slope at low water of about 0.53 foot from Fort Montgomery to White¬ hall, which is impossible, since the flow is from Whitehall toward Fort Montgomery. COOPERVILLE TO HOGANSBURG. This line was run in connection with the topographic survey under the direction of Frank P. Davis, assistant engineer, with Buff & Berger DEEP WATERWAYS. 397 wye level, from beneli mark 1, elevation 104.147, at Cooperville, via Champlain and Valleyfield to United States permanent bench mark P, on Catholic church at Hogansburg, making its elevation 180.302. But since the elevation used of bench mark 1 was 1.109 feet too high, we have 180.392 —1.109 = 179.283 for the elevation of permanent bench mark P, at Hogansburg. These levels were begun in August, 1898, and completed in October, 1898. HOGANSBURG TO CAPE VINCENT. This is a duplicate line of precise levels run in the fall of 1898 and the spring of 1899 by D. A. Molitor, United States assistant engineer, using a Buff & Berger precise level. Starting at United States per¬ manent bench mark P, at Hogansburg, elevation 179.283, and running along the St. Lawrence River to Cape Vincent, the zero of the gauge at that point was determined as 248.803. This gauge was read every ten minutes from 6 a. m. to 6 p. m. July 27 to August 6, 1898, and the mean of the readings for this time was 2.318 feet below the zero of the gauge. Therefore we have 248.803 — 2.318 = 246.485 for the elevation of the mean water surface in Lake Ontario at Cape Vincent July 27 to August 6, 1898, which is 0.114 foot lower than as deter¬ mined for the same dates at Oswego via the Mohawk route. Notable elevations, in feet, above mean tide at New York. Oswego-Mokawk route. Feet. Champlain route. Greenbusli bench mark . 14. 730 Greenbush bench mark. 14. 730 Troy bench mark 1. 22. 084 Troy bench mark 1. . .... 22.084 Herkimer bench mark 75 389. 408 Whitehall bench mark 30... 104.550 Oswego bench mark A. .. 252.853 Putnam station Dench mark 39.. 1< >7.615 Oswego zero of gauge. .. Oswego mean water surface July 27 245.009 Zero of Fort Montgomery gauge.. Rouse Point bench mark on Chapman 93.500 to August 6, 1898_. 240.599 Block . 108.951 Cooperville bench mark 1.... Hogansburg permanent bench rnarkP. Cape Vincent zero of gauge .. Cape Vincent mean water surface July 27 to August 0, 1898 .... 103.038 179.283 248.803 240.485 Assuming that the mean water surface in Lake Ontario was level from July 27 to August 6, 1898, -we would then have 246.599 — 246.485 = 0.114 for the error of closure for the two lines of levels. The lengths of the several lines, measured along the course actually traversed in this circuit, are as follows: Miles. Bench mark 1, at Troy, to Crown Point (wye levels)... 103. 3 Crown Point to Cooperville (water levels)... 64 Cooperville to Hogansburg (wye levels).... 78.5 Hogansburg to Cape Vincent (precise levels)... 118.3 Cape Vincent to Oswego (water levels)... . 47 Oswego to Herkimer (wye levels)..... 98.5 Herkimer to bench mark 1, at Troy (wye levels) ... 89. 2 Total...-.- 598.8 398 DEEP WATERWAYS. Total wye levels .. Total precise levels Total water levels Miles. 369.5 118.3 111 Total..... 598.8 The distance, 7 miles, from bench mark 1, at Troy, to the Greenbush bench mark is common to both lines. Respectfully submitted. C. L. Harrison, Principal Assistant Engineer. The Board of Engineers on Deep Waterways. Appendix No. 11. CHAMPLAIN ROUTE, ST. LAWRENCE DIVISION. Detroit, Mich., September 30, 1899. Gentlemen: I respectfully submit herewith the following report on surveys of the St. Lawrence River from Ogdensburg to Lake St. Francis, and estimates of the cost of constructing a ship canal of 21 and 30 feet depth. A field party was organized and reported for duty at Ogdensburg, N. Y., August 11, 1898. The base line and stadia field work were commenced on the 17th and soundings on the 22d. The boring party was equipped with a Sullivan boring machine for work on either land or water, and began work on the river September 1. Topography was practically completed December 3, soundings December 14, and borings were discontinued December 20, 1898. Four land parties and one river party resumed borings April 11, 1899. The required field work was completed June 20, 1899. A small office force has been engaged on mapping, computations, and esti¬ mates up to September 20, 1899. The field work was done in accordance with the general instructions to field parties, Appendix No. 9. The field work has required an aggregate of 1,031 days’ work by instrument men, recorders, and rodsmen, 681 days’ work by stadia men and foremen, 276 days’ work by teams with drivers, and 2,046 days’ work by laborers. The office work has required 1,028 days’ work by draftsmen and computers. The principal items of topographical work were 39 miles of base line; 103.6 miles of stadia circuits; exclusive of side lines and such portions of base line as were included in the circuits; considerable work on levels which were dependent upon the bench marks estab¬ lished by Mr. David Molitor, United States assistant engineer, and which have not been reduced to miles run; about 20,900 soundings; 148 borings on land, aggregating 7,052 linear feet, of which 65 reached rock, and 151 borings on water, aggregating 2,123 linear feet, of which 94 reached rock. DEEP WATERWAYS 399 Surveys were not carried over such portions of the river as were shown on existing charts to have a sufficient depth of navigable water. In Lake St. Francis estimates are based on data shown on existing charts and on soundings furnished by Mr. Tom S. Rubidge, superin¬ tending engineer of the St. Lawrence district, Cornwall, Ontario. HIGH AND LOW STAGES OF THE RIVER. In order to determine the relations of fluctuations in the river with fluctuations in lake, and high and low water stages, a comparison of three points in the river is made with the three highest and three low¬ est stages of the lake at Oswego between the years 1865 and 1898. The points selected in the river are the head of the Galops Rapids at Lock No. 27, the head of Lake St. Francis at Lock No. 15, and the foot of Lake St. Francis at Lock No. 14. Gauge readings for Lock No. 27 during the year 1870 are not available; also gauge readings for Lock No. 15 are omitted during the first four months of the year on account of the effects of ice jams. The mean reading for the year and for the period of navigation, May to November, inclusive, are given. Table No. 1 .—Tabulation of gauge readings for the three years. HIGHEST ANNUAL MEAN GAUGE READINGS SINCE 1865. Oswego. Lock No. 27. Lock No. 15. Lock No. 14. 1870. 1884. 1886. 1884. 1886. 1870. 1884. 1886. 1870. 1884. 1886. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. January.. 3.14 2.39 3.48 9.18 11.01 12.83 12.11 13.33 3.29 2 3.55 9.62 10.95 12.98 12 13 13 OH March. 3.29 3.44 3.69 10.33 10.87 12.36 12.58 12.34 April. . 4.23 4.05 4.31 11.51 11.83 13.31 13.22 13.46 M*ay. 4.83 4.07 4.52 11.78 12.36 12.43 11.78 12.14 13.42 13.23 13. 18 June... 4.51 3.97 4.32 11.66 12.13 12. 11 11.57 11.97 13.28 12.88 13.10 July--- 4.19 3.76 3.92 11.66 11.70 11.96 11.57 11.69 13.17 12.82 12.82 August.. 3.85 3.53 3.48 11 39 11.33 11.65 11 .35 11 32 12.94 12. 41 12.57 September_ .. 3.16 3.10 3.12 11.01 10.98 11.15 10.99 11.05 12.54 12. 44 12.15 October.. 2.83 2. Ii8 2.83 10.54 10.62 10.58 10. 63 10. 73 12.44 12. 11 12.20 November__ 2.26 2.18 2.39 10.23 10.49 10.05 10.26 10. 64 12.28 11.86 12.12 December.- 2.01 2.03 2.30 10.00 10.20 11.01 11.65 11.49 12.19 11.98 12.38 Mean readings 3.47 3.16 3.49 10.74 11.21 12.81 12.48 12.73 Mean readings. May-Novem- ber. 3.66 3.33 3.51 11.18 11.37 11.42 11.16 11.36 12.87 12.54 12.59 LOWEST ANNUAL MEAN GAUGE READINGS SINCE 1865. Oswego. Lock No. 27. Lock No. 15. Lock No. 14. 1895. 1896. 1897. 1895. 1896. 1897. 1895. 1896. 1897. 1895. 1896. 1897. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. January .37 —.32 —.24 8.02 7.65 7.78 11.50 11.21 10. 47 February.. .31 + .20 -.30 7.19 7.80 7.61 10.57 11.18 10.35 March .21 .37 + .20 7.46 7.70 8.11 10.68 10.69 10. 78 April. .76 1.30 .84 8.50 8.93 8.94 11.95 11.96 11.58 May. .88 1.31 1.28 8.83 9.32 9.21 9.00 9.83 9.89 11.46 11.88 11.63 June. .76 1.22 1.50 8.67 9.14 9.58 9.53 9.73 9. 92 11.28 11.44 11.65 July —.- .47 .96 1.49 8.37 9.03 9.50 9.23 9.56 9.91 10.96 11.25 11.52 August... +.22 .82 1.48 8.20 8.88 9.48 9.17 9.43 9.90 10.98 11.12 11.38 September__ -. 12 .34 .98 7.98 8.40 8.93 8.80 9.06 9.38 10.54 10.77 11.09 October .. —.46 + .11 .35 7.63 8.19 8.43 8.54 8.78 8.94 10.39 10.44 tO. 68 November. -.71 —. 15 .29 7.42 8.08 s. it; 8.37 8.77 8.86 10.17 10.44 10.68 December.- —.68 -. 15 .35 7.45 7.91 8.23 8.83 9.32 9.17 10.47 10.51 10.96 . 17 .50 .68 7.98 8.42 8.00 10.91 11.07 11.06 Mean readings. May - Novem- ber.. .15 .66 1.05 8.16 8.72 9.04 9.04 9.31 9.54 10.83 11.05 11.23 400 DEEP WA LEEWAYS. Taking the difference between the highest and lowest mean monthly readings gives a fluctuation of Lake Ontario at Oswego of 5.54 feet. The months taken are May, 1870,4.83 feet, and November, 1895, —0.71. The fluctuation in the St. Lawrence River at lock No. 27 , between the months of May, 1886, and November, 1895, is 4.94 feet. At locks Nos. 15 and 14 the difference between the mean readings of May, 1870, and November, 1895, gives a fluctuation of 4.06 and 3.25 feet, respectively. The year 1870 gives a slightly lower annual mean reading than that for 1886, but a higher mean for the period of navigation. The curves of gauge readings for these years are nearly similar. More data are available for the year 1886, and it is taken as a typical high-water stage for detailed comparison with the low-water stage of 1895. The general location and approximate stations of the Canadian gauges used in the St. Lawrence River are as follows: General location of Canadian gauges. No. Station. Near the head of Galops Island, Galops Canal... 27 3620 Opposite Point Rock way at Iroquois, Galops Canal.... Near the head of Ogdens Island, Rauide Plat Canal...____ 25 4055 24 4260 Opposite Dry Island at Morrisburg, Rapide Plat Canal_ _ . . _ Near the head of Croills Island. Farrans Plat Canal ... 23 4470 22 5010 Opposite central part of Long Sault Island, Cornwall Canal. ..... Opposite Cornwall Island at Cornwall, Cornwall Canal.... Near the foot of Lake St. Francis, Beauliarnois Canal ..... 21 5280 15 14 5795 The above stations indicate the approximate location of the gauges, but the elevation of the water surface at the gauge does not correspond exactly to the elevation of water surface at the proposed center-line station, on account of its distance from the center line, intervening islands, eddies, etc. Table No. 2 .—Comparison of the high-water stage of 1886 and the low-water stage of 1895 of Lake Ontario with the St. La wrence River through Lake St. Francis. Location of gauges. Year. Average monthly readings. Mean read¬ ing, May to Novem¬ ber, inclu¬ sive. May. June. July. Au¬ gust. Sep¬ tem¬ ber. Octo¬ ber. No¬ vem¬ ber. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Feet. Lake Ontario at Oswe- 11886 4.52 4.:S2 3.92 3.48 3.12 2.83 2.39 3. 51 go, N. V ... 11895 0.88 0. 76 0.47 +0. 22 —0.12 —0.46 -0. 71 +0.15 St. Lawrence River at [1886 12. 36 12.13 11.70 11.33 10.98 10.62 10.49 11.37 lock No. 27__ \1895 8.83 8.67 8.37 8.20 7.98 7.63 7.42 8.16 Lock No.25 .. 11886 14.70 14.72 14.07 13.69 13.29 12.58 12.57 13. 66 11895 10.31 10.26 9.66 9. 40 8.81 8.34 7.81 9.23 Lock No. 24. Si 886 11.83 11.67 11.17 10. 67 10. 42 10.08 9.83 10.81 (1895 7.83 7. 67 7.42 7.08 6.50 6.17 5.67 6. 90 Lock No. 23.. 11886 11.92 11.67 11.33 10. 75 10. 42 10.00 9.75 10.83 11895 7. 75 7.58 7.17 7.00 6. 42 6.00 5.58 6. 79 Lock No. 22. [1886 11.59 11.49 11. 12 10. 67 10. 41 10.08 9.82 10. 74 11895 8.47 8.37 8.08 7.89 i. o2 7.17 6. 70 7.74 Lock No.21.. 11886 12.17 12.00 11.67 11.33 11.00 10. 67 10.50 11.31 11895 9. 00 8.92 8.67 8.58 8.25 7.92 7.67 8.43 11886 12.14 11.97 11.69 11.32 11.05 10.73 10.64 11.36 11895 9.66 9.52 9 22 9. 17 8.80 8.54 8. 37 9.04 Lock No. 14. 11886 13.18 13.10 12.82 12.57 12.15 12.20 12.12 12.59 11895 11.46 11.28 10.96 10.98 10.54 10.39 10.17 10.83 In Table No. 3 the difference between the mean gauge readings for May to November, inclusive, for 1886 and 1895 are reduced to a per¬ centage of the fluctations in Lake Ontario. DEEP WATERWAYS. 401 Differences between the standard high water determined in the 1896 report of the United States Deep Waterways Commission and the high stage of May, 1886, is given; also the differences between stand¬ ard low water and the mean low stage during the period of navigation for 1895. The difference between the high stage of May, 1886, and the low stage of May to November, inclusive, for 1895 is given. The difference between standard low water and that period covered by simultaneous gauge readings by Mr. David Molitor, United States assistant engineer, is also shown. , Table No. 3. —Tabulation of various gauge readings and, differences. Lake On¬ tario at Os¬ wego. St. Lawrence River at Canadian locks. No. 27. No. 25. No. 24. No. 23. No. 22. No. 21. No. 15. No. 14. Mean gauge readings, May-No- vember: Feet. Feet. Feet. Feet. Feet. Feet. Feet. \ Feet. Feet. 1886....... 3.51 11.37 13.66 10.81 10.83 10. 74 11.33 ! 11.36 12.59 1895. 0.15 8.16 9.23 6.90 6.79 7.74 8.43 I 9.04 10.83 3 36 3 21 4.43 3.91 4 04 3.00 2.90 i 2.32 Percentage of differences. 100 96 132 116 120 89 86 ! 69 52 Standard high water of 1896 report. 4.83 12.67 15.67 12.58 10.83 . 12.42 13.42 Mean monthly readings for May, 1886.. 4.52 12.36 14.70 11.83 11.92 11.59 12.17 12.14 13.18 Differences_ 0.31 0.31 0.97 —0.76 . 0.28 0.24 Standard low water of 1896 report. 0.32 8.31 9.49 (7.18) 6.99 7.95 (8.60) 9.13 10.98 Less mean readings of May-No- vember, 1895 .. 0.17 0.15 0.26 0.28 0.20 0.21 0.17 0.09 0.15 Mean of May, 1886, less the mean of May to November, inclusive, 1895_____ 4.37 4.20 5.47 4.93 5.13 3.85 3.74 3.10 2.35 Mean gauge readings, July27-A.u- gust 6,1898 __ 1.53 9.19 10.89 8.50 8.39 8.85 9.49 9.80 11.34 Less standard low water of 1896 report . 1.21 0.88 1.40 1.32 1.40 0.90 0.89 9.67 0.36 In Appendix No. 1, page 128, of the “Annual report of the depart¬ ment of railways and canals,” Dominion of Canada, for the fiscal year from July 1,1890, to June 30,1891, is a “ Statement of the highest and lowest water on the canals in the St. Lawrence district, May to Novem¬ ber in each year,” from 1849 to 1891, for gauges at locks Nos. 15, 21, 23, 24, 25, and 27. A similar statement in the report for the fiscal year from July 1, 1896, to June 30, 1897, on page 161, Part I, con¬ tinues this data to June, 1897. The following readings show the highest daily readings occurring during those months when water stood at or more than 0.5 foot above the mean reading of May, 1886, for the corresponding gauge: Highest gauge reading. Lock No. 27, May, 1870... ____ - 13.00 Lock No. 25: July, 1867 ______--- 15.67 May, 1870 _ _ ___ 16.17 July, 1876--- --- 15.67 May, 1884 __ _ 15.83 May, 1886 ______ 15.25 H. Doc. 149-26 402 DEEP WATERWAYS. Lock No. 24: July, 1867...-.-.... 12.50 May, 1870 .....'.......13.00 May. 1876 .... 12.75 May, 1884 ....... ....... .. 12.50 Lock No. 23: June, 1867...... 12.83 May, 1870 .. .... ... 12.75 July. 1876.. ... 12.50 May, 1884 ___ __ 12.75 LockNo. 22, July, 1876 .......... 12.50 Lock No. 21: May, 1870 .... .... 12.75 June, 1886...... .. 12.92 Lock No. 15, June, 1870_______ 12.67 These maximum daily readings average approximately 1 foot above the mean high stage of May, 1886. The following readings show the lowest daily readings occurring during those months when the water stood at or lower than 0.5 foot below the mean readings of 1895, May to November, inclusive: • Lowest gauge reading. Lock No. 27: November, 1889......i_7.08 November, 1895.......... 6.75 November, 1896.... ..... 7.50 Lock No. 25: November, 1875............. 8.33 November, 1887_________ 8.33 October. 1891 ..... 7.83 November, 1895........'.__ 6.67 November, 1896.........8.25 Lock No. 24: November, 1875.... ... 6.25 November, 1889........... 6.00 November, 1895..........4.42 November, 1896_ ______ 5.58 Lock No. 23: November, 1875... 6.00 November, 1895_ 4.83 November, 1896. 5.75 Lock No. 22: November. 1895.......... 5.83 November, 1896.... 6.92 Lock No. 21: November, 1895....... 7.33 November, 1896........7.67 Lock No. 15: November, 1895.......... 8.00 November, 1896...........8.54 Table No. 4 shows the approximate location of the gauges estab¬ lished by Mr. David Molitor, United States assistant engineer, over DEEP WATERWAYS. 403 that stretch of the river under consideration, and the reduction of the mean elevations of the water surface to standard low water. The percentage corrections are determined by multiplying 1.21, which is that stage of Lake Ontario at Oswego during the period cov¬ ered by the simultaneous gauge readings, by a factor obtained from the percentages of fluctuations found between the mean high water, 1886, and the mean low water of 1895, at the various Canadian gauges during the period of navigation. The approximate elevation of stand¬ ard low water so determined is used for a comparison with the reduc¬ tions otherwise obtained. All elevations used in this report are based on a line of levels run under direction of the United States Board of Engineers on Deep Waterways from the Greenbush bench mark elevation, 14.73 feet above mean tide at New York. Table No. 4. No. General location. Station. Mean ele¬ vation water sur¬ face. Percent¬ age cor¬ rection. Elevation standard low water. 15 Ogdensburg. 3272 245.811 1.16 244.65 14 Butternut Island. 3586 244.868 1.16 243.71 13 Sheldon Island. 3811 235. 805 1.22 234.59 12 Point Rockway... 4070 + 228. 543 1.60 226.94 11 Leishmans Point... 4225 + 226. 412 1.42 224.99 111 Waddington, below dam.... 4346 215. 202 1.42 213. 78 9 Murphys Island .. 4510 + 213. 675 1.45 212.23 8 Above Bradfords Hill. 4735 ± 208.983 1.32 207.66 7 Below Bradfords Hill._.. . 4770 + 207. 657 1.26 206.40 6 Louisville Landing.. 4975 206.188 1.(19 205.10 5 Richards Landing... 5123 204. 792 1.05 203.74 O Grass River..... 5633 156. 844 0.91 * 155.93 1 Racket River... 5855 154.595 0.82 153.78 Table No. 5, giving the standard high and low water after comple¬ tion of proposed work, has been computed from the elevation of the actual water surface, taken on various dates, reduced to standard low water by a comparison with the Oswego gauge readings for the corresponding dates, reduced to standard low water at the desired point by applying the percentage correction shown in Table No. 3, by a comparison with the standard low water obtained by applying the same percentage to the mean elevations determined by Mr. David Molitor, United States assistant engineer, and by a comparison with the difference between the nearest Canadian gauge reading and its standard low water. The various results thus obtained practically agree. The standard high water, after completion of the proposed works, is determined by a comparison of the high stage of Lake Ontario at Oswego, reduced to an elevation at the desired point by applying the percentage correction, and by a comparison with standard high water of the nearest Canadian gauges. The results thus obtained agree within reasonable limits. 404 DEEP WATERWAYS. The grade used for the channels of 30 or 21 feet depth, on which estimates depend, are determined by subtracting 30 or 21 feet from the standard low water, except as is noted under details of the esti¬ mates of the respective channels. Table No. 5. Number of nearest gauge. Elevation. General location. Station. Molitor’s. Canadian. Stand¬ ard high water. 1 Stand¬ ard low water. 1 State hospital grounds... 3149 14 27 248.8 244.4 Head of Galops Island.. 3629 14 27 247.1 242.7 Lock. Sheldon Island--__ 3779 13 13 27 247.1 242.7 Do .. 3779 27 237. 5 233.0 Leislimans Point.-.. 4213 11 24 231.0 225. 7 Lock, Clarks Point.. 4372 10 23 231.0 225.7 Do . . 4372 10 23 219.4 213.9 Off Dry Island ___ 4495 9 23 218.4 212.8 Below Bradfords Hill... 4705 22 210.9 200.2 Off Chat Island ... 4925 6 22 209.3 205.2 Richards Point .. 5075 6 22 209.0 205.0 St,. Lawrence Power Company Canal_ 5233 208.5 204.5 Lock, Massena Valley.. 5508 208.5 204.5 Do 5508 160.2 156.8 Near the mouth of Grass River_ 5033 2 15 160.2 156.8 Off Cornwall Island.. 5815 i 15 157. 0 153.7 Off St. Regis Island... ... 6035 i 15 15 150. 7 153.4 Lower portion Lake St. Francis... 7280 i 155.5 153.0 1 After improvement has been completed. Stations showing location of head of locks as given in Table No. 5 are for the 30-foot channel. The location of the locks for the 21-foot channel does not correspond exactly with the locks of the 30-foot channel. EFFECT OF ICE JAMS. Ice jams in the vicinity of St. Regis Island occur during the winter months, occasionally beginning in December or extending into April. The rise of water is generally from 10 to 20 feet above standard low water. The following readings show the highest readings during the months indicated, wherein the gauge readings of the Cornwall Lock No. 15 were over 30 feet, or 21 feet above standard low water, between the years 1802 to 1899, inclusive: Date. Reading. Height above standard low water. February, 1870._. 31.42 90 January, 1871. 30.43 21.29 January, 1884. 32.17 23.04 January, 1885. January, 1887..... 30.67 38.33 21.54 29.20 March, 1891. 30.25 21.12 Effects produced by ice jams in the vicinity of proposed construc¬ tion between Ogdensburg and the entrance to the land canal at Richards Point can be disregarded. DEEP WATERWAYS. 405 GENERAL TOPOGRAPHY AND MATERIAL ENCOUNTERED. From Ogdensburg to Waddington the ridge adjacent to the river is wide and varies from 40 to 70 or more feet above the water surface. It is broken by a few small creeks. In the vicinity of Gooseneck Island and Louisville Landing vallej^s extend southward toward the Grass River, which, at the latter place, approaches within about 3 miles of the St. Lawrence. From this point the trend of the ridges or valleys is nearly parallel to the St. Lawrence for a distance of about 11 miles, to the mouth of the Grass River. Several cross-overs exist between the valleys. The higher portions of points and islands crossed by the proposed center line have a rich but shallow loamy surface. The Massena val¬ leys are in general fertile bottom lands. The usual formation of all points, islands, and ridges is a core of bowlder clay approaching close to the surface on the upstream side, while on the downstream side considerable clay and sand of softer material is frequently found. A portion of the bowlder clay is more or less cemented and is classified as “liardpan” in the detailed esti¬ mates of yardage and cost. The remaining portion is classified as “hard material.” It consists of sand, gravel, and bowlders in vary¬ ing proportions, mixed with clay, but not cemented, liardpan is imposed upon several feet of soft blue clay in a few cases. The bulk of the clay is soft blue clay; a small proportion is a tough yellow clay. All the gravel found is angular, and appears to have been formed from local rock. It generally contains a small amount of clay, but not enough to cement it together. The rock appears to be a heavy dark-gray limestone with a clear ring and good weathering properties, as is shown by various outcrop¬ pings in the vicinity of Lisbon, Waddington, and Ilogansburg. It is called a calciferous sandstone; the percentage of lime varies through a large range. Rock excavation was in progress on the St. Lawrence Power Company’s canal and on the Canadian Canal improvements. It was easily drilled and conveniently broken for loading into cars, skips, etc. The strata are practically horizontal, and vary in thick¬ ness from a few inches up to about 5 feet. Some public buildings and dwellings are constructed from stone taken from local quarries. A similar stone is used in the construc¬ tion of a portion of the masonry in the Canadian locks. A considerable area of rock outcroppings appears on river bottoms at the Galops Rapids and the Rapide Plat. Other river bottoms in the vicinity of swift water are thickly covered with medium-sized bowlders. GENERAL ROUTE AND ESTIMATES. Those stretches of the river requiring radical improvement are the Galops Rapids, the swift water in the vicinity of Long Point and Point 40b DEEP WATERWAYS. Rockway, the Rapide Plat, and the swift water extending from Croflls Island to Cornwall Island, including the Long Sault. Estimates are based upon the standard sections and designs, except that the 600-foot river channel has been contracted to a 400-foot width between Lalone Island and the mainland, channel limits being above high water. In a few other places, where a side slope would extend into a high shore, a contraction of 100 feet on that side has been made. Additional excavation in river sections has been esti¬ mated at all curves between the outer curve and its tangents. No slope walls are estimated adjacent to river sections, where the excavated shore is hard material, as is the case at Lalone Island, Long Point, Point Rockway, and a few other places. For convenience in tabulating quantities and cost the proposed work is divided into sections, as follows: Section 1 .— Station 3481 to station 3763; from the St. Lawrence State hospital grounds to the dock at entrance of the Sheldon Island lock. Section 2. —Station 3763 to station 3819; Sheldon Island lock and excavation out to deep water. Section 3. —Station 3819 to station 3885; general river improvements at Long Point. Section Jf. —Station 3999 to station 4066; general river improvements at Point Rockway. Section 5. —Station 4206 to station 4233; Leishmans Point. Section 6. —Station 4233 to station 4261; Leishmans Point to land canal in Ogdens Island. Section 7. —Station 4261 to station 4278; land canal in Ogdens Island. Section 8. —Station 4278 to station 4356; from Ogdens Island canal to dock at entrance of lock in Clarks Point. Section 9. —Station 4356 to station 4396; lock at Clarks Point. Section 10. —Station 4396 to station 4933; from the end of dock below Clarks Point to deep water beyond Chat Island. Section 11. —Station 5055 to station 5491; from Richards Point to the dock for the Massena Valley lock. Section 12. —Station 5491 to station 5532; the Massena Valley lock. Section 13. —Station 5532 to station 5635; from the Massena Valley lock to the deep water in the St. Lawrence River near the mouth of Grass River. Section 11 ^.— Station 5668 to station 7028; from below the mouth of Grass River to opposite McKees Point, Lake St. Francis. The estimates for the proposed river improvements for the North Galops channel and for Long Point and Point Rockway are the same for the canals of 30 or 21 feet depth. DEEP WATERWAYS. 407 DETAILED ESTIMATES, 30-FOOT CHANNEL. Estimates are based on slack-water navigation from the head of Galops Island to the foot of Sheldon Island by means of rock-filled dams, connecting Galops, Benedict, Ray craft, Lalone, and Sheldon islands, and Sheldon Island with the mainland opposite, the lock of 0.7 feet lift, standard low water, being located in Sheldon Island. It is proposed to enlarge the North Galops channel by excavating a 20-foot channel 1,300 feet wide through the adjacent portions of Galops and Adams islands. The present channel at this point is estimated at 5,000 square feet area. The proposed channel would be 26,000 square feet, giving an increased area of 21,000 square feet. The present area of the North Channel between Adams Island and the guard wall of the Canadian canal is 16,000 square feet. The area of the South Channel near Red Mill Point is 19,500 square feet. This improvement of the North Channel, with the slack-water navigation in the South Channel, would make the cross section available for dis¬ charge to the Galops Rapids practically unchanged. The yardage and classification of material has been determined from the data on existing charts, and the estimated cost of the same is regarded as equivalent to the cost of any required improvements at this point. The yardage of material required for rock-filled dams is based upon a dam 4 feet above high water, having a top width of 20 feet and side slopes of 14 to 1. No deduction is made in these quantities for the material that could be deposited by scows, etc., from the “exca¬ vation under water.” The river bottom at the dam sites is either rock or hard material. The present currents between the islands have a maximum surface velocity of from 3 to 4 miles per hour, and are directed northward, except between Lalone and Sheldon islands. In general, the bulk of the material necessary for dams, back filling, embankments, and crushed stone for concrete is available from the material to lie excavated from channel. Table No. 6. ESTIMATE, 30-FOOT CHANNEL. Section 1 .—The estimated yardage and cost of material to be exca¬ vated and dams constructed are as follows: Excavation under water: Rock, 2,939,642 cubic yards, at $2.50.... $7,349,105 Hard material, 1,098,583 cubic yards, at 25 cents.. 274,646 Gravel, 106,050 cubic yards, at 15 cents.. 15,908 Excavation in the dry: Rock. 25,036 cubic yards, at 65 cents... 16,273 Hard material, 103,018 cubic yards, at 25 cents. 25,755 Gravel, 41,112 cubic yards, at 15 cents. 6,167 408 DEEP WATERWAYS. For 1 lie improvement of the North Galops Channel: Excavation under water: Rock. 566.908 cubic yards, at $2.50...$1,417,270 Hard material, 226,805 cubic yards, at 25 cents. 56,701 Excavation in tlie dry: Rock, 592.444 cubic yards, at 65 cents... ... 385,089 Hard material. 860,389 cubic yards, at 25 cents . 215,097 Rock-filled dams: Between Galops and Benedict islands. 20,737 cubic yards, at 50 cents_ __—..... 10.368 Between Benedict and Raycraft islands, 21,972 cubic yards, at 50 cents .......... 10.986 Between Raycraft and Lalone islands. 46,425 cubic yards, at 50 cents. ...... 23.212 Between Lalone and Sheldon islands. 30.310 cubic yards, at 50 cents.... 15,155 Right of way: On Galops Island, including the acreage required for river improve¬ ments at the northwest portion, 114 acres, at $75.... 8. 550 On mainland opposite Galops Island, 10 acres, at $75.. 750 Six islands, including Adams Island.... 6,500 Total estimated cost of section 1..... 9,837,532 Section 2 .—Docks and cribs in approaches to the lock, the lock, and the retaining-wall quantities are based on the standard designs. The lock is located between station 3777+29 and station 3787+00 in Sheldon Island. The lift at standard low water is 9.7 feet. The fluctuation of high water is 4.4 feet and 4.5 feet. Both above and below the lock, respectively, the foundations are on rock. The estimated quantities and cost of construction are as follows: Excavation under water: Rock, 33,459 cubic yards, at $2.50..... $83,647 Hardpan, 112,788 cubic yards, at $1.... 112,788 Hard material, 311,926 cubic yards, at 25 cents.. 77,981 Clay, 89,643 cubic yards, at 15 cents..... 13,446 Gravel, 6,144 cubic yards, at 15 cents.... 922 Excavation in the dry: Rock, 267,850 cubic yards, at 65 cents .... 174,103 Hardpan. 106,492 cubic yards, at 35 cents_-. . . 37.272 Hard material, 85,993 cubic yards, at 25 cents.. 21,498 Clay, 48,911 cubic yards, at 15 cents.. 7,337 Approaches to lock (docks): Timber— Oak. 79.680 feet B. M., at $50 per M .. 3,984 Hemlock, 5,588,962 feet B. M., at $23 per M... 128.546 Pine, 1,594,432 feet B. M., at $30 per M. .. _ 47,832 Stone fill, 88,441 cubic yards, at 60 cents.. 53,065 Iron, driftbolts, 652,005 pounds, at 3 cents ..... 19,560 Retaining walls, 4,849 cubic yards, at $4... 19,396 Rock-filled dam from Sheldon Island to mainland, 50,046 cubic yards, at 50 cents....... 25,023 Back fill, 105,128 cubic yards, at 25 cents. 26,282 DEEP WATERWAYS. 409 Lock No. 1 (Sheldon Island).. $816,888 Lock operating machinery.. 100,000 Right of way: Sheldon Island, 62 acres, at $100 ..... $6,200 Mainland, 20 acres, at $75. 1,500 - 7,700 Total estimated cost of section 2 ..... 1,777,270 Sections 3 and 4-—-It is proposed to improve the river at Long Point and at Point Rockway by so increasing the cross section that the slope shall be practically uniform from Sheldon Island to Leishmans Point. Simultaneous gauge readings taken above and below Long Point give a difference in elevation of water surface of 2.25 feet in a dis¬ tance of 4,000 feet, estimated along the central thread of the river. The present estimated mean available area at an elevation of 234 feet, corresponding to that stage of the river during which discharge measurements were made at Ogdensburg, is 37,540 square feet. This gives, with the measured discharge of 219,100 cubic feet per second, a mean velocity of 5.84 feet per second. The maximum velocity observed by using rod floats submerged about 6 feet was 10.3 feet per second. The proposed river improvements upon which estimates are based increase the area to 56,860 square feet and decrease the mean velocity to 3.9 feet per second. Similarly, at Point Rockway, a difference in elevation of 3.70 feet in an estimated distance of 7,000 feet was determined. The eleva¬ tion of water surface for the mean present section of 33,150 square feet for a stage corresponding to that used for Long Point is 229.6 feet. The mean velocity is 6.6 feet per second. The maximum observed velocity was 9.8 feet per second. The area at this point is increased to 51,730 square feet and tlie mean velocity reduced to 4.2 feet per second. Eddies exist below both Long Point and Point Rockway, and the estimated distances of 4,000 feet and 7,000 feet, respectively, may be too small for representing present conditions. The proposed improvements would destroy or largely decrease these eddies. These increased areas at Point Rockway and Long Point will decrease the elevation of the water surface below Sheldon Island about 1.6 feet. The increased area at the head of the North Galops Channel will tend to decrease this effect. The decrease of the veloc¬ ity of approach to the pools below Long Point and Point Rockway will tend toward a relative increase in the elevation of the water sur¬ face at the head of these pools. The resulting slopes at the improved points and through the pools are nearly the same. Therefore a uni¬ form slope for estimating the yardage of excavation is used from the lock approaches below Sheldon Island to Leishmans Point. 410 DEEP WATERWAYS. *The estimate is based on excavating all material from a side line 100 feet to right of the center line, slopes 2 on 1, out to deep water, and the quantities are the same for the proposed canals of either 21 or 30 feet depth. The estimated quantities and cost are as follows: Section 3: Excavation under water— Hardpan, 818,948 cubic yards, at $1.... $818, 948 Clay, 7,160 cubic yards, at 15 cents.. 1,074 Excavation in the dry— Hardpan, 680,631 cubic yards, at 35 cents.. 238,221 Clay, 277,110 cubic yards, at 15 cents.. .. 41,567 Right of way, 36 acres, at $75.... 2, 700 Total estimated cost of section 3__ _ 1,102, 510 Section 4: Excavation under water— Rock, 61,348 cubic yards.at $2.50...... 153,370 Hard material, 1,101,611 cubic yards, at 25 cents.. 275,403 Excavation in the dry— Rock, 105,317 cubic yards, at 65 cents___ 68,456 Hard material, 595,036 cubic yards, at 25 cents.... 148, 759 Right of way, 37 acres, at $75 ... . 2,775 Total estimated cost of section 4. ... 648,763 At the Rapide Plat it is proposed to obtain slack-water navigation by a rock-filled dam between Clarks Point and Ogdens Island and a lock located in Clarks Point. A land canal, earth section, will be necessary through Leishmans Point and through a portion of Ogdens Island. Otherwise the 600-foot river section through the Little River is used in estimating quantities to be excavated. The maxi¬ mum velocity observed just below the entrance to the land canal at Leishmans Point was 8 feet per second, about 700 feet off the shore— a velocity greater than that existing on the proposed sailing course at and above this entrance. A guard crib 300 feet long, of the standard section, is estimated for at this point. This proposition would destroy the existing dam and water power at Waddington. It is estimated that 15,000 theoretical horsepower could be developed by properly improving the dam and the highway crossing above the same and the canal. Three mills, including the electric-lighting plant, are located on the dam, which is 950 feet long, and eight mills are located on the canal, which is 950 feet long. These mills consist of two flour and grist mills, a lighting plant, milk receiver, blacksmith shop, planing, shingle, and saw mills, etc. Their outputs are mainly to meet local demands. The total present capacity of the above mills will not exceed 300 horsepower. The valuation of this property is based upon its possible future improvement. DEEP WATERWAYS. 411 The lock at Clarks Point is on rock foundation; the lift at standard low-water is 11.8 feet; the high water above the lock is 5.3 feet, and below the lock it is 5.5 feet above standard low water. The division of excavation under water and in the dry has been determined from the present shore line. The estimated quantities and cost for sections 5 to 9, inclusive, are tabulated as follows: Section 5: Excavation under water— Rock, 23,626 cubic yards, at $2.50..... $59,065 Hard material, 46,245 cubic yards, at 25 cents.... 11,561 Clay, 52,860 cubic yards, at 15 cents... 7,929 Gravel, 28,275 cubic yards, at 15 cents .. 4,241 Excavation in the dry— Rock, 104,195 cubic yards, at 65 cents.. .. _ 67,727 Hard material, 638,665 cubic yards, at 25 cents__ 159, 666 Clay, 75,560 cubic yards, at 15 cents.. . 11,334 Guard crib at entrance— Timber— Hemlock, 538,620 feet B. M.. at $23 per M. 12,388 Pine, 163,536 feet B. M., at $30 per M... 4,906 Stone fill, 8,400 cubic yards, at 60 cents....... 5,040 Iron driftbolts, 62,585 pounds, at 3 cents...... 1,878 Slope walls. 8,125 square yards, at $1.10.... 8,937 Right of way, 47 acres, at $75 ...... 3,525 Total estimated cost of section 5....... 358,197 Section 6: Excavation under water— Rock, 340,557 cubic yards, at $2.50 ... 851, 393 Hard material, 497,450 cubic yards, at 25 cents... 124,362 Clay, 90,350 cubic yards, at 15 cents ___13,552 Gravel, 12,868 cubic yards, at 15 cents...... 1,930 Total estimated cost of section 6....... 991.237 Section 7: Excavation in the dry— Rock, 61,129 cubic yards, at 65 ceqts..._ 39, 734 Hard material. 217,969 cubic yards, at 25 cents... 54,492 Clay, 3.404 cubic yards, at 15 cents.. 511 Gravel, 595,707 cubic yards, at 15 cents.... 89,356 Slope walls, 11,500 square yards, at $1.10.... 12,650 Right of way, 76 acres, at $75 ...... 5,700 Total estimated cost of section 7. .... 202,443 Section 8: Excavation under water— Rock, 1,396,943 cubic yards, at $2.50. . 3,492,357 Hard material, 301,827 cubic yards, at 25 cents. 75.457 Clay, 715,477 cubic yards, at 15 cents.. .... 107,322 Gravel, 236,336 cubic yards, at 15 cents. 35,450 412 DEEP WATERWAYS. Section 8—Continued. Excavation in the dry— Rock, 113,888 cubic yards, at 65 cents .. . $74,024 Hard material, 106.736 cubic yards, at 25 cents.. 26,684 Gravel, 80,808 cubic yards, at 15 cents.... 12,121 Right of way, 20 acres, at $75----- 1,500 Water power and mill properties. 100,000 Total estimated cost of section 8... 3,924,915 Section 9: Excavation under water— Rock. 12,230 cubic yards, at $2.50 ...... 30.575 Hard material, 144,100 cubic yards, at 25 cents .. 36.025 Clay, 172,069 cvibic yards, at 15 cents .. 25,810 Excavation in the dry— Rock, 129,515 cubic yards, at 65 cents...-. 84,185 Hardpan, 221,494 cubic yards, at 35 cents.. 77,523 Hard material, 34,555 cubic yards, at 25 cents. ... 8. 639 Clay, 515,923 cubic yards, at 15 cents... 77,388 Approaches to locks, docks, and cribs— Timber— Oak, 71,289 feet B. M., at $50 per M .... .. 3,564 Hemlock, 5,919.868 feet B. M., at $23 per M __ 136,157 Pine, 1,577,428 feet B. M., at $30 per M__„ 47,323 Stone fill. 90.985 cubic yards, at 60 cents ..... 54,591 Iron driftbolts, 660,987 pounds, at 3 cents.. 19,830 Retaining walls, 4,490 cubic yards, at $4..... 17,960 Rock-filled dam, 88,532 cubic yards, at 50 cents.. 44,266 Back fill, 195,878 cubic yards, at 25 cents.__ 48,969 Lock No. 2 (Clarks Point)..... 864,003 Lock-operating machinery.... ... 100,000 Right of way, 54 acres, at $75...._ 4,050 Total estimated cost of section 9..... 1,680,858 Section 10. —The higher velocities occur off Goose Neck Island and Bradford Ilill. The maximum observed velocities at these points were 6.8 and 8.8 feet per second, respectively. The estimated quantities and cost are as follows: Excavation under water: , Rock. 9,750 cubic yards, at $2.50._ ...... ...$24,375 Hard material. 2,437.748 cubic yards, at 25 cents__.. 609, 437 Excavation in the dry, on Dry Island, hard material. 75,720 cubic yards, at 25 cents ...... 18,930 Right of way, 5 acres, at $75..... . 375 Total estimated cost of section 10 ...... 653,117 It is proposed to pass the swift water existing between Croills Island and Cornwall, including the Long Sault, with a land canal, earth sec¬ tion, leaving the St. Lawrence at Richards Point and entering the same near the mouth of Grass River. The distance between these points is 10.57 miles. The maximum observed velocity off Richards Point below the entrance of the proposed canal was 4.8 feet per sec- DEEP WATERWAYS. 413 ond. A guard crib 300 feet long at this entrance is included in the estimate. The entrance to the St. Lawrence Power Company’s canal is about 2 miles below the entrance to the proposed canal. The difference in elevation of the water surface at these entrances is 3.9 feet. The sectional area of the St. Lawrence Power Company’s canal is 5,625 square feet; average width, 225 feet; depth of water, 25 feet, and its mean velocity is estimated to be 4.47 feet per second when its maximum water power is utilized. It would require a velocity of 3.2 feet per second in the proposed ship canal to supply the maximum needs of the power canal. This velocity will give a fall of 0.5 foot between stations 5075 and 5223. The elevation of standard low water at the entrance of the proposed canal is 205. At the intersection of the two canals it would be 204.5. The elevation of water at the entrance to the power canal is 201.1. Its elevation at the intersection would be 200.1. An increased head to the power canal of 4.4 feet would exist at the point of intersection. No estimate has been made of the effect of this increased head or of the cost of controlling gates on the power canal between the proposed ship canal and the St. Law¬ rence River. To reduce the velocity and aid navigation at the intersection of these canals the area of the proposed canal is practically doubled, decreasing to a standard section 1,000 feet above the point of inter¬ section. The yardage estimated as necessary for this purpose is 459,330 cubic yards more than that required for a standard section. Above the lock for a distance of 9,700 feet embankments are neces¬ sary. The estimated yardage for this purpose is based on an embank¬ ment 50 feet wide, 2 to 1 slopes, and the top at the same elevation as the adjacent slope wall. Its average height is 11 feet, its maximum height for a short distance is 28.5 feet. The embankment will rest upon a stratum of soft blue clay over a portion of its length. At such points it may be necessary to construct the embankment at a safe distance from the excavated prism. Right of way is ample for such modification, and the estimated quantities would remain unchanged. No rock exists at the entrance at Richards Point or at the crossing of the St. Lawrence Power Company’s canal, and no separation has been made for the relatively small amount of material to be excavated under water. The estimated quantities and cost of the same for section 1L are as follows: Excavation: Rock, 187,060 cubic yards, at 65 cents ..... $121,589 Hardpan, 6,895,286 cubic yards, at 35 cents..... 2,413,350 Hard material, 3,964,744 cubic yards, at 25 cents ___ 991,186 Clay, 8,518,415 cubic yards, at 15 cents ...... 1,277,762 Sand, 956,364 cubic yards, at 15 cents...... 143,455 Gravel, 610,933 cubic yards, at 15 cents... 91,640 414 DEEP WATERWAYS. Excavation—Continued. Guard crib— Timber— Hemlock. 683,876 feet B. M., at $23 per M.. $15,718 Pine. 163,536 feet B. M.,at $30 per M...- 4.906 Stone fill, 10,350 cubic yards, at 60 cents .... 6,210 Iron drift bolts, 76,385 pounds, at 3 cents _..-. 2,291 Embankment, 586,927 cubic yards, at 15 cents.... 88,039 Slope walls, 212,120 square yards, at $1.10.... 233,332 Surface drainage.. ....-.. 200 Right of way, 1,110 acres, at $75___ 83,250 Highways, 4.6 miles, at $200 ....... 920 Highway crossing, steam ferry-..---- 20,000 Total estimated cost of section 11____ 5,493,848 Section 12 .—The lift of the lock is 47.7 feet at standard low water. The elevations of rock average about 2 feet above the grade of the bottom of the canal below the lock. The high water at Richard’s Point is 4 feet, and in the vicinity of the mouth of Grass River it is 3.4 feet above standard low water. Ice jams will raise the water at the lower gates in exceptional cases nearly 30 feet above standard low water. The estimated quantities and cost of same are as follows: Excavation in the dry: Rock, 99,810 cubic yards, at 65 cents.-.._ $64,876 Hardpan, 775,043 cubic yards, at 35 cents... 271,265 Hard material, 306,172 cubic yards, at 25 cents.... 76,543 Clay, 1,100,101 cubic yards,at 15 cents.......... 165,015 Sand, 72,577 cubic yards, at 15 cents _........ 10,886 Approaches to locks, docks, and cribs: Timber: Oak, 79,680 feet B. M., at $50 per M..... 3,984 Hemlock, 7,016,304 feet B.M., at $23 per M___ 161,375 Pine, 1,778,124 feet B. M., at $30 per M.._... 53,344 Stone fill, 115.348 cubic yards, at 60 cents . ___.__ 69,209 Iron drift bolts, 749,020 pounds, at 3 cents........ 22,470 Lock No. 3 (Massena) ...... ... 1,667,422 Lock-operating machinery____ __ 100,000 Back fill, 488,410 cubic yards, at 25 cents.. 122,103 Right of way, 183 acres, at $75....... 13,725 Total estimated cost of section 12______ 2,802,217 Section IS .—Considerable soft blue clay exists on this section. The maximum cut is 79 feet. Rock for slope walls is not found in the excavation. The estimated quantities and cost of same are as follows: Excavation in the dry: Hardpan, 1.888,411 cubic yards, at 35 cents__ $660,944 Hard material, 41,473 cubic yards, at 25 cents.. 10, 368 Clay, 6,295,996 cubic yards, at 15 cents... 944,399 Sand, 451,686 cubic yards, at 15 cents... 67,753 Slope walls. 53,853 square yards, at $1.45.... 78,087 Right of way, 393 acres, at $75_____ 29,475 Total estimated cost of section 13... 1,791,026 DEEP WATERWAYS. 415 Section 1J +.—The average width of Grass River near its mouth is 600 feet at standard low water, or at an elevation of 157 feet; at an elevation of 170 feet, the average width is about 850 feet. When the St. Lawrence Power Compan 3 T ’s canal has been developed to its maxi¬ mum capacity, it will increase the discharge of Grass River by 25,300 cubic feet per second. Polly’s Gut, to the north of the proposed sail¬ ing course, has an observed velocity of 10 feet per second. The maxi¬ mum observed velocity about 1,000 feet above the New York and Ottawa Railway bridge was 6.5 feet per second. The headroom at this bridge is approximately 40 feet at standard low water. Borings taken in the vicinity of St. Regis Island and near the sailing course off Clark’s Island indicate that hard material may be encoun¬ tered down to Hamilton Island, and that below this island the material is sand. Therefore, the classification of material below Station 5930 has been made upon this basis. The quantities to be excavated under water are: Hard material, 2,007,839 cubic yards, at 25 cents .. $501,900 Sand, 1,980,485 cubic yards, at 15 cents.... 297,073 New York and Ottawa Railway bridge....._ 222,888 Total estimated cost for section 14 ...... 1,021,921 SUMMARY FOR 30-FOOT CHANNEL. Section 1..-.... $9,837,532 Section 2......:... 1,777,270 Section 3........ . _ 1,102,510 Section 4......... 648,763 Sections. .....—_ 358,197 Section 6. 991,237 Section 7. 202,443 Section 8. 3,924,915 Section 9. 1,680,858 Section 10.. '53,117 Section 11. 5,493,848 Section 12. 2,802,217 Section 13____ ___ . 1,791,026 Section 14. 1,021,921 Total.. . 32,285,854 Note.— The difference between the total and that given in Table No. 8 is due to not carrying the mujtiplication out further. DETAILED ESTIMATES— 21-FOOT CHANNEL. The center line, the right of way, the dams, and the general river improvements at the North Galops Channel and at Long Point and Point Rockway are identical with the same items in the estimate of the channel of 30-foot depth. The principal exception to the modifi¬ cation of quantities caused by raising the grade line 9 feet is a canal of 25 feet instead of 21 feet depth from the entrance to the land canal at Richards Point down to the intersection of the proposed canal with the St. Lawrence Power Company’s canal. The width at grade is 215 41G DEEP WATERWAYS. feet, and its sectional area 0,680 square feet. This increase in area is made in order to decrease the velocity through the proposed canal when the power canal is being utilized to its full capacity. The resulting velocity in the proposed canal is 3.78 feet per second, and the fall from Richards Point to the intersection of the two canals is 1 foot. This reduces the difference between the grade lines of the proposed canals of 21 and 30 feet depth to 8.5 feet instead of !) feet from the intersection with the power canal down to the Massena Valley lock. The yardage of rock to be excavated in the dry is not as large as will be required for concrete masonry, stone filling in docks and cribs, and in rock-filled dams on sections 1, 2, and 9. Considerable rock, classified under “Excavation under water,” could be excavated in the dry by building protection levees. Also on these sections rock outcroppings appear within short distances. Rock excavation on sections 5 and 7 will about equal the amount of rock required for slope walls and stone filling in the guard crib. The rock excavation on section 11 will furnish but a small per cent of the amount required for slope walls and lock construction, while no rock exists in section 12. No rock outcroppings were found in the vicinity of these sections. The estimated quantities and cost of excavation and construction for the various sections follow: Table No. 7.— Estimate 21-foot channel. Section 1: Excavation under water— Rock, 921,005 cubic yards, at $2.50 ._. ' .... $2,302,512 Hard material, 400.091 cubic yards, at 25 cents___ 100,023 Clay, 29,292 cubic yards, at 15 cents..... 4,394 Gravel, 49,610 cubic yards, at 15 cents____ 7,441 Excavation in the dry— Rock, 3,354 cubic yards, at 65 cents .... .... 2,180 Hard material, 100.414 cubic yards, at 25 cents. 25,104 Gravel, 6,747 cubic yards, at 15 cents........ 1,012 The improvement of the North Galops Channel, the same as for the 30-foot channel________ 2,074,157 Rock-filled dams, 119,444 cubic yards, at 50 cents.. 59,722 Right of way, same as for the 30-foot channel. 15,800 Total estimated cost of section 1 .... 4,592,345 Section 2: Excavation under water— Hardpan, 12.200 cubic yards, at $1 ..... .. 12,200 Hard material, 32,010 cubic yards, at 25 cents____ 8,002 Clay, 96,585 cubic yards, at 15 cents..... 14,488 Excavation in the dry— Rock, 126,544 cubic yards, at 65 cents... 82,253 Hardpan, 91.589 cubic yards, at 35 cents____ 32,056 Hard material, 82,122 cubic yards, at 25 cents.. ... 20,531 Clay, 45,258 cubic yards, at 15 cents ____ 6, 788 DEEP WATERWAYS. 417 Section 2—Continued. Approaches to locks (docks) — Timber— Oak. 79,680 feet B. M., at $50 per M. $3,984 Hemlock, 4,356,320 feet B. M., at $23 per M_. 100,195 Pine. 1,606,632 feet B. M., at $30 per M ... 48,199 Stone fill. 71,716 cubic yards.at 60 cents.-... .. 43.010 Iron drift bolts, 535,090 pounds, at 3 cents.... 16,053 Retaining walls, 6,233 cubic yards, at $4.... 24.932 Rock-filled dam. 50,046 cubic yards, at 50 cents. .... 25,023 Back fill, 66,768 cubic yards, at 25 cents. . 16,692 Lock No. 1 (Sheldon Island) . 541,141 Lock operating machinery.. 100,000 Right of wav. same as for 30-foot channel.. 7,700 Total estimated cost of section 2 ... 1,103,267 Section 3, same as for 30-foot channel........ 1,102,510 Section 4, same as for 30-foot channel... 648,763 Section 5: Excavation under water— Rock. 368 cubic yards, at $2.50 ... 920 Hard material, 25,301 cubic yards, at 25 cents. 6,325 Clay, 20,850 cubic yards, at 15 cents.. 3,128 Gravel, 7.965 cubic yards, at 15 cents.. 1,195 Excavation in the dry— Rock, 16,493 cubic yards,at 65 cents.... 10,722 Hard material, 563,546 cubic yards, at 25 cents.. 140, 886 Clay, 70,005 cubic yards, at 15 cents.... 10,501 Guard crib at entrance— Timber— Hemlock, 498,460 feet B. M., at $23 per M . 11,465 Pine, 163,536 feet B. M..at $30 per M .. 4,906 Stone fill, 7,860 cubic yards, at 60 cents.. 4,716 Iron drift bolts, 58,485 pounds, at 3 cents.. .. 1. 755 Slope walls, 8,125 square yards, at $1.10 ... 8,937 Right of way, same as for the 30-foot channel... 3,525 Total estimated cost of section 5.. 208.981 Section 6: Excavation under water— Rock, 104,513 cubic yards, at $2.50. . . 261,282 Hard material, 322,999 cubic yards, at 25 cents. 80,750 Clay, 16,640 cubic yards, at 15 cents. 2,496 Gravel, 10,753 cubic yards, at 15 cents... 1,613 Total estimated cost of section 6 .. 346,141 Section 7: Excavation in the dry— Rock, 7,745 cubic yards, at 65 cents. .. 5,034 Hard material. 197,880 cubic yai’ds. at 25 cents. 49,470 Clay, 2,756 cubic yards, at 15 cents.. . 413 Gravel, 481,258 cubic yards, at 15 cents... 72,189 Slope walls, 11.500 square yards, at $1.10... 12,650 Right of way, same as for 30-foot channel... 5, 700 Total estimated cost of section 7... 145,456 H. Doc. 149-27 418 DEEP WATERWAYS. Section 8: Excavation under water— Rock, 447,990 cubic yards, at §2-50... SI, 119.975 Hard material, 95,894 cubic yards, at 25 cents... 23,849 Clay, 555,395 cubic yards, at 15 cents..... . 83,309 Gravel, 153.856 cubic yards, at 15 cents.- 23,078 Excavation in the dry— Rock, 49,027 cubic yards, at 65 cents..... 31,868 Hard material, 98,516 cubic yards, at 25 cents-... 24,629 Gravel, 74,604 cubic yards, at 15 cents. 11,191 Right of way, same as for the 30-foot channel.. 1. 50.) Water-power and mill properties, same as for the 30-foot channel. 100. OJJ Total estimated cost of section 8.... 1.419,399 Section 9: Excavation under water— Rock, 432 cubic yards, at $2.50... 1,080 Hardpan, 52,771 cubic yards, at $1...... 52,771 Clay, 23, 253 cubic yards, at 15 cents. 3, 488 Excavation in the dry— Rock, 51,446 cubic yards, at 65 cents.... 33.440 Hardpan, 206.135 cubic yards, at 35 cents.... 72.147 Clay, 402,038 cubic yards, at 15 cents.. .. 60,306 Approaches to locks, docks, and cribs— Timber— Oak, 19,680 feet B. M., at $50 per M. 3, 984 Hemlock, 5,635,240 feet B. M.,at $23 per M.. 129.6 1 1 Pine, 1,857,504 feet B. M., at $30 per M... 55,725 Stone till, 95.967 cubic yards, at 60 cents... 57.580 Iron drift bolts, 632,760 pounds, at 3 cents. . 18, 983 Rock-filled dam, 88,532 cubic yards, at 50 cents... 44,266 Back fill. 124,636 cubic yards, at 25 cents. .. 31.159 Lock No. 2 (Clarks Point)...._.. 571,506 Lock operating machinery.. 100,000 Right of way, same as for the 30-foot channel._. 4.010 Total estimated cost of section 9_..... . 1,240. 096 Section 10: Excavation under water, hard material, 494,765 cubic yards, at 25 cents....... 123,691 Excavation in the dry, hard material, 49,120cubic yards, at25 cents. 12,280 Right of way, same as for the 30-foot channel... 375 Total estimated cost of section 10 . . 136. 346 Section 11: Excavation in the dry— Rock, 14,823 cubic yards, at 65 cents ... 9.635 Hardpan, 5,490,943 cubic yards, at 35 cents... .. 1,921,830 Hard material. 3,226,583 cubic yards, at 25 cents. . 806,646 Clay. 7,479.665 cubic yards, at 15 cents.... 1,121,950 Sand, 935,842 cubic yards, at 15 cents.... 140,376 Gravel, 637,275 cubic yards, at 15 cents. 95,591 DEEP WATERWAYS. 419 Section 11—Continued. Guard crib— Timber— Hemlock, 498,460 feet B. M., at $23 per M... $11,465 Pine, 163,536 feet B. M., at $30 per M. 4.906 Stone fill, 7,860 cubic yards, at 60 cents.. 4,716 Iron drift bolts, 58,485 pounds, at 3 cents.... 1,755 Embankment, 574,420 cubic yards, at 15 cents__ 86,163 Slope walls, 212.518 square yards, at $1.10. 233,770 Surface drainage, same as for the 30-foot channel__ 200 Right of way. same as for the 30-foot channel... 83,250 Highways, same as for the 30-foot channel. . . . 920 Highway crossings, same as for the 30-foot channel.. . 20,000 Total estimated cost of section 11... Section 12: Excavation in the dry— Rock, 18,888 cubic yards, at 65 cents. Hardpan, 666,739 cubic yards, at 35 cents ... Hard material, 201,647 cubic yards, at 25 cents .. Clay, 816,135 cubic yards, at 15 cents .. Sand, 71,855 cubic yards, at 15 cents.. Approaches to locks, docks, and cribs— Timber— Oak, 79, 680 feet B. M., at $50 per M .. Hemlock, 5.393,320 feet B. M., at $23 per M. Pine, 1,184,664 feet B. M., at $30 per M_ Stone fill, 92.120 cubic yards, at 60 cents ... Iron driftbolts, 608,820 pounds, at 3 cents_ Lock No. 3 (Massena)..-.. Lock-operating machinery.... Back fill, 399,890 cubic yards, at 25 cents_ Right of way. same as for the 30-foot channel _ 4,543,173 12,277 233.358 50.412 122.420 10,778 3, 984 124.046 35,540 55.272 IS.264 1,086.042 100,000 99.973 13,725 Total estimated cost of section 12.... 1, 966,091 Section 13: Excavation in the dry— Hardpan, 983,931 cubic yards, at 35 cents. 344,376 Hard material, 239,485 cubic yards, at 25 cents. 59,871 Clay, 5,722,842 cubic yards, at 15 cents. 85S, 426 Sand, 463,422 cubic yards, at 15 cents.... 69,513 Slope walls, 54,100 squar yards,at $1.45 ... . ._ 78,445 Right of way, same as for the 30-foot channel.. 29, 475 Total estimated cost of section 13...... 1,440,106 Section Ilf .—From the existing charts no excavation is found neces¬ sary for a ship canal of 21 feet depth from the end of section 13 to Station 7283. New York and Ottawa Railway bridge $222.888 DEEP WATERWAYS. 420 SUMMARY FOR 21-FOOT CHANNEL. Section 1... .... .. $4,592,345 Sec ion 2..-.. ...1,103,267 Section 3. ...-..... 1,102,510 Section 4. _ ... ... . ._. 648, 763 Sections.. .......-.. 208,981 Section 6. . ... 346,141 Section 7... . .. . 145,456 Section 8... . ......__ 1,419,399 Section 9......... 1,240,096 Section 10..... 136,346 Section 11... .... 4,543,173 Section 12...... . .... 1,966,091 Section 13...... ... ... 1,440,106 Section 14..... .. . 222,888 Total ..... .... . 19,115,562 Note.— The difference between the total and that given in Table No. 9 is due to not carrying the multiplication out farther. Tables Nos. 8 and 9 give a summary of tjie quantities and costs of the 30 and 21 foot channels, respectively. Table No. 10 gives the location, lifts, and costs of the various locks for both 30 and 21 foot channels. Tables Nos. 11 and 12 give a classification of the kinds of channels of the proposed 30 and 21 foot channels, respectively. Table No. 8.— Summary of quantities and cost — 30-foot channel. Classification of material. Unit of quan¬ tity. Quantities in section — 1. 9 3. 4. 5. 6. Excavation under water : Rock ... . Cubic yards.. .do . 3,506,550 33,459 112,788 311,926 89.643 6,144 267,850 106,492 85,993 48,911 ’818,"948’ 61,348 23,626 340.557 Hardpan .. Hard material . .do . 1,325,388 1,101,611 46,245 53,860 28,275 104,195 497,450 90,350 12,868 Clay.. ...do. 7,160 Gravel. .do. 106,050 617,480 Excavation in the dry: Rock. . .do _ 105,317 Hardpan . 680,631 '277,'iio" Hard material . 063,407 595,036 638,665 75,560 Clay . . .. -do . ' Gravel . .do . 41,113 Back fill. .do . 105,128 4, 849 Retaining walls .. . _do.. Slope walls . . Square yards Feet B. M ... 8,125 Docks and cribs: Oak . 79,680 5,588,962 1,594,432 88,441 652,005 50,046 20 §6,2o0 Hemlock . .do . . 538,620 163,536 8,400 62,585 Pine. Stone fill. Cubic yards.. Iron, driftbolts. Pounds . Rock-filled dams. Cubic yards.. Acres__ 119,444 124 §6,500 Right of way. 36 37 47 Islands near Galops. DEEP WATERWAYS. 421 Table No. 8. — Summary of quantities and cost — 30-foot channel —Continued. Quantities in section— Classification of material. Excavation under water: Rock. Hard material. Clay. Gravel. . Excavation in the dry: Rock. Hard pan.. Hard material. Clay. Sand. Gravel. Embankments. Back fill. Retaining walls.. Slope walls. Docks and cribs: Oak ... . Hemlock. Pine. Stone fill.. Iron, driftbolts. Highways. Ferry (at highway cross¬ ing). Rock-filled dams. Right of way .. Water power, etc. Drainage.. Unit of quan¬ tity. Cubic yards. .do. _do. .do. _do. _do. _do. .do. _do. .do. _do. .do. .do_ . . Squareyards Feet B. M — _do. .do. Cubic yards. . Pounds . Miles. Cubic yards. Acres. 61,129 217,969 3,40-1 595,707 11,500 8. 1,396,943 301,827 715,477 236,336 113,883 106,736 80.808 20 $100, ooo 12,230 144,100 172,069 129,515 221,494 34,555 515,923 195,878 4,490 71.280 5,919,868 1.577,428 90,985 660.987 88,532 54 10 . 9,750 2,437,748 11 . 187. .'6,895 75,720 3,964 .8,518 . 956 610 586 060 286 744 415 :164 933 927 212,120 683,376 163,536 10,350 76.385 4.6 1 1,110 $200 12 . 99,810 775.043 306,172 l, 100.101 72,577 488,410 79.680 7,016,8144 1.778,124 115,348 749,020 ■I 183 Classification of material. Excavation under water: Rock . Hardpan . Hard material. Clay.- ... Sand . Gravel .. Excavation in the dry: Rock .. . Hardpan.. Hard material . Clay .. Sand . Gravel. Embankments .. Backfill.... .. Retaining walls. Slope walls . Is- Docks and cribs: Oak.. Hemlock. Pine. Stone fill... Iron, driftbolts Lock No. 1 (Sheldon land), operating ma¬ chinery included. Lock No. 2 (Clarks Point), operating ma¬ chinery included. Lock No. 3 (Massena), op¬ erating machinery in¬ cluded. Highways . Ferry (at highway cross¬ ing). Rock-filled dams. Right of way. Islands near Galops. Water power, etc. Drainage ... Railroad bridge. Total. Unit of quan¬ tity. Cubic yards. .do.. .do. .do. .do. .do. .do. .do. .do. do_ .do. .do_ .do. do_ .do_ Square yards .M Feet B. . do . . do . Cubic yards Pounds . Miles. Cubic yards. Acres. Quantities in sec¬ tion— 13. 1,888,411 41,473 6,295,996 451,686 53.853 393 14. 2,007,839 1,980,485 Total quantity. 5,384,463 931,736 8.174.134 1.127.559 1.980,485 389,673 1,686,239 II1.567,352 7,030,470 16,835,420 1,480.627 1.328.560 586.1*27 789,416 9, *39 285.598 230.640 19,747, 430 5,277,056 313.524 2,200,982 4.6 1 258.022 2,105 $12,700 100. (MX) $200 1 Unit price. $2.50 1.00 .25 .15 .15 .15 . 65 .35 .25 .15 .15 . 15 . 15 .25 4. (X) l.lo 1.45 1 50.00 1 23.00 * 30. 00 <..00 .03 Amount. 200.00 .50 75.00 $13,461,157 931,736 2,043.534 169,133 297,073 58,451 1,096,056 3,698,573 1,757,617 2,525.313 222,094 199,284 88.039 197,354 37.356 333,006 11.532 454.184 158,311 188.114 66.929 916,888 964,003 1,767,422 920 20.000 129.011 157,875 12.700 100 , 00(1 200 222,888 32,285,853 1 Per 1,000 feet. 422 DEEP WATERWAYS Table No. 9 .—Summary of quantities and cost — ,21-foot channel. Classification of material. Unit of quan¬ tity. Quantities in section— 1. O 3. 4. 5. 6. Excavation under water: Cubic yards.. .do. 1,487,913 61,348 368 104,513 12,200 32,010 96,585 818,948 7,160 HarcYmaterial. Clay .. .do. .do. 626,896 29,292 49,610 595,798 1,101,611 25,301 20,850 7,965 16,496 322,999 16,640 10,753 ...do. Excavation in the dry: .do. 126,544 91,589 82,122 45,258 105,317 ...do.. 680,631 . do. _ 960,803 595,036 563,546 70,005 ... do. 277,110 Gravel... _.do. 6,747 Back fill .. .do. 66,768 6,233 .do. Square yards Feet B. M_ 8,125 Docks and cribs: Oak... 79,680 4,356,320 1,606,632 71,716 535,090 50.046 20 86,200 Hemlock. .do.. 498,460 163,536 7,860 58,485 Pine .. ...do. Stone fill Cubic yards.. Iron, driftbolts. Pounds _ Rock-filled dams_ Cubic yards.. Acres. 119,444 124 86,500 Right of way. Islands near Galops. 36 37 47 Classification of material. Unit of quan¬ tity. Quantities in section — 7. 8. 9. 10. 11. 12. Excavation under water: Rock .. Cubic yards.. 447,990 432 52,771 Hardpan. .. do_ Hard material. _do . 93,394 555,395 153,856 49,027 494,765 Clay__ .do. . 23,253 Gravel. .do. Excavation in the dry: Rock.. .do. 7,745 51,446 206,135 14,823 5,490,943 3,226,583 7,479,665 9:15,842 637,275 574,420 18,888 666,739 201,647 816,135 71,855 Hardpan. _.. .do. Hard material.. .do. 197,880 2, 756 . 98,516 49,120 Clay.. _ _. .do. 402,038- Sand ... _ do._. Gravel. .do. 481,258 74,604 Embankments.. _do_ Back fill... _do... 124,636 399,890 Slope walls. Squareyards Feet B. M ... 11,500 212,518 Docks and cribs: Oak. 79,680 79,680 5,393.320 1,184,664 92,120 608,820 Hemlock. do. 498,460 163,536 7,860 58,485 4.6 1 Pine. .do. Stone fill Cubic yards 95,967 632,760 Iron driftbolts_ Pounds _ Highways. Miles. Ferry (at highway cross¬ ing). Rock filled dams. Cubic yards. 88,532 54 Right of way. Acres . 76 20 8100,000 5 1,110 183 Water power, etc. Drainage . 8200 DEEP WATERWAYS 423 Table No. !). —Summary of quantities and cost — 21-foot channel —Continued. Classification of material. Unit of quan¬ tity. Quantities in sec¬ tion— Total quantity. Unit price. Amount. 13. 14. Excavation under water: Rock .... Cubic yards. . 2,102,564 883,919 2,698.976 749,175 222,184 986,084 8,119,968 6,214,738 14,815,809 1.471,119 1,199,884 574,420 591,294 6,233 286,243 239,040 16,381.800 4,975,872 275,523 1,893,640 $2.50 1.00 .25 .15 .15 :S .25 . 15 .15 . 15 .15 .25 4.00 f 1.10 l 1.45 >50.00 123.00 $5,256,410 883,919 674, 744 112.376 33,327 640,954 2,841,988 1,553,684 2.222,371 Hardpan.. .do ... Hard material. .do. Clay .... .do .. Gravel. .do. Excavation in the dry: Rock .. .do. Hardpan._. .do.. 983,931 239,485 5,722,842 463,422 Hardmaterial. Clay . .do. .do. . Sand... .do .. 220,667 179,982 86,163 147,823 24,932 | 3:53,802 11,952 376. 781 Gravel . .do. Embankments. .do.. Back fill. .do. Retaining walls . __do. Slope walls ... . Square yards Feet B. M . 54,100 Docks and cribs: Oak. Hemlock .. .do. Pine... .do. >30. (Kl 149.276 Stone fill_ _ Cubic yards.. . 60 .03 165,314 56,809 641,141 671,506 1,186,042 920 20,000 129,011 157,875 12,700 100,000 200 222,888 Iron driftbolts,-. Pounds . Lock No. 1 (Sheldon Is¬ land), operating ma¬ chinery included. Lock No. 3 (Clarks Point), operating ma¬ chinery included. Lock No. 3 (Massena), operating machinery in¬ cluded. Highways.... Miles. 4.6 1 258,022 2.105 12.700 $100,000 $200 1 200.00 Ferry (at highway cross¬ ing)- Rock-filled dams. Cubic yards .. ..50 $75.00 Right of way - - 393 Islands near Galops. Water power, etc . Drainage .. Railroad bridge_ 1 Total .. 19,115,557 J Per 1,000 feet. Table No. 10. No. of lock. Location. Station. Kind. Elevation (stand¬ ard low water). Lift, in feet. Length of level, in miles. Above. Below. 1. Sheldon Island.. 3779 4372 5508 Single.... . -do 212.7 225. 7 204.5 233.0 213.9 156.8 9.7 11.8 47.7 Clarks Point. 11.2 21.5 3. Massena. _do_ Cost of locks. No. of lock. 30-foot 21-foot channel. channel. Operating machinery 1.. $816,888 864,003 1,667,422 $541,141 471,506 1,086,042 $100,000 100,000 100,000 3. Total... .. . 3,348,313 300,000 2,198,689 300,000 300,000 Operating machinery.. Total cost... 3,648,313 2,498,689 Stations showing location of head of locks, as given in Table No. 10, are for the 30-foot channel. The location of the locks for the 31-foot channel does not correspond exactly with the locks of the 30-foot channel. 424 DEEP WATERWAYS Classification of channel, 30-foot depth. Station. Open water. Improved river. Canal, Lock and earth sec- ap- tion. proaches. From— To- o . 3481.. 348,100 34H1 1,700 3572.. 7,400 3570 3762 + 29 .. 19,029 3762+29 3802. 3,971 3809 3819 . .. 1,700 3819 3,700 3850 . 3885 .. 2.900 3885 11,400 3909 . 4066 .. 6,700 4066 .-. 4200. 14,000 4*>06 .. 4213 . 700 4‘>]3 . 4229 1,600 4'>29 . 4257 2,800 4257 . 4278.. . 2,100 4*>7S . 4350+10.... 7,810 4350+10 4395 + 81.. 3,971 4395+81 . 4460-. 6,419 . 44H0 . 4470. 1,000 4470 4481 . 1,100 4481 . 4490. 900 4490 ... 4610.. 12,000 4610 ... 4020.. 1,600 4643.... 1,700 46413. 4721.... 7,800 4721 .. 4700..... 3,900 4760 . 4910. 15,600 4916 4933 . 1,700 4933 12,200 5Q55 . 5075 .. 2,000 5075 . 5490+60.. 41,560 5490+00 5532 +31. 4,171 55:r>+31 5633 10,069 563:} . 200 5635 . 566S .-. 3,300 5674 .. 600 5682... 800 5082 1,300 5695 .- 5702. 700 5762. 5717. 1,500 5717 5831 . 11,400 5^31 5844..... 1,300 5844 . 5851 ... 700 5851 _ 5872..'.... 2,100 5872 . 5885... 1,300 5885 800 5909 . 1,600 5909 .. 5932.... 2,300 5955 2.300 5955.... 5983 . 2,800 5983 6204 . 22.100 0204.... 6240.... 3,600 6240.... 6268 ..•_ 2,800 626^ - 6288 2,000 6288 .. 6327 . 3,900 6327 4,300 6370. 800 6378. 6388 . 1,000 6388. 6,500 1,100 6404. 6473. 900 6473. 6482 . 900 6482. . 6630 . 14,800 6703. 7,300 6703... 6975. 27,200 6975...'.. 7010 3,500 7010.. 7016. 600 7010. 7028 . 1,200 7028. 7083. 25,500 Total feet. 550,900 104.34 109,958 20.82 55,329 10.48 12.113 2 30 Total miles. DEEP WATERWAYS. 425 Classification of channel, 21-foot depth. Station. Open water. Improved river. Canal earth sec¬ tion. Lock and ap¬ proaches. From— To- 0. 357,400 3574.*. 3763 + 29. i8,929 3763 + 29. 3801 + 30. 3,801 3801 + 30. 5,470 3885 .. 2,900 3885.... 3999 .. 11,400 3999 ____ 4066. 6,700 4066... 4209 . 14,300 4209. 4213 . 400 4213.. f>->9 . 1,600 4229 2,800 4257. 4278 2,100 4278. 4348 . 7,000 4348 . 4362 + 10. 1.410 4362 +10 4400 + 11 3,801 4400 + 11. 4455 '... 5,489 4455 . . 4531 7,600 5431... 400 4535.. 3,400 4569... 900 4578. . . 4726 . 14,600 4748. 2,200 4748... 4916. 16,800 4916. 4919 300 13.600 5055..... 5062.. 700 5062. . 300 5065. 1,000 5075 5491 +60 41,660 5491 + 60 4,001 5531 +61 . 10.139 5633. . . 72S) .. 105.000 Total feet . _ . 611,480 115.81 49,718 9. 42 55.499 10.51 11.603 2.20 Total miles _ In conclusion, Edw. B. Hitchcock, Glenn D. Holmes, instrument men; Charles F. Howe and George A. Hammond, superintendents of borings, and John Y. Bayliss, George D. Williams, and William P. Boright, recorders, should be mentioned. I would acknowledge the favors extended and the information fur¬ nished by Mr. Tom S. Rubidge, superintending engineer of the St. Lawrence district, Cornwall, Ontario, regarding Canadian gauge read¬ ings and soundings through a portion of Lake St. Francis. Very respectfully, J. W. Beardsley. Assistant Engineer. The Board of Engineers on Deep Waterways. Appendix No. 12. CHAMPLAIN ROUTE, NORTHERN DIVISION. Detroit, Mich., July 27 , 1899. Gentlemen: I have the honor to submit the following report in regard to the surveys and estimates of the cost of a ship canal between Lake Champlain and the St. Lawrence River DEEP WATERWAYS. 426 A study of maps of the region between Lake Champlain and the St. Lawrence River in the State of New York shows the country to be mountainous and entirely impracticable for a ship canal. The country immediately north of the international boundary line is quite high and rolling, with deep valleys. The drainage of this region is to the north and northeast. The most important topo¬ graphical features are a series of nearly parallel glacial valleys, with ridges of rock between, running nearly north and south in the eastern portion, but swinging to about northeast and southwest as the St. Lawrence is approached. At a distance of about 10 miles north of the boundary line the hills break down and the country becomes generally level. In the hilly country adjacent to the boundary line, and about 10 miles west of the village of Champlain, N. Y., are found the head¬ waters of the Chazy River, which empties into Kings Bay, Lake Cham¬ plain ; the La Colle River, which flows a little north of east and empties into the Richelieu River; the Little Montreal River, which flows first north, then northeast, and empties into the Richelieu River; Norton Creek, which flows parallel to and about 2 miles west of the Little Montreal for about 10 miles, where it turns to the west and empties into the English River, which flows nearly northwest and empties into the Chateauguay River. A careful examination of the country bounded on the north by Nor¬ ton Creek, south by the international boundary line, east by Norton Creek, and west by the Chateauguay River showed that there was no practicable line for a canal south of Norton Creek. From a point on Norton Creek near Aubrey Station to the St. Law¬ rence River at Lake St. Francis the country is in general a plain, much of the distance below the level of Lake St. Francis. About midway between these points is the valley of the Chateauguay River, which is 40 feet below the level of Lake St. Francis. A line drawn from the mouth of the Chazy River to a point on Norton Creek where it turns to the west was found to pass near low divides between the heads of adjacent streams and to offer a practica¬ ble route. This is essentially the route suggested by the commission of 189(3. From Aubrey to Lake St. Francis the route suggested by the com¬ mission of 1896 did not appear to be the best, inasmuch as it kept to the north, getting on ground so low that embankments from 15 to 25 feet in height would be required for many miles. On this account it was thought best to keep farther south and get on higher ground. Standard low water in Lake St. Francis, which is practically low water during season of navigation, is 152.39 feet above mean tide at New 5 ork. The summit level for the canal has therefore been taken as 152.4, and an effort made to keep on ground which would give embankments of reasonable height. From Aubrey to Ormstown the country immediately adjoining the plains is a solid mass of rock at DEEP WATERWAYS. 427 an elevation of 220 feet or more. The surveys were kept as close to this rock as practicable. From Aubrey to Champlain the country is all above the level of the lake, and the line was kept on the lowest ground. The surveys have been made with a degree of care commensurate with the importance of the work, and in accordance with the “General instructions to held parties” (Appendix No. 0). The levels were run in one direction and checked in the reverse direction. The limit of error allowed was C=0.05 feet x V distance in miles. The error was not allowed to exceed this amount, either for the whole or any part of the line. The levels were started from the United States Coast and Geodetic Survey bench mark on the Chapman Block at Rouse Point, N. Y., with an elevation of 110.00. The field maps and profiles have all been referred to this datum. A later determination of the elevation of this bench mark, made under the direction of Mr. C. L. Harrison, gives the elevation above mean tide in New York Harbor as 108.95, when referred to the bench mark at Greenbush, N. Y., with an eleva¬ tion of 14.73. All elevations used in this report will be referred to the later datum. At Valleyfield connection was made with a bench mark on lock No. 14, Beauharnois Canal, established by Mr. Thomas Monro in 1891, by a line of levels brought across from the Chapman Block bench. Our elevation w r as 0.121 foot higher than his. From one of our benches near St. Stanislas we carried a line of levels to Hogansburg, N. Y., to connect with a bench mark which had previously been established by Mr. David Molitor. Wherever the country seemed to offer a choice of locations, alternate lines were run and the topography developed. The original base line between Ormstown and Lake St. Francis w T as run via St. Stanislas. An alternate line beginning about 2 miles west of Ormstown was run to the lake, a distance of 9.3 miles. This line is the one used for the location. Another line beginning about 4 miles east of Ormstown was run to the lake, a distance of 13 miles. This line kept to the north of the other lines on lower ground, and the embankment was found to be excessive. On this account no topography was taken. For most of the distance it was outside the limits of the maps and is not shown. The total length of base line run and leveled was 74.51 miles. In addition to this a stadia line about 9 miles long was run, and topography taken to develop what is known as the “ Bogtown line.” It was thought that this line would give less rock excavation than the base-line route, but this proved not to be the case. The proposed location of the canal in Lakes Champlain and St. Francis was covered by soundings. These soundings are reduced to elevations above mean tide at New York. They were carried out to 428 DEEP WATERWAYS. a point such that the elevation of the bottom was well below that of the established grade. The United States Coast and Geodetic charts of Lake Champlain and the Canadian charts of the northern end of Lake St. Francis, furnished by Mr. Thomas Monro, together with a line of soundings from the southern end of the Canadian chart to Cherry Island, show a sufficient depth of water along the sailing line. Borings to determine the character of the materials to be excavated have been made at intervals of 1,000 feet or less on all lines projected and covering a width of from half a mile to a mile. These have in all cases been carried to rock or to the proposed bottom of the canal. Samples of the materials from all holes have been preserved and properly labeled for identification. In all 1,387 borings were made. In addition to these, four diamond- drill borings have been made to determine the character of the rock. These holes were located as follows: No. 1, near station 9904 of the proposed canal location. No. 2, near station 9348 of the proposed canal location. No. 3, near station 9075 of base line, at Holton Station, on the Canadian Atlantic Railway. No. 4, near station 810G of canal location at the crossing of the Chateauguay River. A report of these borings has been submitted, in Appendix No. 19, by Mr. R. C. Smith, under whose direction they were made. CONDITIONS GOVERNING THE LOCATION. Lake Champlain .—The lowest water ever recorded in Lake Cham¬ plain was on October 12,1880, when the reading on the gauge was 0.14. (Appendix No. 7, Report of the United States Coast and Geodetic Survey, 1887, p. 170.) The zero of the gauge is 93.501, and low water 93.3G1. The Coast Survey report gives the elevation of the zero of the gauge as 94.53. I had a line of levels run from the Chapman Block bench to zero of gauge and checked by reverse running, the two agreeing within 0.028 foot. Using the mean of the two runnings, the elevation of the zero of the gauge as compared with the Chapman Block bench is 94.61, or 0.08 foot higher than the elevation as given by the Coast Survey. Applying the correction of 1.109 to reduce to Greenbush bench, our elevation of the zero becomes 93.501, as stated, and low water 93.361. The highest water recorded appears to have been on May 4, 1869. This was prior to any records being kept at Fort Montgomery. It was recorded at the railroad bridge of the Central Vermont Railroad at Rouse Point, and is given as being 9.25 feet above low water of Octo¬ ber 12, 1880 (Report of United States Deep Waterways Commission, 1896, p. 124), making extreme high water 102.611. This elevation appears to be well authenticated. Lake SI. Francis .—The report of the Canadian Deep Waterways DEEP WATERWAYS. 429 Commission gives, on the authority of Thomas Monro, esq., member of the commission and engineer of the Soulanges Canal: Extreme low water, November, 1895, 151.88—0.988 = 150.892. Extreme high water, 150.80—0.988 = 155.812. The correction of 0.988 was arrived at as follows: At Valle}'field our levels were 0.121 foot higher than the bench mark established by Mr. Monro, which being subtracted from 1.109, the cor¬ rection for Rouse Point bench, gives 0.988 foot. The elevations of some of the principal points will be given below for more convenient reference: Bench mark on Chapman block, Rouse Point... 108.951 Bench mark. United States Engineers, Fort Montgomery .... 94.998 Upper miter sill, lock No. 1, St. Johns 1 - - .. ... 86.278 Zero of United States Engineers' gauge, Fort Montgomery. 93.501 High water, Lake Champlain... 102.611 Low water, Lake Champlain .-..... 93.361 High water, Lake St. Francis.-....155.812 Low water. Lake St. Francis ..... . 150.892 Standard low water, Lake St. Francis . 152.390 Bench mark on lock No. 14. Valleyfield..... . .. 155.682 Datum plane for Coast Survey charts, Lake Champlain... 93.861 In order to lessen the excavation between Whitehall and Albany, as well as the dredging in Lake Champlain, it is desirable to keep the lake at as high an elevation as practicable. For this reason it was directed by the Board that estimates be based on maintaining the low-water stage of Lake Champlain at an elevation of 100 feet above mean tide at New York by regulating works near the foot of the lake. It would not be practicable to raise the level of Lake St. Francis, nor would it be desirable to increase the height of the embankments across the yalley of the Chateauguay River and the Aubrey Plains. The elevation of water surface in the summit level has, therefore, been fixed at 152.4, which is the elevation of standard low water in Lake St. Francis. As the difference between high and low water in Lake St. Francis is 4.92 feet, a guard lock at the entrance of the canal will be necessary. As before stated, it is desirable to regulate the surface of Lake Champlain so that it will not fall below 100 and at the same time not go materially above the present high-water mark of 102.61. By putting a dam across from Stony Point to Windmill Point and mak¬ ing the crest of the dam 100 the surface would be maintained at 100 'The elevation of the miter sill is given in the report of the Deep Waterways Commission for 1896 as 87.41. This appears to be an error. Mr. R. Steckel. engi¬ neer in charge of the Canadian geodetic leveling, gives it as 7.22 feet below the zero of the United States Engineers’ gauge at Fort Montgomery. The zero of this gauge as used by the commission was 94.53, giving 87.31 for the level of the lock sill, or 86.278 when reduced to the Greenbush datum. 430 DEEP WATERWAYS. or more, provided the inflow during the dry season was sufficient to supply the evaporation and the minimum amount now flowing in the Richelieu River. As the Chainbly Canal takes its water from the river, and as there are valuable power plants at Chainbly, the regular flow of the river could not be interfered with. The report of the United States Deep Waterways Commission for 1896 shows the monthly mean of water levels above the zero of the United States Engineers’ gauge at Fort Montgomery for the years 1871 to 1895, inclusive. Col. J. N. Barlow, Corps of Engineers, U. S. Army, furnished the daily record at the same place from January, 1896, to May, 1899, inclusive. The record of the daily fluctuations for twenty-seven years at lock No. 1, St. Johns, Province of Quebec, was obtained from Ernest Marceau, superintendent engineer of railways and canals, Montreal, Canada. Information was also furnished by Mr. P. P. Benoit, super¬ intendent of the Chambly Canal. A study of these records for many years shows that the lake is ris¬ ing from November to May, inclusive, and falling from June to Octo¬ ber, inclusive. During the months of August, September, and October the lake appears to fall on the average about 0.25 foot per month, and the river appears to be drawing on the reservoir supply in the lake to a certain extent. A fall of 0.25 foot per month over the area of the lake would furnish a supply of about 1,000 cubic feet per second. Thus it would seem, in order to supply the present flow and maintain the lake at a constant level during the months of August, September, and October, a supply of 1,000 cubic feet per second would have to be supplied through the canal from the St. Lawrence River. In addi¬ tion to this, enough would have to be drawn from the St. Lawrence to supply the water needed to operate the upper portion of the canal from Whitehall to Albany. As the sectional area of the canal is about 7,500 square feet, this could not create a current of more than a half to three-quarters of a foot per second, which would not be objectionable. Plate 91 shows the general location and details of the proposed regulating works. At Stony Point and at Windmill Point there are outcrops of rock which is said to be UJica shale. At Windmill Point the rock appears to be close to the surface for some distance back from the point. Between Stony Point and the mainland there is a swamp, in which borings showed mud 5 feet deep and clay 57 feet without striking rock. On the mainland a short distance to the west rock is found in wells at an elevation of about 100. The borings show a stratum of clay extending from Stony Point to Windmill Point. As the head on the dam would never exceed about 7 feet, a crib dam founded on this clay stratum would undoubtedly be perfectly safe. A lock and sluices can be put in on rock founda¬ tions at Stony Point. DEEP WATERWAYS. 481 As the Canadian government is enlarging its canals to a depth of 14 feet, with locks 42 by 280 feet, the lock would probably have to be of the same dimensions in order to accommodate their traffic on the Richelieu River. Between Stony Point and the mainland an earth embankment would be perfectly safe. The details and estimate of cost of the Champlain regulating works are given in Appendix No. 8. I have secured the following gauging of the Richelieu River: October 16, 1862, by Mr. Charles Legge; 1 measurement near St. Johns: Discharge in cubic feet per second_.....4,257 September 10, 1894, by Mr. Henry Holgate; 2 measurement near St. Johns: Reading of gauge at St. Johns . ........ 7.58 Surface of water at St. Johns above tide ... . 93.85 Discharge in cubic feet per second _..... 6,390 August 22,1895, by Mr. Cecil B. Smith; * * 5 measurement near St. Johns: Reading of gauge at St. Johns .-.-.. 7.67 Surface of water at St. Johns above tide....93.94 Discharge in cubic feet per second.____ 6.102 October 17. 1895. by Prof. C. H. McLeod:’ measurement near St. Johns: Reading of gauge at St. Johns ._ 1...-... 7.00 Surface of water at St. Johns above tide....93.27 Discharge in cubic feet per second. .... 3,750 1895. by Shanly and Quirk; 1 measurement near St. Johns: Discharge in cubic feet per second..7,000 I have not been able to get the date of this, but the water is reported to have been at about the same stage as when measured by Messrs. Ilolgate and Smith. April 28, 1899. by J. W. Macklin; weir measurement at dam of Chambly Power Company, Chambly, Quebec: Length of weir ..... ... ...feet.. 1,625 Depth of water on weir measured to surface of still water above . ..do... 3.20 Reading of gauge at St. Johns...... 11.75 Surface of water at St. Johns above tide. . 99.13 Discharge in cubic feet per second, as computed by Mr. Macklin (pass¬ ing over weir). ... .... .. 29,050 Add passing through sluice to mill. 800 Total ... . 29,850 At the time the above gauging was made there was no wind and t lie river had been at about the same stage for several days. This is probably near a high-water discharge. The gaugings by Messrs. Holgate and Smith were made with tubes weighted so as to float vertically and reaching nearly to the bottom with but little projecting above the water. That by Professor Mc¬ Leod was made with an “Amsl.er mechanical meter.” 1 Furnished by Ernest Marceau. esq., engi neer of railways and canals. Montreal Canada. 5 Furnished by Mr. J. W. Macklin. engineer of the Chambly Power Company. / 432 DEEP WATERWAYS. The low-water measurements are somewhat discordant and hard to reconcile, as there was only a difference of 0.67 foot between the read¬ ings of the gauge when the difference was greatest. As the river near St. Johns has a width of about 1,200 feet and a mean depth of 12 to 13 feet, a variation of 0.67 foot ought not to make so great a difference as is shown by the measurements. Lochs .—A lock with a lift of 52.4 feet has been located about 4 miles from Lake Champlain and just southeastof the village of Cham¬ plain. At this point rock is found at an elevation of 110, giving a rock foundation with the lower 40 feet in rock. If it should be thought best to substitute two locks with lifts of 26.2 feet each, the lower lock can be put about 6,500 feet farther east on rock foundation. The cost would probably be about the same in either case, but the single lock possesses the advantage of less deten¬ tion and will somewhat simplify the passing of the water supply for Lake Champlain. In the event of one lock being used, the water can be taken out above the lock and discharged into the Chazy River after being used to generate power for operating the lock, etc. The esti¬ mates were made for a canal with only one look at this point. As high water in Lake St. Francis is 3.4 feet above standard low water, a guard lock with a maximum lift of 3.4 feet will be necessary near the entrance to the canal. A favorable location has been selected at a point about three-quarters of a mile from the lake, where the sur¬ face of the rock has an elevation of 140. A by-pass of sufficient dimensions to pass 4,000 to 5,000 cubic feet per second will have to be constructed around this lock. The power for operating the lock can conveniently be brought from a power house located at the crossing of the Chateauguay River, where a head of 30 feet would be available. Receiving weirs .—At the crossing of the Chazy River above Cham¬ plain the surface of the water in the river is about 12 feet above the level of water in canal. Rock is found at an elevation of about 160. Just below this point there is a dam with a head of 20 feet. The vil¬ lage of Champlain has acquired the right to use whatever water is needed for a supply, as well as the power for pumping. The surplus power, if any, is owned by Whiteside Brothers, of Champlain. When this stream was gauged on April 25, 1899, it was carrying 1,038 cubic feet per second. This may be increased in time of high water to 3,500 cubic feet, and in time of low water may fall as low as 100 cubic feet or less. As taking the water in the canal would destroy the power at the waterworks and Whiteside’s mill, it is proposed to move the water¬ works to the point of intake. This will necessitate laying about 4,000 feet of discharge main and the building of a new pump house. The DEEP WATERWAYS. 433 turbine and pump now in use could be moved. The present pump house is a brick structure about 15 feet square. The head now util¬ ized at the pump house is 12 feet. The same head can be obtained where the river is taken into the canal. The supply can be taken from the river above the canal and the discharge main carried under, thus furnishing the same quantity and quality of water as they now have. To do this, the two branches of the Chazy ought to be united above the canal and a dam about 200 feet in length and 5 feet in height founded on rock be built to turn the water through the pump house with a tailrace discharging into the canal. As the surface of the rock is at an elevation of about 160, no special construction is needed for receiving whatever water passes over the dam into the canal. Whenever all the water is not used by the waterworks, the surplus belongs to Whiteside Brothers and has an available head of 20 feet. Formerly this entire power was used to run a strawboard mill; but this has not been in operation since 1895, and since that date the right to use whatever water is needed has been transferred to the village of Champlain. It is probable that 100 horsepower might be obtained from the sur¬ plus water for six or seven months in the year; but, as this would neces¬ sitate a steam plant of equal capacity for several months, it is doubt¬ ful if the surplus water has much commercial value. About a mile farther down the stream there is another power. It would be necessary to return water to the streams above this dam equal to the amount taken in above. This can be easily done. At the crossing of the Little Montreal River and Norton Creek the streams will have to be taken into the canal. As rock is found near the surface at both points, no receiving weirs will be needed. These streams will have to be carried across the canal in flumes and diverted during construction. The bed of English River, where the canal crosses it, has an eleva¬ tion of about 125. When gauged April 26, 1899, it was carrying 392 cubic feet per second. The maximum discharge may reach 1,000 cubic feet per second. As its elevation is above the bottom of the canal, to pass it under would require an inverted siphon, which would be not only expensive but objectionable on account of the heavy ice which forms in that region during the winter. For these reasons it was thought best to flood the valley above the canal and pass the water over a wasteweif. The area which would be flooded is about 1,000 acres. A waste weir has been located at a point about three-fourths of a mile east of the river crossing. The water will be discharged into Norton Creek about a mile above its junction with the English River. The waste weir will be founded on rock at an elevation of 140. H. Doc. 149-28 434 DEEP WATERWAYS. The Chateauguay River is the largest stream crossed, and presents one of the most difficult problems on the line. The following data in regard to the stream have been obtained: Above the dam at Ormstown: High water, April 11, 1887. in shoe store of W. Maw.. 136.45 Surface of water September 14, 1898 ...120.43 Crest of dam, about..... 120.00 Below the dam: High water. April, 1887, on Murphy's home, G miles below Ormstown. 136.00 High water, April, 1899... ...... 119.00 Surface of water September 14, 1898 ..110.14 On April 19, 1899, the river was carrying 2,720 cubic feet, with water surface at 113.29, 1^ miles below the dam. T do not know of any high-water gauging. The maximum discharge may reach as high as 5,000 or 6,000 cubic feet per second. Extreme high water is always caused by an ice gorge, which usually forms at the Grand Trunk Railway bridge, about 3 miles below town. T have to suggest three methods of carrying the canal across the river: First. By an aqueduct, the river being passed under the canal. Second. By putting a dam across the river and flooding the valley above. 1 Third. By a dam and dikes on both sides to a point above the flowage line. The various projects will now be considered. First. To pass the river under the canal an aqueduct from 600 to 800 feet in length would be required. The conditions for this are not as favorable as could be desired. Low water in the river is about 110, ordinary high water 120, and extreme high water 136. The grade of bottom of canal with 30 feet of water is 122.4. Owing to the large amount of ice which the river sometimes carries, the openings ought to be kept as large as possible; but on account of the great weight to be carried they ought to be of moderate spans. Spans of about 30 feet will probably best fulfill the conditions. From the elevations given above it will be seen that there is very little headway, and at ordinary high water the crown of the arch will be submerged. In order to give all the headway possible, the feasibility of carrying the aqueduct on steel girders with 30-foot span was considered. To do this with 24-incli I-beams weighing 80 pounds to the linear foot would require them to be spaced 10 inches between centers under the body of the aqueduct. As the side walls would weigh about two and one-lialf times as much as the water, the walls would have to be car¬ ried on deep girders built into the walls. As beams spaced 10 inches apart could not be painted, they would have to be encased in concrete to protect them. This leads at once to the Melan system of arches. DEEP WATERWAYS. 435 If arches were made 5 feet thick at the crown, which seems to be as much as is practicable under the circumstances, they would be lack¬ ing in strength to resist the pressure of the water tending to force the side walls of the aqueduct apart. This thrust for the 30-foot channel amounts to over 28,000 pounds per linear foot of the canal. As this force acts with a lever arm of 10 feet, there would be a moment of 280,000 foot-pounds for each linear foot of the aqueduct, tending to rupture the arch in a plane parallel to the axis of the aqueduct. A con¬ crete arch under these conditions would not be stable. It would be necessary to strengthen the arches with steel rods or beams in a direc¬ tion transverse to the axis of the aqueduct. Steel rods would prob¬ ably also be required in the side walls to prevent cracks from tem¬ perature strains. Ice forms in the streams in this region to a depth of 3 feet or more. It would undoubtedly form to a greater depth in the prism of the aque¬ duct and might rupture the walls unless the water was drawn off at the end of navigation. As the English River and other streams are to be taken in and discharged over wasteweirs, the entire canal can not be emptied. On this account gates will be necessary at both ends of the aqueduct, and a pipe with a gate carried through the wall to empty the water. The ice which runs in the river passes over a dam about half a mile above our crossing, and would be pretty well broken up before reach¬ ing the aqueduct; but as an additional safeguard it would be well to put in masonry ice breakers a short distance above the aqueduct, spaced less than 30 feet apart, so that any ice which had passed them would pass through the openings in the aqueduct. By doing this the ice would probably pass in safety at all ordinary times. There is, however, danger of an ice jam forming at the Grand Trunk Railway bridge, 3 miles below, and setting back to this point. In 1887, and again in 1888, this happened, and the water rose to a height of 136. This might cause an ice gorge above the aqueduct and raise the water still higher. If the aqueduct was filled it would be stable, even though the water went over the top of it; but any rise of water above that of 1887 would do serious damage to people in Ormstown and vicinity. By adopting either the second or third plan proposed the danger from floods and ice would be lessened. By putting a dam across the river above town and flooding the valley there would be submerged, as nearly as can be ascertained from data at hand, about 8,000 acres of land, which would probably be valued at from $75 to $100 per acre. In addition to this, the Grand Trunk Railway would have to be reconstructed for several miles. By building dikes along both sides of the river to a point where there would be no danger of flowage, the stream could be taken in DEEP WATERWAYS. 436 and discharged over a waste weir with safety, and probably at no greater cost than for an aqueduct. 1 * * * * * For a 21-foot channel there would be no trouble in passing the river under the canal. The line lias been so located that by building the western end first and then changing the channel the whole aqueduct can be built in the dry. Siphons .—There are several small brooks and ditches which cross the line at such an elevation that they will have to be passed under through inverted siphons. The} 7 will appear in the estimates. Discharge sluices .—Discharge sluices ought to be built at the fol¬ lowing points: At lock east of Champlain, capable of discharging 4,000 cubic feet per second. Between stations 9821 and 9791 of canal, one capable of passing 2,000 cubic feet per second. At station 8666 of canal line, one to discharge not more than 3,000 cubic feet per second. This is to pass the water of English River. I doubt if it would be safe to discharge more at that point. It would be liable to flood the valley below. Railroad crossings .—The following railroads are crossed: Delaware and Hudson, Ogdensburg and Lake Champlain, Ilemmingford Branch of Grand Trunk, Canada Atlantic, Massena Branch of Grand Trunk, and St. Lawrence and Adirondack. Estimates have been made for a change of location of the Delaware and Hudson, both branches of the Grand Trunk, and the Ogdens¬ burg and Lake Champlain railways to secure crossings at right angles. All railway crossings are by means of swing bridges. The Canada Atlantic crosses the proposed canal at two points, and an esti¬ mate has been made of a new location of this road for a distance of about 9 miles to avoid any crossings. All the embankments on these changes can be made from waste material from the canal excavation. Highway crossings .—All highways have been estimated to be car¬ ried over on swing spans or fixed spans with a clear headway of 85 1 After Mr. Davis had severed his connection with the Board, this subject was further considered and additional surveys made in October, 1899. by Mr. James J. Overn. This survey consisted of lines run along the high ground adjacent to both banks of the Chateauguay River from Huntingdon to Ormstown, and the infor¬ mation secured was sufficient to enable an estimate to be made of the embank¬ ments required by the third method mentioned above. This survey is not shown on the accompanying maps. The proposed location of the canal was changed, as shown on plates 45 and 46, in order that the dam might be located on rock foundation above the village of Ormstown. The estimated cost of the embank¬ ment is $405,000, and of the required right of way $117,000, making a total of $522,000. Method No. 8 is cheaper than No. 2 and avoids the uncertainties of the aqueduct plan. The location and estimates have been made in accordance with method No. 3 for both 21 and 30 foot channels. DEEP WATERWAYS. 437 feet, or by steam ferries, with pontoon bridges to be put across in place of the ferries during the period when the canal is not in opera¬ tion. The ferries and pontoons would be not only much cheaper but less liable to any accident which would block the canal than swing bridges. Ferry crossings have been estimated for the following points: At Cooperville, at Champlain, near Aubrey Station, at Fertile ('reek road, at Ormstown. Table No. 1 gives the location, length, cost, etc., of all railway and highway bridges estimated. Classification of excavation. —The excavation can be classified under the following heads: Dredging, which would cover all materials which could best be removed with a dredge; earth excavation, cover¬ ing all earth to be removed otherwise than by dredging; solid rock under water; solid rock above water. All the excavation required in Lake Champlain will be dredging in earth. In Lake St. Francis there will be a small amount of rock. Character of materials. —The earth is mostly a stiff clay. Between the following stations there will be found many bowlders: 9881 to 9241, 8941 to 8741, 7841 to 7721. Between 9241 and 8941 there will be found a considerable amount of black muck. The depth is so very irregular that no attempt has been made to separate it from clay in the estimates. I think most of the clay will stand at slopes of 2 to 1. However, there is some in the vicinity of the Chateauguay River and near the St. Louis River which may give trouble. Last spring some of this became almost like quicksand, the roads being absolutely impassable. On the Soulanges Canal there have been several bad slides on sections which have been excavated to the entire depth and left for two or more years without water being let in. I think the material at the Soulanges and near Ormstown is somewhat similar. Character of rock to he excavated. —The surface rock appears to have been originally Utica shale, below which was limestone, chang¬ ing almost imperceptibly into a calciferous sandstone, one of the Potsdam group, and finally into quartzite. Glacial action has worn this down so that the shale is onl}' found along Lake Champlain and the Richelieu River. Near Champlain and Barrington limestone is found in limited quantities. Near Valleyfield the rock is usually limestone of such a quality that it is used for the manufacture of lime. Quartzite is found outcropping in every ledge of any impor¬ tance along the line between Lake Champlain and Ormstown. It is safe to classify all this rock as hard sandstone or quartzite. Between Ormstown and Lake St. Francis the rock may be classified as lime¬ stone. Samples from most of the ledges in the vicinity of the line are submitted. 438 DEEP WATERWAYS. The rock will all make excellent concrete, and enough building stone for most purposes can be found in the immediate vicinity of the line. At Champlain blocks 2 feet in thickness and 10 to 12 feet square can be quarried. Photographs showing the stratification and seams in some of the ledges are given. The quartzite occurs in strata varying from a few inches to several feet in thickness, and is broken up by many vertical seams. It can be channeled unless the vertical seams interfere. Ground water in cuts .—As several of the borings develop flowing wells, I anticipate that much water will be found in the deep cuts. Sand .—There is no good bank sand in the vicinity of the line. Sand for building purposes is found in both lakes, but what I saw was not of a first-rate quality. An excellent sand for concrete could be made by crushing quartzite. Details of alignmen t. Length of line from 30 feet of water in Lake Champlain to 30 feet in Lake St. Francis ... ...miles.. 53.674 Distance between shores of lakes ... .. do... 48.66 Length of tangent. ...... ..do— 38.49 Total length of canal: Radius 5,000 feet.....feet.. 15,567 Radius 6,000 feet.... .do... 11,115 Radius 7,000 feet......do... 13,175 Radius 10,000 feet...... * ..do 10,715 Radius 12,000 feet.. ... ..do... 12,700 Radius 17,088 feet........do... 5, 269 Radius 10,111 feet. ...,. do... 11,634 Total length of curve....miles.. 15.18 Character of channel. Earth, and earth with rock less than 5 feet above bottom .miles.. 17. 784 Rock section. . . . ..... do 30.871 Lake section with dredging in bottom..do 5.019 Profile. Proposed level of Lake Champlain ..... 100.00 Proposed summit level....... 152.40 Lift of Lock No. 1. . ... 52.4 Low water, Lake St. Francis... . 150. 89 High water, Lake St. Francis.. ... 155.81 Extreme lift of guard loca. ... 3. 4 Table No. 1 shows the location and cost of the proposed bridges; Table No. 2 shows the location and cost of the proposed locks; Table No. 3, detailed estimates of 30-foot'channel; and Table No. 4, the estimate for 21-foot channel. DODGE'S QUARRY, CHAMPLAIN, VT. Upper 4 feet sandstone Below that limestone. QUARTZITE OUTCROP, 1 MILE WEST OF DODGE'S QUARRY QUARTZITE OUTCROP, 1 MILE SOUTH OF HOLTON STATION, P. Q. QUARTZITE OUTCROP, 1 MILE SOUTH OF HOLTON STATION, P. Q DEEP WATERWAYS 439 Table No. 1.— Champlain route, northern division. BRIDGES. Location. Station. Kind of bridge. Number of j tracks. Swing or fixed. Number of spans. 30-foot channel. 21-foot channel. Total length. Esti¬ mated cost. Total length. Esti¬ mated cost. 7571 Highway. Swing.. i 235 $19,986 195 $16,650 7674 Railway. 1 _do i 537i 139,882 5171 118,772 777(1 Highway. _do i 555 68,476 531 63,430 Ormstown_ 8(171 Railway. 1 .do i 537} 139.882 5171 1 18,772 8923 Highway_ .do.. i 600 72,910 575 67,470 9259 Railway and 1 .do. - 3 600 217,730 5761 201,023 highway. 1 9446 Highway. Fixed .. 1 300 31,499 282 28,679 9652 _do __. Swing.. 1 570 69,572 546 6)3,683 Champlain . 9752 Railway i .do.. 1 5t>4 112,838 540 103,810 Cooperville. 10048 .do .1. l .do.. 1 5371 111,536 5171 104,320 Bridges not over canal. Ormstown.. 8121 Highway. Fixed .. 1 150 10,395 150 10,395 8682 .do.. .do.. 1 100 8,350 100 8,350 Dewittville.. .do.. .. .do.. 120 9,893 120 9,893 Above Ormstown 5,000 2 5,000 Total. — 1,017,949 920,247 1 Double-deck drawspan, two 63-foot girders. 2 New abutments. Note.—H ighway bridge, 22 feet clear opening; single-track bridge, 14 feet clear opening; double-track bridge, 26 feet clear opening; single-track double-deck bridge, 26 feet clear opening. Table No. 2. —Champlain route, northern division. LOCKS. 1 No. Station, upperend Length of level. Elevation, low-wa¬ ter surface. Lift. Remarks. masonry. Above. Below. 4. 7566 Miles. 39.0 152.4 152.4 00 Guard lock at Valley Field. Champlain lock. 5. 9876 43.7 152.4 100.0 52.4 1 Lock at Lake Champlain regulating works is given in Appendix No. 8. COST. Lock No. 30-foot chan¬ nel. 21-foot chan¬ nel. Operating machin¬ ery. 4 ... $725,346 1,801,307 $450,634 1,156,198 $100,000 100,000 5. Operating machinery...... w ss gg S £ 1,606,832 200,000 200,000 Total . .... 2,726,653 1,806,832 Table No. 3. —Champlain route, northern division. ESTIMATE OF COST OF CONSTRUCTION OF 30-FOOT CHANNEL. Section No. 1, Station 7395 to 7532, Lake St. Francis: Excavation— Earth, wet, 2,755,626 cubic yards, at 15 cents. $413, 344 Rock, wet, 279,337 cubic yards, at $2.50 . 698,343 Total... . 1,111,687 440 DEEP WATERWAYS. Section No. 2, Station 7532 to 9953. Lake St. Francis to Great Chazy River: Excavation— Earth, dry, 41,179,324 cubic yards, at 15 cents ....$6,176,899 Rock, dry, 57,724, 771 cubic yards, at 65 cents.. 37,521,101 Embankment, excavation necessary furnished from canal prism, 9,133,645 cubic yards, at 15 cents........... . .. 1,370.047 Embankment, Chateauguay River, required excavation not com¬ puted separately, 1,343,500 cubic yards, at 25 cents .. ... _ 335,875 Retaining wall, 437,384 cubic yards, at $4..... 1,749,536 Slope wall, 501,528 square yards, at $1.10.... 551,681 Back fill, 1,315,565 cubic yards, at 25 cents... 328,891 Timber crib— Oak, 26,640 feet B. M., at $50 per M __ 1.332 Hemlock. 3,790,080 feet B. M., at $23 per M... 87,172 Pine, 540,000 feet B. M., at $30 per M .. 16,200 Stone fill, 59,400 cirbic yards, at 60 cents.... 35,640 Iron, 369,720 pounds, at 3 cents... 11,092 Right of way— Town land, 274 , acres, at $1,500.... 41,250 Farm land, 5,453 acres, at $91.50 . 497,440 Farm land, Chateauguay, 1,170 acres, at $100.. 117,000 Railroad changes, 14.07 miles ...... 218,599 Entrance of streams— Excavation, rock, dry, 5,600 cubic yards, at 65 cents.. 3,640 Excavation, earth, dry, 186,700 cubic yards, at 15 cents. 28,005 Gates (sluice and by-pass)..... 26,970 Bridges, 13....... 1,017,949 Steam ferries, 5, at $20,000...... 100,000 Lock No. 4, Yalleyfield (guard lock)...... 725,346 Lock-operating machinery, 1 set single... 100,000 Lock No. 5, Champlain... 1,801.307 Lock-operating machinery, 1 set single...... 100,000 Dam at Ormstown— Concrete. 17,670 cubic yards, at $6...... 106,020 Gate.. ... 3,000 Dam at Champlain, concrete, 1,315 cubic yards, at $6. 7,890 Champlain waterworks extension- iron pipe, 2,471.5 tons, at $40.. .. . . 98,860 Pumphouse........ .. 500 Total.... .. 53,179,242 Section No. 3, station 9953 to station 10101, Great Chazy River: Excavation— Earth, wet, 3,940,896 cubic yards, at 15 cents. 591,134 Rock, wet, 1,048,571 cubic yards, at $2.1. .. 2,097,142 Right of way, farm land, 259 acres, at $100. .... 25,900 Total... 2,714,176 Section No. 4, station 10101 to station 10231, Lake Champlain: Excavation— Earth, wet, 2,880,559 cubic yards, at 10 cents... 288,056 Rock, wet, 2,631 cubic yards, at $2.50_____ 6,578 Total.. . .... 294,634 Auxiliary work, Lake Champlain regulating works... .. 890,244 DEEP WATERWAYS. 441 SUMMARY. Section. Station to station. Total cost. 1.... 7395- 7532 7532- 9953 9953-10101 10101-10231 $1,111,687 53,179,242 2.714,176 294,634 890,244 2 . 3.. 4. .... Lake Champlain regulating works. Total. 58,189,983 Table No. 4. — Champlain route, northern division. ESTIMATE OF COST OF CONSTRUCTION OF 21-FOOT CHANNEL. Section No. 1, station 7403 to station 7533, Lake St. Francis: Excavation— Earth, wet, 392,652 cubic yards, at 15 cents.. $43,898 Rock, wet, 74,041 cubic yards, at $2.50 ... 185,103 Total... 229,001 Section No. 2, station 7532 to station 9953, Lake St. Francis to Great Cliazy River: Excavation— Earth, dry, 34,767,719 cubic yards, at 15 cents. 5,215,158 Rock, dry, 43,533,587 cubic yards, at 65 cents. 28,296,832 Embankment— Excavation necessary furnished from canal prism, 9,540,775 cubic yards, at 15 cents. . ._. . 1,431,116 Chateauguay River, required excavation not computed sepa¬ rately, 1,343,500 cubic yards, at 25 cents.. 335,875 Retaining wall, 262,239 cubic yards, at $4. 1,048,956 Slope wall, 527,487 square yards, at $1.10. 580,236 Back fill, 699,974 cubic yards, at 25 cents. 174,994 Timber crib— Oak, 26,640 feet B. M., at $50 per M. 1,332 Hemlock, 2,764,080 feet B. M., at $23 per M__. 63,574 Pine, 540,000 feet B. M., at $30 per M.... 16,200 Stone fill, 43,200 cubic yards, at 60 cents.... 25,920 Iron, 273,600 pounds, at 3 cents. . .. 8,208 Right of way— Town land, 27£ acres, at $1,500. 41,250 Farm land, 5,453 acres, at $91.50. 497,440 Farm land, Chateauguay River, 1,170 acres, at $100... 117,000 Railroad changes, 14.07 miles. 218,599 Entrance of streams— Excavation, rock. dry. 5,600 cubic yards, at 65 cents. 3,640 Excavation, earth, 186,700 cubic yards, at 15 cents. 28,005 Gates (sluice and by-pass).. 26,970 Bridges, 13 ... 920,247 Steam ferries, 5, at $20,000... 100,000 Lock No. 4, Valleyfield (guard lock). 450.634 Lock-operating machinery, 1 set single lock. 100,000 Lock No. 5, Champlain.. . ... 1,156,198 Lock-operating machinery, 1 set single lock. 100,000 442 DEEP WATERWAYS. Section No. 2, station 7532 to station 9953, Lake St. Francis to Great Chazy River—Continued. Dam at Ormstown— Concrete, 17,670 cubic yards, at $6...... $106,020 Gate....- 3,000 Dam at Champlain, concrete, 1,315 cubic yards, at $6.. 7,890 Champlain waterworks extension— Iron pipe, 2,471.5 tons, at $40........ 98,860 Pumphouse.. ....... 500 Total........ 41,174,654 Section No. 3, station 9953 to station 10101. Great Chazy River: Excavation— Earth, w T et, 3,061,214 cubic yards, at 15 cents. 459,182 Rock, wet, 336,535 cubic yards, at $2 .... 673,070 Right of way, farm land, 259 acres, at $100... 25,900 Total.... . 1,158,152 Section No. 4, station 10101 to station 10171. Lake Champlain: Excavation, earth, wet, 1,049,293 cubic yards, at 10 cents .. 104,929 Auxiliary work, Lake Champlain regulating works . 890,244 SUMMARY. Section. Station to station. Total cost. 1 ..... 7403- 7532 7532- 9953 9953-10101 10101-10171 ' $229,001 41.174,654 1,158,152 104,929 890,244 2 . ...... 3 ......... 4 ....... Lake Champlain regulating works. Total.-.-... 43,556,980 I wisli to acknowledge the many courtesies which I received from the Canadian customs officials, who allowed free entry to all of our equipment without question or hindrance of any kind; also to the following gentlemen for valuable data furnished: Mr. J. W. Maeklin, engineer of the Chambly Power Company, Montreal, Canada; Mr. Ernest Marceau, superintendent-engineer of railways and canals, Montreal, Canada; Mr. Thomas Monro, engineer of the Soulanges Canal, Coteau, Canada; Mr. J. F. Beique, superintendent of the Beau- harnois Canal, Valleyfield, Canada; Mr. J. P. Benoit, superintendent of the Chambly Canal, Chambly, Canada; Capt. D. White, Rouse Point, N. Y. The names of the assistants who filled the most important positions on this division are P. H. Aslimead, John J. L. Houston, Y. W. Kline, M. G. Barn es, E. B. Wheeler, S. D. Woodward, instrument men; Charles G. Weyl, draftsman; Jason F. Stearns, superintendent of borings. Very respectfully, Frank P. Davis, Assistant Engineer. The Board of Engineers on Deep Waterways. DEEP WATERWAYS. 443 Appendix No. 13. OSWEGO-MOHAWK ROUTE, WESTERN DIVISION. Detroit, Mich., September 30, 1899. Gentlemen : I have the honor to report as follows upon the western division of the Oswego-Mohawk route: GENERAL DESCRIPTION. The Oswego route has been generally described as leaving Lake Ontario at Oswego, N. Y., following the Oswego River southward from the lake about 21.5 miles to Phoenix, then turning eastward to Oneida Lake, running across the lake to Wood Creek, along this creek to Rome, and down the Mohawk Valley from Rome to the Hud¬ son River. The portion of this route which has been designated as the western division extends from Oswego to Herkimer, a small village, about 14 miles east of Utica, N. Y. It lies mostly in the counties of Oswego, Oneida, and Herkimer, but also passes through the northern part of Onondaga County. Oneida Lake is bordered on the southeast by Madison County. The country traversed is thought to be the bed of an ancient lake, greater than and including Lake Ontario, whose outlet to the sea was through the Mohawk Valley. It is supposed that the region north of the Adirondacks was at that time covered by a held of ice and that the St. Lawrence River was yet unmade. At a still earlier date this field of ice probably extended farther south, so as to cover the route described. 1 In Oswego County the surface of the country is composed of a series of ridges and valleys extending in a southeasterly direction from Lake Ontario. Oneida Lake is 125 feet above Lake Ontario and 371 feet above the sea. It is 21 miles long, averages about 4 miles in width, and receives the drainage from an area of about 1,348 square miles. It is bounded on the north by a strip of table-land, which separates it from Lake Ontario, and on the south by the Niagara escarpment. On the west an outlet is afforded by the Oneida River, which unites with the Seneca at Three River Point, 3 miles south of Phoenix, to form the Oswego River. On the east lies a level valley some 5 miles wide at the lake, but rising gradually and growing narrower as it approaches Rome. At Rome there is a low, short divide separating Wood ('reek from the Mohawk. It is the highest land on the route, excepting Sand Ridge, and has an elevation of 430 feet above the sea. From Rome to Little Falls the river winds in a sinuous course along a narrow alluvial plain, 1 See paper by G. K. Gilbert in Sixth Annual Report of the Commissioners of the New York State Reservation at N agnra. 444 DEEP WATERWAYS. and bears no resemblance to the rapidly descending and rocky channel below the falls. The ridges in Oswego County are composed of gravel, sand, and clay in varying proportions and degrees of hardness. Rock does not generally appear in the ridges, and it is easy to think of them as the moraines of ancient glaciers, deposited upon the surface of the under¬ lying rock. The comparatively recent formation of the topography is strikingly illustrated by the test borings made at Minetto, which show the bed rock to be higher in the river channel than beneath the hills on either side. The argillaceous sandstone of the Utica formation is found in the river bed at Oswego. Between Oswego and Fulton is found the gray, brown, or mottled Medina sandstone, and between Fulton and Oneida Lake are found the Clinton shales. The material to be excavated in Oneida Lake is mostly soft mud. There is also a little sand and gravel, and near the outlet at Brewerton a small amount of rock was discovered. East of Oneida Lake the rock encountered is a shale belonging to the Utica formation. A few miles west of Rome there is a deposit of rock and hardpan, which separates the Wood Creek and Mohawk valleys. West of this ridge there is a deep deposit of sand and clay resting upon the Utica shale. East of Rome and along the Mohawk Valley there is a deep channel between the hills, filled in with sand, clay, and beds of gravel. This channel is somewhat tortuous, and in places spurs of rock jutting into the valley from the adjacent hills are intersected by the located canal channel. The location and depth of proposed rock excavation can be readily seen on the profile of the line. The country adjacent to the route is fertile and productive, but the route itself lies generally in swamps and water courses, which are of less value than ordinary farming land. Building sand is abundant along the route. The sandstone at Oswego is unsuited for heavy masonry, but could be used for slope wall. East of Oswego, at Chaumont, and north of Rome, on the Black River Canal, there exists an abundance of good limestone that could be brought to the route by water transportation. Granite is quarried on Grindstone Island, in the St. Lawrence, and is quite abundant on the islands at the head of the river. The route appears to be well suited for the construction of a water¬ way of large dimensions. The vital question of water supply has been discussed elsewhere, and there remains but one feature concern¬ ing which any serious doubt need be expressed. Between Rome and Frankfort, in the Mohawk Valley, there is perhaps 10 miles of line along which much of the material to be excavated is loose sand and DEEP WATERWAYS. 445 clay, and the slopes of the channel would undoubtedly cause some trouble by caving into the canal. But such a condition means, at the worst, an increase of excavation that is small compared with the total, and its importance is not sufficient to weigh heavily against the route. The particular line along this route which has been selected as best suited for the construction of a canal may be briefly described as follows: The line leaves Lake Ontario about 1.1 miles west from the Oswego light-house, at a place known locally as Sheldons Point, and passes through the outskirts of the city along the westerly slope of a narrow valley. It crosses the Rome, Watertown and Ogdensburg Railroad tracks near that company’s repair shops. The Delaware, Lackawanna and Western Railroad is intersected by it in the southwestern part of Oswego city, near the divide between the Oswego River and Lake Ontario, and it then follows the Delaware, Lackawanna and Western Railroad to the village of Minetto, where it enters the Oswego River, about 5.7 miles from its mouth. From Minetto it follows the river about 4.9 miles, to the northern part of the village of Fulton. There it crosses the New York, Ontario and Western Railway and enters the valley of a small creek on the east side of the river. It passes up this creek, along the easterly side of Fulton, crosses the New York, Ontario and Western Railway again about four-fifths of a mile west of a flag station called Ingalls, and, continuing 34 miles farther, enters a swamp having an area of about 2f miles, and which is known as Peter Scott’s Swamp. L T p to this place the direction has been southeasterly. Here the course changes to easterly and crosses a glacial deposit called Sand Ridge, and the Oneida River, which skirts the ridge on the east. Then it crosses a second ridge and enters the Oneida River, the channel of which it follows about 1.7 miles, to the foot of Oneida Lake, at Brewerton. The line passes through Oneida Lake a little north of its axis and enters Wood Creek Valley at Sylvan Beach, a summer resort on the eastern shore of the lake. About 5.5 miles from the lake the line begins to rise on the south side of the valley in order to pass over the Rome summit. The Erie Canal is crossed about 3.4 miles west of Rome and again just east of that city. The New York Central and Hudson River Railroad and the New York, Ontario and Western Railway are crossed in the southern outskirts of the city; the New York Central and Hudson River Railroad is crossed a second time about 3 miles east of Rome, and the line passes down the valley, sometimes on one side and sometimes on the other. The Black River branch of the New York Central and Hudson River Railroad is crossed just west of Utica, and the main line is crossed again about 4.7 miles east of Utica. From there to Herkimer the valley is very narrow and is DEEP WATERWAYS. 44b already occupied by the New York Central and Hudson River Rail¬ road and West Shore railroads. There is room between them for the canal, but there is no choice of location, and the present channel of the Mohawk must give place to that of the proposed waterway. DETAILED DESCRIPTION AND ESTIMATES. Estimates have been made for two channels having depths of 21 and 30 feet, respectively, and two radically different plans have been considered for passing the summit at Rome. The high-level plan is to pass over the summit with a minimum amount of excavation, to use a short high level at Rome, and to bring the water supply from the Black River Valley through a feeder 92.75 miles long. The sur¬ vey and estimate for the feeder were not a part of the w'ork of this division. The low-level plan is to convert Oneida Lake into a great storage reservoir and excavate a channel through the Rome divide that would permit the waters of the lake to flow eastward into the Mohawk Valley. This description is written chiefly with reference to a 30-foot chan¬ nel on the high-level plan, but it applies generally to a 21-foot channel as well, and the portions of the route north of Fulton and east of Frankfort are common to both of the above-mentioned plans. The elevation of water surface of the river level from Minetto to Fulton is determined by the crest of the lower Fulton dam. There are two dams at Fulton, which supply water to the Oswego Canal and to numerous mills and factories. It is considered that if either dam is to be affected by the proposed waterway it will be more economical to wipe out one dam entirely than to either partly submerge one or to wholly submerge one and part of the other. As a general proposition, it is more economical to raise the water surface of a stream and buy the property submerged than to obtain the necessary depth of water by excavation. With the water surface raised to the crest of the lower Fulton dam the submerged territory will include some low ground west of Minetto, a small cemetery just south of Minetto, a small amount of land along the river, a portion of the Black Creek Valley, the Battle Island dam, and the lower part of Fulton, which is thinly settled, but includes several shops and factories. Black Creek is a small stream which enters the river from the east about halfway between Minetto and Fulton, and Battle Island dam is located in the river just below the mouth of Black Creek. At present it is only used to furnish water to the Oswego Canal, but the right to the surplus waters for power purposes is held by private parties. It is proposed to build a dam a little south of Minetto to raise the water to the desired elevation, and, by means of an embankment through the village and an excavated prism beyond, to extend the DEEP WATERWAYS. 447 same level northward across the summit between the river and the lake to the outskirts of Oswego, where the slope of the ground makes it necessary to lock down toward the lake. 1 If the water surface were fixed at a lower elevation, there would be a great increase of rock excavation in the river channel and also in the earth excavation north of Minetto, the Fulton dam would be only partially submerged, and other conditions would not be varied enough to materially lessen the expense. If it were fixed at a higher eleva¬ tion, the rock excavation in the river channel would not be materi¬ ally lessened; the earth excavation in the canal prism north of Minetto would be lessened, but the Minetto embankment would be increased in height, which is objectionable; the additional property flooded in Fulton would be more valuable in proportion than that which it is proposed to submerge, and the head available for power at the upper Fulton dam would be lessened. These conditions were deemed sufti- cient to determine the elevation without making any comparative estimates. The low-water elevation of Lake Ontario is 245.4, and the elevation proposed for the water surface of the river level is 331. The differ¬ ence—85.0 feet—has been divided between four locks, each having a lift of 21.4 feet. The first is located on the shore of the lake, so as to save excavation by rising as quickly as possible above the lake level. The next two locks are combined in one structure, and located near the Rome, Watertown and Ogdensburg Railroad roundhouse, where there is a sudden rise of ground, which makes such an arrangement economical. The fourth lock is located in the southwestern part of the city of Oswego. 2 The channel connecting the first lock with deep water in the lake is about 1,700 feet long and 600 feet wide. The material to be excavated is solid rock. Timber cribs have been planned for mooring vessels and guiding them into the lock, but the question of breakwaters to protect the entrance has not been considered in this report. That subject has been treated separately in Appendix No. 3. The design of locks Nos. 2 and 3 is such that the water may not always pass 1 See footnote following for change of this proposed plan. 2 After having been further considered by the Board, the location of the fourth lock has been changed since this report was completed. The new location is in the village of Minetto, a short distance below the proposed dam at this place. By the first location of the lock the canal would be carried through the village for a distance of about a mile between earth embankmentshaving an average height of about 20 feet and a maximum height of 30 feet. The interests involved by the waterway itself and the village below the embankment, should a break occur in the same, have been deemed sufficient to warrant the change. With this location the entire canal section is below the surface of the ground. The adopted estimates have been changed accordingly, the increase in cost over the first location being, for the 30-foot channel, $1,190,413. and for the 21-foot channel $.">93,471. See plate 18 for this relocation. 448 DEEP WATERWAYS. through to the level below, as required, so sluice gates and a by-pass have been provided to feed water around the structure. The dam at Minetto is 42 feet high above the river bed and 50 feet above the rock, and has a spillway 750 feet long. It is built of con¬ crete and founded on solid rock, but the wings are anchored in banks of compact gravel and clay. While the construction of the wings would be less difficult and expensive if they could be built in rock, it is be¬ lieved that they can be made perfectly secure in the existing material. No guard lock has been planned where the canal leaves the river, because the fluctuation of flow in the Oswego River is not great, and can be provided for at a less cost by raising the lock walls at the lower end of the level and the canal banks through the village of Minetto. The usual flood in the Oswego River is about 25,000 cubic feet per second; the maximum flood recorded is about 42,000 cubic feet per second. It is proposed at times to turn 10,000 cubic feet per second westward from the Mohawk into Oneida Lake. With a spillway 750 feet long, a flow of 52,000 cubic feet per second will produce a depth on its crest of about 8 feet. The lock walls at the lower end of the level and the slope walls along the level are planned 10 feet higher than the crest of the dam. 1 But a flood of 10,000 cubic feet per second from the Mohawk would spread out over Oneida Lake and probably be exhausted before it could create a flow of 10,000 cubic feet per second in the Oneida and Oswego rivers. Moreover, to cause the maximum flood on the Oswego would require that a maximum flood should occur on the Seneca, the Oneida, and the Upper Mohawk simultaneously. Maximum floods usually occur when a warm rain falls on a layer of snow. It seldom happens that a heavy rain occurs on these three streams at once, and it is still more seldom that they are all covered with snow. If Oneida Lake is made a storage reservoir, the condition of a full reservoir would be added to those necessary for a maximum flood, and the probability of its occurrence would be still further lessened. In the river level, where it is necessary to increase the depth by excavation, the submerged shoulders of the excavated prism would be invisible to navigators, and the river currents would make boats more difficult to guide than when in a regular canal prism. To lessen these difficulties the bottom width of the channel in the river has been made 400 feet instead of 203 feet or 250 feet, as shown in the regular canal sections. The fifth lock is situated on Waterhouse Creek, in the northeastern part of Fulton, and lias a lift of 224 feet. The volume of excavation 1 The results of the Cornell experiments, which have become available since Mr. Himes wrote his report, show that a coefficient of 4 may be used for ogee dams. Recomputing the height of high water over the Minetto dam, we obtain 6.7 instead of 8. This allows the height of lock walls and slope walls to be reduced 1 foot and also decreases the amount of territory estimated as flooded. (See also foot¬ note on the preceding page.) DEEP WATERWAYS. 449 could be lessened and an equally good foundation secured by placing the lock nearer the river. The reason this was not done is because the canal prism above the lock would be excavated through beds of gravel so coarse and porous that it is doubtful whether it would hold water. A more thorough examination of the material might justify the location of the lock nearer the river, but it is safer with present information to base the estimate on the location selected. The valley above lock No. 5 rises quite rapidly, and another lock, with a 224-foot lift, is planned 1.193 miles farther along the line. In locating the locks no attempt lias been made to economize exca¬ vation by placing them so as to require earthen embankments to form the sides of the prism just above, it being considered that f he interests which would justify the construction of so great a waterway should not be jeopardized by the use of earthen darns to retain 21 feet or 30 feet of water. Throughout the entire plan the water surface has been kept as near as practicable to the natural earth surface, and where embankments can not be avoided there is always a great surplus of excavation which may be used to give them an excess of strength. Lock No. 6 raises the canal to the proposed low-water elevation of Oneida Lake, which is 376. The lake level is 40.833 miles long, and reaches from near Fulton across Oneida Lake and up the Wood Creek Valley 5.5 miles beyond Sylvan Beach. The line crosses the New York, Ontario, and Western Railway 1.4 miles beyond lock No. 6, near the summit of the divide between Waterhouse Creek and Peter Scott’s swamp. The creek drains into the Oswego River, and the swamp drains into the Oneida River. The surface of the swamp is about 12 feet lower than the canal water surface, and this is one of the few places where an embankment has been planned. The canal passes near the center of the swamp, where the earth is very soft to a depth of 70 feet, and the construction of an embankment upon it to hold 12 feet of water would be rather hazardous; but along the banks of the Oneida River, on the southern margin of the swamp, the earth is much firmer and the distance between high ground on either side is much shorter than along the line of the canal. It has therefore been planned to submerge* the entire swamp by building an embankment along the river. Two small creeks run through the swamp into the river and should be cut off by dams before building the embankment. The embankment is to be 10 feet higher than low-water surface in the lake and well lined with riprap. There is an abundance of clay and gravel with which to build it, and the great volume of material to be excavated in Sand Ridge on the east side of the swamp may be used to give it a cross section largely in excess of any possible requirement. The deepest cut on the line is encountered in Sand Ridge, the maxi¬ mum being about 84 feet. H. Doc. 149-29 450 DEEP WATERWAYS. At the crossing of the Oneida River on the east side of the ridge a weir has been planned to raise the water surface of the river and lake perma¬ nently to an elevation of 376, which is the highest known stage of the lake. It is proposed to build a timber structure, 1 with gates to be used when necessary, to maintain the low-water flow of the stream. The foundation, as shown by the nearest borings, is compact, gravelly earth. To provide for evaporation and to maintain the present supply for power purposes at Phoenix and Fulton during the summer months, the crest of the weir is raised to 378, thus making a storage depth of 2 feet. The excess of evaporation over precipitation during naviga¬ tion season will not exceed 16 inches, 2 and the balance, 8 inches, may be used for power purposes. The data to determine just how much water is needed to maintain the low-water flow of the stream are not available, but, when it is determined, the storage depth could be varied sufficiently to furnish the required amount without materially affect¬ ing the present estimate. The crest of the weir is to be 800 feet long and will discharge a flow of 22,000 cubic feet per second with a depth of 4.1 feet on its crest. The maximum flood in the Oswego River is about 42,000 cubic feet per second. The area of the watershed is 5,002 square miles, 10.6 per cent of which lies north of Three River Point. If we assume that the maximum flood at Three River Point is 42,000 cubic feet per sec¬ ond, and that it is divided between the Oneida and Seneca rivers in proportion to the areas of their respective watersheds, the flow from the Oneida is 12,000 cubic feet per second, and that from the Seneca is 30,000 cubic feet per second. But in the Oneida basin the pondage area is 15.55 per cent of the total and in the Seneca basin the pondage area is 8.9 per cent of the total, so it is probable that the discharge of the Seneca would be the greater in proportion and that the flow assumed for the Oneida is excessive. It has already been stated that in time of maximum flood it is intended to turn 10,000 cubic feet per second westward from the Mohawk, and the sum of the two volumes, 22,000 cubic feet per sec¬ ond, is the amount for which the weir is planned. In the case 3 of ordinary floods there would be no flow from the Mohawk, and the Oswego flood of 25,000 cubic feet per second pro¬ portioned between the Seneca and Oneida rivers, as before, would produce a depth on the weir of 1.92 feet. 4 1 In the finals this clam was estimated to be built of concrete. 2 See low-ievel plan, water supply. 3 See Report on Water Supply, Appendix No. 16. 4 Since Mr. Himes’s report was completed, the control of the Oneida River has been further considered by the Board, and the adopted estimates changed, as out¬ lined below. By Mr. Himes's plan, flash boards or some form of movable dam would be required to raise the crest of the weir 2 feet at low water. It was thought that flash boards 2 feet high would hardly be in accordance with a work of this character. Since a movable dam must be built, a dam with 2-foot head DEEP WATERWAYS. 451 By this arrangement, if it should happen that an extreme flood occurred simultaneously in the Mohawk, Oneida, and Seneca rivers, the extra volume of 10,000 cubic feet per second from the Mohawk would make a higher flood than has yet occurred at Three River Point. Such a contingency is not likely to occur, but if it did, little damage would be done beyond the flooding of an increased area of farming land. Even this damage could be avoided by building a bear trap in the present Phoenix dam; but it is thought unnecessary and has not been included in the estimate. A flood in the Oswego River below Phoenix, 10,000 cubic feet per second in excess of the maximum flood recorded, would probably do a little damage at Fulton. Below that, if the canal should follow the river to the lake, as described later, the increased section of waterway would carry the flood without difficulty. If the canal should leave the river at Minetto, as above described, an excessive flood would do some damage at Minetto and Oswego; but such a flood would occur so rarely, if at all, and the damage would be so small, that it would be cheaper to pay the damage than to provide works to avoid it. The lowest stage of the water in Oneida Lake is probably 368.5, and the highest 376. No very valuable property would be injured by rais¬ ing the water surface to 378, except at Sylvan Beach. There the ground is so low that practically the whole place would be destined. The tracks of the New York, Ontario and Western and the Lehigh Valley railroads, which cross the valley near Sylvan Beach would need to be raised a few feet and the embankments heavily riprapped. The line adopted through the lake is that of least excavation. It has but two curves, one a little west of the center of the lake and the other at the entrance to the canal at the eastern end of the lake. A straight channel could be made through the lake at a somewhat greater expense for excavation; and in the case of a 21-foot channel the increased excavation would be so slight that the straight channel might be preferable to the one adopted. The lake is less than 30 feet deep for a distance of 6.8 miles, meas¬ ured from its western end. At the eastern end there is only one-half mile of water less than 30 feet deep. Where it is necessary to dredge a channel the proposed bottom width is 600 feet, so that open-water navigation will practically extend the whole length of the lake. Here and in Peter Scott’s swamp the material is so soft that the excavation has been estimated on slopes of 1 on 3 instead of 1 on 2, as shown on the standard cross section. would not cost materially less per linear foot than one with 5 feet of head. It is therefore proposed to reduce the length of weir to 300 feet and the elevation of crest of the fixed portion to 373. The movable portion of the dam is designed to raise the elevation of crest to 378. The sluice gates mentioned by Mr. Himes are designed for a high-water discharge of 10,000 cubic second-feet. This leaves 12,000 as the maximum discharge over the weir. Using 3.5 as the coefficient of discharge, the depth of flow on the fixed weir at high water would be 5.1, or the high-water elevation would be 378.1. 452 DEEP WATERWAYS. Fish Creek, which unites with Wood Creek just before it enters Oneida Lake, is a much larger stream than the latter and carries large volumes of silt into Oneida Lake. It is therefore impracticable to receive it into the canal, and an estimate has been made for a diver¬ sion channel to turn its waters into Oneida Lake, north of the canal entrance. Two timber crib piers are provided at Sylvan Beach to prevent the loose sand on the shore from being washed into the channel. At the west end of the lake the material is more compact and no such protection is required. The next lock, which is the seventh from the beginning, is located at the first place east of the lake, where the ground is high and firm enough to build at a higher level and where a suitable lock founda¬ tion exists. The lock is a little south of Wood Creek and has a lift of 20 feet. The eighth lock, which is the last of the ascending series, has also a lift of 20 feet, and is located 4.356 miles beyond Lock No. 7 and about 4 miles west from Rome. The water surface of the summit level has an elevation of 416 feet above the sea and 170.6 feet above low water in Lake Ontario. The subject of floods on the Mohawk has been investigated by Mr. Rafter, and it is understood that a flow of 35,000 cubic feet per sec¬ ond at Little Falls can be taken safely down the river. It is to pro¬ vide for the contingency of a greater flood that plans have been made to take 10,000 cubic feet per second westward from the summit level into Oneida Lake. To do this by-passes have been designed to con¬ vey the water around Locks Nos. 7 and 8, and the discharge through the passes would be controlled by gates, so that no water at all would go that way except when desired. West of Rome the line crosses a wide plain through which it fol¬ lows the line of greatest depression of the underlying rock. At Rome, where there is no rock, it passes through the southern outskirts of the city along the lowest ground between the Wood Creek and Mohawk basins. From Rome eastward to the end of the division the Mohawk is too small to be considered in the location of the canal. The river is so crooked that it is frequently crossed by the canal, but the section of the latter is so much the greater, and near Rome its water surface is so much lower, that it would carry all the drainage and the river channel would be of no further use. About a mile east of Rome, on the south side of the valley, the canal passes through a rock cut in which it is proposed to receive the water of the Mohawk. The channel is enlarged to lessen the disturb¬ ance that may be caused by the currents, and a basin and weir are planned to intercept the gravel and silt that may be borne down by the stream. DEEP WATERWAYS. 453 At Oriskany there is planned a weir and lock, having a lift of 20 feet. This lock is the first in the descending series east of Rome. Oriskany Creek is to be received into the pool above the weir, and will in no way affect the canal. A similar weir and dam are planned at Frankfort, the lift being 20 feet as at Oriskany. The locations of both these locks are determined by the presence of rock that may serve as foundations. The Frankfort lock is the tenth and last on the western division. The elevation of water surface below the lock is 376, the same as on Oneida Lake. Some additional work is required to collect the waters of the small streams between Rome and Frankfort and convey them to places where they can be received into the canal without injury to its slopes. This work has been included in the estimates. In making the excavations between Rome and Frankfort, it will be necessary in those places where the material is very soft to carry it a considerable distance from the channel in order to prevent the super¬ imposed weight from forcing in the banks of the canal. With the exception of the Oneida River dam, all of the structures have foundations of solid rock. No borings have been made since the various structures have been located, save in the case of the upper lock at Fulton, and consequently it may be expected that a detailed examination of the lock sites might show an occasional slight change of location to be desirable, but the borings that were made are suffi¬ cient to show the presence of rock, so that the locations could be selected within narrow limits. The estimates have been made according to standard plans and prices, which are fully described in the report of your Board, and need not be discussed here. Table No. 1 gives the lengths of levels, lifts, and costs of the locks; Table No. 2 gives the location, kind* and cost of the bridges; Table No. 3 shows the alignment; Table No. 4 gives the different classes of navigation with the percentages of each, and Tables Nos. 5 and 6 give the total estimated costs of 21 and 30 foot channels, respectively. These six tables all pertain to the high-level plan. Table No. 1 .—Locks for high-level plan. Xo. of lock. Location of lock. Single or double Elevation of low water. Lift, in feet. Length of level, lock. Below. Above. in miles. 1 _ Oswego, X. Y. Single .... Double... Single .... . do_ 245.4 200.8 309.6 331.0 353.5 370.0 390.0 390.0 376.0 266.8 309.6 331.0 353.5 370.0 390.0 416.0 410.0 396.0 21.4 42.8 21.4 22.5 22.5 20.0 20.0 20.0 20.0 2 and 3. 4_ _do .. .. 0.890 4.384 5.900 1.193 40.833 4.350 13. 048 15.000 4.529 Minetto, N. Y... Pulton, X. Y. 6. 7... .... do ... .do _ 6 miles east of Sylvan Beach. .do_ 8 . 4 miles west of Rome.. . .do_ 9 Oriskany. N. Y. Frankfort, N. Y.... .do_ .. .do.. 10 Head gates, lock 10, to end of division. Total.. 210.5 90.733 454 DEEP WATERWAYS Table No. 1. —Locks for high-level plain — Continued. COST. No. of lock. 30-foot channel. 21-foot channel. Operating machin¬ ery, 30- foot and 21-foot channels. 1 . $1,054,942 3,456,980 1,109,991 1,076, 693 1,072, 443 1,031.659 1,031,659 1,042,387 1,042,387 $641,108 2,179.291 681,426 662,834 662,808 634, 733 626,933 634,481 634,481 $100,000 175,000 100,000 100,000 100,000 100,000 100,000 100,000 100,000 2 and 3_._ 4. t). 8 ... 9 . lo::::::::::::;;::;;;:::;::::::::::;:::::::::::;;:::::::::::.::: Operating machinery-__-_ 11,919,141 975,000 7,358,095 975,000 975,000 Total.... ... 12,894,141 8,333,095 Note.—I n giving the lifts of locks no account is made of the fluctuations of water surfaces due to natural causes. Table No. 2. —Bridges for high-level plan. Location. Oswego, N. Y . Dc. Near Oswego, N.Y Minetto, N.Y. Fulton, N.Y. Do. Caughnedoy.N.Y 1 Brewerton, N.Y. Sylvan Beach,N.Y Rome, N.Y. Do. Utica, N. Y. Do. Near Utica, N. Y.. Do. Frankfort, N. Y .. Ilion, N. Y .. Herkimer, N, Y... Total . Sta¬ tion. Kind of bridge. Num¬ ber of tracks. Fixed or swing. Num¬ ber of spans. 15 Highway Swing.... 1 87 Railway . 2 .do_ 1 no Higliwav .do_ 1 275 _do_ .do_ 1 554 Railway . 1 .do_ 1 582 Highway _do_ 1 _do Fixed.... 9 1403 Railway . 1 Swing.... 1 2550 _do_ 1 .do- 1 3292 Highway .do_ l 3304 Railway . 1 .do_ 1 3995 .do_ 1 .do_ l 4038 Highway .do_ l 4236 Railway - 2 .do_ 1 4242 .do — 2 .do_ 1 4555 Highway _do_ l 4648 _~do_ .do_ l 4747 .do_ .do_ 1 30-foot channel. 21-foot channel. Total length. Esti¬ mated cost. Total length. Esti¬ mated cost. 545.0 $68,24(3 525.0 $66, 796 550.0 221,429 530.0 196,161 545.0 116.534 525.0 98,138 235.0 19,986 195.0 16.6,50 537.5 152,916 517.5 132.174 235.0 19,986 195.0 16,650 400.0 25,090 400.0 25.090 537.5 141.362 517.5 134,14(3 537.5 141,184 517.5 125,796 545.0 95,456 525.0 80,820 537. 5 142,548 517.5 117,236 537.5 137,444 517.5 117.236 545.0 95,456 525. 0 80,820 550.0 221,429 530.0 196,161 550.0 221,429 530.0 196,161 235.0 19,986 195.0 16,650 545.0 95,456 525.0 80,820 545.0 95,456 525. 0 80,820 2,031,387 1,782,500 1 Not over channel. Table No. 3. — High-level plan. Length of tangent. Length of curve. Minimum radius of curvature. Degree of curvature. Percent¬ age of tangent. Percent¬ age of- curvature. Miles. 70,990 Miles. 19,743 Feet. 4,523.4 O / 687 31 78.24 21.76 Table No. 4. — High-level plan. Miles. Per cent. Open water... 20.928 23.06 Improved waterways. 4.962 5.47 Canal prism.. 64.843 71.47 Total. 90.733 100.00 DEEP WATERWAYS. 455 Table No. 5.— Estimate — Oswego-Mohawk route, western division. HIGH-LEVEL PLAN—30-POOT CHANNEL. Lake Ontario section (station —26 + 00 to —9 +20): Excavation— Submerged rock, 349,670 cubic yards, at $2... $699, 340 Breakwater (Appendix No. 3) ....... 1,190,317 Crib work— Pine, 2,415,600 feet B. M., at $30 per M_._.. 72,468 Hemlock, 5,840.620 feet B. M., at$23 per M. 134,334 Oak, 97,920 feet B. M., at $50 per M.. 4,896 Iron, 742,870 pounds, at 3 cents.... 22,286 Stone fill, 102,034 cubic yards, at 60 cents.. 61,220 Total......... 2,184,861 Oswego-Minetto section (station —9 + 20 to 288 + 00): Excavation— Earth, dry, 11,866,321 cubic yards, at 20 cents. 2,373,264 Rock, dry, 1,969,596 cubic yards, at 65 cents. . 1,280, 237 Retaining wall, 143,479 cubic yards, at $4. 573,916 Masonry in by-passes. 4.300 cubic yards, at $4. 17,200 Slope wall, 68,448 square yards, at $1.10 . 75.293 Back fill, 449,336 cubic yards, at 25 cents. 112,334 Crib work— Pine, 2,374,400 feet B. M., at $30 per M . 71,232 Hemlock, 6,992,000 feet B. M., at $23 per M. 160,816 Oak. 96,920 feet B. M..at $25 per M.... 2,423 Iron, 786,632 pounds, at 3 cents .... 23, 599 Stone fill, 125,856 cubic yards, at 60 cents. 75,514 Railroad and highway changes... 29,660 Locks— No. 1. 1,054,942 Nos. 2 and 3. 3,456,980 No. 4. 1,109,991 Lock-operating machinery— 2 sets single, at $100,000. 200,000 1 set double, at $175,000. 175,000 Bridges... 426,189 Right of way— City property, 277 acres, at $2,080. 576,160 Farm land, 729 acres, at $200... 145,800 Total.. ... 11,940,550 Minetto-Fulton section (stations 288+00 to 550+00): Excavation— Earth, 1,911,170 cubic yards, at 18 cents. 344,011 Rock, dry, 117,268 cubic yards, at 65 cents. 76,224 Slope wall, 14,508 square yards, at $1.10. 15,959 Minetto dam (No. 1) — Excavation, 25,567 cubic yards, at 20 cents. 5,113 Masonry in dam and abutments, 44,913 cubic yards, at $6. 269,478 Cofferdams, estimated cost..... 55,000 Railroad and highway changes.. 25,670 456 DEEP WATERWAYS. Minetto-Fulton section—Continued. Right of way: Farm land, 1,154 acres, at $100-.. $115,400 Swamp land, 408 acres, at $12.50.... 5,100 Water rights, Battle Island dam... 150,000 Total.......—... 1,061,955 Fulton-Brewerton section (stations 550 + 00 to 1410 + 00): Excavation — Earth, dry, 29,190,187 cubic yards, at 18 cents...... 5,254,234 Rock, dry, 2,047,113 cubic yards, at 65 cents_ 1,330,623 Dike in Peter Scott’s swamp: Excavation, 383,000 cubic yards, at 18 cents .. . 68,940 Embankment, 588,000 cubic yards, at 25 cents_ 147,000 Riprap, 71,000 cubic yards, at 90 cents__ 63,900 Retaining wall, 129,353 cubic yards, at $4... .. 517,412 Slope wall, 201,946 square yards, at $1.10... _...- 222,141 Back fill, 548,894 cubic yards, at 25 cents._ _ ___ 137,224 Crib work— Pine, 1,903,200 feet B. M.. at $30 per M... 57,096 Hemlock, 7,423,680 feet B. M., at $23 per M. 170,745 Oak, 52,560 feet B. M., at $50 per M..... 2,628 Iron, 788,404 pounds, at 3 cents..... 23,652 Stone. 125,172 cubic yards, at 60 cents..... 75,103 Oneida River dam (No. 2) — Excavation, 14,000 cubic yards, at 18 cents __ 2, 520 Embankment. 43,700 cubic yards, at 15 cents .. 6, 555 Masonry in dam and abutments, 6,018 cubic yards, at $6. 36,108 Two Stoney gates to discharge 10,000 cubic feet per second_ 46,250 Movable dam, 5 feet high, 300 feet long, at $9.40 per linear foot. 2,820 Operating machinery_ _ _ .... _ 1,000 Railroad and highway changes..... 73,400 Steam ferries, 4, at $20,000_____ 80,000 Locks— No. 5.......... 1,076.693 No. 6......... 1,072,443 Lock-operating machinery, 2 sets single, at $100,000 ... 200.000 Bridges. .... 339, 354 Right of way— Village property, Fulton, N. Y., 55 acres, at $2,600. 143. 000 Farm land, 5,910 acres, at $100... 591,000 Swamp land. 2,446 acres, at $12.50____ 30,575 Water rights, Fulton. N. Y... 806.000 Total.. . 12,578,416 Brewerton to lock No. 7 section (station 1410 + 00 to 2782 + 00): Excavation— Earth, 15,736,440 cubic yards, at 8 cents.. 1,258,915 Rock, wet, 719,100 cubic yards, at $2___ __ 1.438,200 Fish Creek diversion, earth, 308,000 cubic yards, at 8 cents_ 24. 640 Slope wall. 94,746 square yards, at $1.10 .. 104,221 DEEP WATERWAYS. 457 Brewerton to lock No. 7 section, etc.—Continued. Crib work— Pine, 1,560,000 feet B. M., at $30 per M. . $46,800 Hemlock, 5,769,100 feet B. M.,at$23 per M. 132,689 Oak, 43,560 feet B. M., at $50 per M. 2,178 Iron, 616,342 pounds, at 3 cents .. 18,490 Stone fill, 97,566 cubic yards, at 60 cents . .. 58,540 Railroad and highway changes .. .„. . 112,638 Bridges. .. . 141,184 Right of way— Farm land, 10,541 acres, at $100...... 1,054. 100 Swamp land, 4,800 acres, at $12.50 . 60,000 Village property, Sylvan Beach. . . 109,000 Total ..... 4,561,595 Lock No. 7, Fort Bull section (station 2782+00 to 3164+00): Excavation— Earth, 15,702,581 cubic yards, at 20 cents ... . 3,140,516 Rock, dry. 257.522 cubic yards, at 70 cents .. 180,265 Retaining wall, 47,330 cubic yards, at $4 ... 189,320 Masonry in by-pass, 10,000 cubic yards, at $4... 40.000 Slope wall, 100,467 square yards, at $1.10. 110,514 Back fill, 425,084 cubic yards, at 25 cents... 106,271 Crib work— Pine. 2,854.800 feet B. M., at $30 per M. 85,644 Hemlock. 10,897,220 feet B. M., at $23 per M ... 250, 636 Oak. 78.840 feet B. M., at $50 per M.... 3,942 Iron. 1.160,606 pounds, at 3 cents .. 34,818 Stone fill, 184,098 cubic yards, at 60 cents. 110, 459 Highway changes ...... 680 Steam ferry, 1...... 20.000 Locks— No. 7....... 1,031,659 No. 8.. .. 1,031.659 Lock-operating machinery. 2 sets single, at $t00,000.. 200, 000 Right of way— Farm land, 943 acres, at $100..... 94,300 Total .. . 6,630,683 Fort Bull—Herkimer section (station 3164 + 00 to 4789 + 91.5): Excavation— Earth, 61,434,573 cubic yards, at 18 cents.... 11,058,223 Rock, dry, 494,380 cubic yards, at 70 cents. 346.066 Retaining wall, 73,685 cubic yards, at $4... 294,740 Masonry in by-passes, receivers, etc., 27,400 cubic yards, at $4_ 109,600 Slope wall, 841,185 square yards, at $1.10..... 925,303 Back fill. 378,858 cubic yards, at 25 cents .... 94,715 Crib work— Pine, 2.277,600 feet B. M.. at $30 per M.... 68,328 Hemlock, 8,832,200 feet B. M., at $23 per M.. 203,141 Oak, 62,880 feet B. M., at $50 per M. 3,144 Iron, 964,024 pounds, at 3 cents. 28,921 Stone fill, 143,297 cubic yards, at 60 cents... 85,978 458 DEEP WATERWAYS. Fort Bull—Herkimer section, etc.—Continued. Dams— Oriskany, N. Y. (No. 3)— Excavation, 787,000 cubic yards, at 18 cents. $141. 660 Masonry in dams and abutments, 21,835 cubic yards, at $6. 131. 010 Timber in grillage foundation, 1,101,372 feet B. M., at $22 perM.... 24,230 Piles in foundation, 155,100 linear feet, at 20 cents. 31,020 Sheet piling. 128,000 feet B. M., at $33 per M. 4,224 Iron. 80,000 pounds, at 3 cents ...._ 2,400 Stoney gate to discharge 500 cubic feet per second. 2, 000 Frankfort, N. Y. (No. 4)— Excavation, 639,000 cubic yards, at 18 cents. .115,020 Masonry in dam and abutments, 21,835 cubic yards, at $6. 131,010 Timber in grillage foundation, 1.101,372 feet B. M., at $22 perM..._ ______ 24,230 Piles in foundation, 155,100 linear feet, at 20 cents. 31,020 Sheet piling, 128.000 feet B, M., at $33 per M. 4,224 Iron, 80,000 pounds, at 3 cents.... 2,400 Railroad and highway changes. 450,170 Steam ferry, 1. 20,000 Locks— No. 9 . 1,042.387 No. 10. 1,042,387 Lock-operating machinery, 2 sets single, at $100,000 .... 200,000 Bridges.. 1,124,660 Right of way— Farm land, 10,335 acres, at $100.. 1, 033, 500 Total. 18,775,711 TOTAL COST. Lake Ontario'section.—...$2,184,861 Oswego-Minetto section....... 11,940,550 Minetto-Fulton section.. 1,061,955 Fulton-Brewerton section...... 12.578,416 Brewerton-Lock No. 7 section. .,..... 4,561,595 Lock No. 7-Fort Bull section.... 6,630, 683 Fort Bull-Herkimer section....... 18,775,711 Water Supply, Appendix No. 16. 18,080,752 Total. 75,814,523 Table No. 6.— Estimate — Oswego-Mohawk route, western division. HIGH LEVEL PLAN-21-FOOT CHANNEL. Lake Ontario section (station — 20 + 09 to —9 + 20): Excavation— Submerged rock, 152.361 cubic yards, at $2... .. $304,722 Breakwater, Appendix No. 3. . 721,380 Crib work— Pine, 2.416,000 feet B. M., at $30 per M. 72,480 Hemlock. 3,096.800 feet B. M., at $23 per M. _ 71,226 Oak, 83.600 feet B. M.. at $50 per M ... 4,180 Iron, 490,585 pounds, at 3 cents... 14,718 Stone fill, 63,325 cubic yards, at 60 cents... . 37,995 Total..... 1,226,701 DEEP WATEKWAYS. 459 Oswego-Minetto section (station — 9 4 - 20 to 288 4 - 00): Excavation— Earth, dry, 9,300,777 cubic yards, at 20 cents.$1,860,155 Rock, dry, 843,884 cubic yards, at 65 cents. 548,525 Retaining wall, 97,906 cubic yards, at $4... 391,624 Masonry in by-passes, 4,300 cubic yards, at $4... 17,200 Slope wall, 88,288 square yards, at $1.10. 97,117 Back fill, 348,900 cubic yards, at 25 cents. 87,225 Crib work— Pine. 2,374,400 feet B. M., at $30 per M. 71,232 Hemlock, 4.893,400 feet B.M.. at $23 per M. 112,571 Oak. 53,760 feet B. M., at $50 per M .. 2,688 Iron, 593,600 pounds, at 3 cents... 17,808 Stone fill, 94,576 cubic yards, at 60 cents. 56,746 Railroad and highway changes. 29,660 Locks— No. 1. 641,108 Nos. 2 and 3. 2,179,291 No. 4. 681,426 Lock-operating machinery— 2 sets single, at $100,000... 200,000 1 set double, at $175,000. 175,000 Bridges.. 377,745 Right of way— City property, 277 acres, at $2,080. 576,160 Farm land, 729 acres, at $200... 145,800 Total. 8.269,081 Minetto-Fulton section (station 288—00 to 550—00): Excavation— Earth, 914.740 cubic yards, at 18 cents .. 164,653 Slope wall, 13,414 square yards, at $1.10. 14,755 Minetto Dam (No. 1) — Excavation, 25,567 cubic yards, at 20 cents. 5,113 Masonry in dam and abutments, 44,913 cubic yards, at $6. 269,478 Cofferdams.... 55,000 Railroad and highway changes ....... 25,670 Right of way— Water rights, Battle Island Dam. 150,000 Farm land, 1,154 acres, at $100....... 115,400 Swamp land, 408 acres, at $12.50. 5,100 Total. . ... 805,169 Fulton-Brewerton section (station 550—00 to 1410—00): Excavation— Earth, 21,192,092 cubic yards, at 18 cents. 3,814,576 Rock, dry, 941,910 cubic yards, at 65 cents.. 612,241 Peter Scott’s Swamp Dike— Excavation, 383,000 cubic yards, at 18 cents .... 68,940 Embankment, 588,000 cubic yards, at 25 cents. 147,000 Riprap, 71,000 cubic yards, at 90 cents. 63,900 Retaining wall, 71,756 cubic yards, at $4 . 287,024 Slope wall, 196,551 square yards, at $1.10. . 216,206 Back fill, 356,371 cubic yards, at 25 cents .... 89,013 DEEP WATERWAYS. 460 Fultou-Brewerton section, etc.—Continued. Crib work— Pine, 2,449,200 feet B. M.. at $30 per M. $73,476 Hemlock. 6,836,300 feet B. M.,at $23 per M.. 157,235 Oak, 70,200 feet B. M.,at $50 per M. 3,510 Iron, 776,277 pounds, at 3 cents.... 23,288 Stone fill, 120.978 cubic yards, at 60 cents.... 72, 587 Oneida River Dam (No. 2) — Excavation, 14,000 cubic yards, at 18 cents... 2, 520 Embankment, 43,700 cubic yards, at 15 cents.. 6, 555 Masonry in dam and abutments, 6,018 cubic yards, at $6. 36,108 Two gates (Stoney to discharge 10,000 cubic feet per second).. 46,250 Movable dam 5 feet high, 300 feet long, at $9.40 per linear foot. 2,820 Operating machinery .. ...__ . 1,000 Railroad and highway changes. ...... 73,400 Steam ferries, 4, at $20,000 .. 80,000 Locks— No. 5...-.... 662,834 No. 6.... .. 662,808 Lock-operating machinery— Two sets single..... 200, 000 Bridges.-. -..... 308.030 Right of way— Village property, Fulton. N. Y., 55 acres, at $2,600. 143,000 Farm land, 5,910 acres, at $100_..;. 591,000 Swamp land, 2,446 acres, at $12.50____ 30,575 Water rights, Fulton. N. Y .... 806.000 Total....... 9,282,006 Brewerton—Lock No. 7 section (station 1410+00 to 2782+00): Excavation— Earth, 8,089,246 cubic yards, at 8 cents___ 647,140 Rock, wet, 171,046 cubic yards, at $2.. 342, 092 Fish Creek diversion— Excavation, earth, 308,000 cubic yards, at 8 cents.. 24,640 Slope wall, 108,540 square yards, at $1.10.__ 119,394 Crib work— Pine, 1,210,000 feet B. M., at $30 per M..... 36,300 Hemlock, 3,483,760 feet B. M.. at $23 per M___ ___ 80.126 Oak, 31,680 feet B. M., at $50 per M.-.-. 1.584 Iron, 366,080 pounds, at 3 cents. ... 10,982 Stone fill, 56,760 cubic yards, at 60 cents.. 34,056 .* . • Railroad and highway changes... 112,638 Bridges __ ... . 125, 796 Right of way— Farmland, 10,541 acres, at $100...... 1,054,100 Village property, Sylvan Beach. 109,000 Swamp land, 4,800 acres, at $12.50 ..... 60.000 Total... 2,757,848 Lock No. 7, Fort Bull section ( station 2782+00 to 3164+00): Excavation— Earth, 12,145,959 cubic yards, at 20 cents . 2.429,192 Rock, dry, 33,420 cubic yards, at 70 cents.... 23,394 DEEP WATERWAYS. 4G1 Lock No. 7, Fort Bull section, etc.—Continued. Masonry in by-pass, 10,000 cubic yards, at $4 ... $40,000 Slope wall, 101,784 square yards, at $1.10... 177,962 Back fill, 281,840 cubic yards, at 25 cents .. 70,400 Crib work— Pine, 3,806,400 feet B. M., at $30 per M.. 114,192 Hemlock, 10,191,600 feet B. M., at $23 per M.. 234,407 Oak, 105,120 feet B. M., at $50 per M. . .. 5,256 Iron, 1,165,448 pounds, at 3 cents . ... 34,963 Stone fill, 181,536 cubic yards, at 60 cents.. 108,922 Highway c uanges... . 680 Steam ferry, 1 ... 20,000 Locks— No. 7....... 634. 733 No. 8......... 626.933 Lock-operating machinery— Two sets single, at $100,000 ... 200,000 Right of way— Farm land, 943 acres, at $100..... 94,300 Total....... 4,815,394 Fort Bull-Herkimer section (station 3164 + 00 to 4789 + 91.5): Excavation— Earth, 45,963,879 cubic yards, at 18 cents.. 8,273,498 Rock. dry. 85,828 cubic yards, at 70 cents.. 60,080 Masonry in receivers and by-pass, 27,400 cubic yards, at $4_ 109, 600 Slope wall, 790,867 square yards, at $1.10 ........ 869,954 Back fill, 247,120 cubic yards, at 25 cents... 61,780 Crib work— Pine, 3.806,400 feet B. M., at $30 per M... 114,192 Hemlock, 10,677,600 feet B. M., at $23 per M_. 245,585 Oak, 105,120 feet B. M., at $50 per M_. 5,256 Iron. 1,244,488 pounds, at 3 cents. 37,335 Stone, 180,016 cubic yards, at 60 cents.-.... 108,010 Dams— Oriskany, N. Y. (No. 3): Excavation, 787,000 cubic yards, at 18 cents. 141,660 Masonry in dam and abutments, 21,835 cubic yards, at $6. 131,010 Timber in grillage foundation. 1,101,372 feet B. M., at $22 per M... .. 24,230 Piles in foundation, 155,100 linear feet, at 20 cents. 31,020 Sheet piling. 128,000 feet B. M., at $33 per M... 4,224 Iron, 80,000 pounds, at 3 cents. 2,400 Stoney gate to discharge 500 cubic feet per second. 2,000 Frankfort, N. Y. (No. 4): Excavation, 639,000 cubic yards, at 18 cents.t. 115,020 Masonry in dam and abutments, 21,835 cubic yards, at $6. 131,010 Timber in grillage foundation, 1,101,372 feet B. M., at $22 per M .... .. 24,230 Piles in foundation, 155,100 linear feet, at 20 cents. 31,020 Sheet piling, 128,000 feet B. M., at $33 per M.. 4,224 Iron, 80,000 pounds, at 3 cents.. 2,400 Railroad and highway changes. 450,170 Steam ferry, 1..... 20,000 462 DEEP WATERWAYS. Fort Bull-Herkimer section, etc.—Continued. Locks— No. 9 .... No. 10.... Lock-operating machinery— 2 sets single, at $100,000.... Bridges..... Right of way— Farm land, 10,335 acres at $100. $634. 481 634,481 200.000 966,724 1.033,500 Total 14,469,094 TOTAL COST. Lake Ontario section.......$1,226,701 Oswego-Minettosection.... .... 8.269,081 Minetto-Fulton section....... 805,169 Fulton-Brewerton section ........ 9,282,006 Brewerton-Lock No. 7 section.... 2,757,848 Lock No. 7-Fort Bull section ... 4,815,394 Fort Bull-Herkimer section... 14,469, 094 Water supply, Appendix No. 16......... 18.080.752 Total..... 59,706,045 ALTERNATIVE ROUTES AND PLANS. Minnetto to Oswego .—It was at first proposed to enter the river directly from the lake at Oswego, to excavate a channel as far as the present Oswego dam, and build there a dam and lock which would raise the water surface 30.6 feet above low water in the lake; to build another dam and lock having a lift of 30 feet about three-fourths of a mile farther up the river, and a third lock near Minetto, having a lift of 25 feet, which would raise the water surface to 331, the eleva¬ tion of the crest of the proposed Minetto dam, which has already been described. The last lock was located at the north end of a small val¬ ley lying on the east side of the ridge against which the east end of the proposed Minetto dam abuts, and which is known as Seneca Hill. Above the lock the canal was to pass through the valley in a southerly direction and enter the river again 14 miles beyond the lock. The use of this valley would avoid a rather sharp bend in the river and permit the location of t-lie lock considerably farther downstream than would otherwise be practicable, thus escaping a large amount of rock excavation in the river bed. The locks and dams were all so located as to be constructed readily and have good rock foundations. The river banks at the dam sites are composed of rock, which would afford secure abutments for the dams. The comparative estimate given below shows that the cost for excavation alone on the line adopted is, fora 30-foot channel, $865,097 less than on the river line: DEEP WATERWAYS. 463 Table No. 7. —Comparative estimate for excavation for 30-foot channels between Minetto and Lake Ontario. ADOPTED LINE. 1 Earth excavation, 12,080,380 cubic yards, at 20 cents.$2,416,076 Submerged rock excavation in Lake Ontario, 349,670 cubic yards, at $2. 699,340 All other rock excavation, 1,969,596 cubic yards, at 65 cents. 1,280,237 Total.... 4,395,653 RIVER LINE. Gravel dredging in Lake Ontario, 491.000 cubic yards, at 20 cents_ $98, 200 Earth excavation along Oswego River. 3,709,000 cubic yards, at 20 cents. 741,800 Submerged rock excavation in Oswego River, 929,000 cubic yards, at $2. 1,858,000 Rock excavation along Oswego River, 1,991,000 cubic yards, at $1.25_ 2,488,750 Removal of old cribs, 36,000 cubic yards, at$l .... 36,000 Removal of old masonry, 38,000 cubic yards, at$l . 38,000 Total........ 5,260,750 Co3t of excavation on adopted line - .. ... 4,395,653 Difference of cost in favor of adopted line'- 1 ...... 865,097 This difference in cost maybe partly explained by saying that where the proposed channel follows the river its estimated bottom width is 400 feet instead of 203 feet or 250 feet, the width of base used for earth and rock respectively in the regular canal prism, and also that because of the valuable property along the river banks it is imprac¬ ticable to build a dam and raise the water surface within 14 miles of the lake, and it therefore would be necessary to excavate a very large volume of rock under water, an expensive class of work, which is largely avoided on the other line. 3 There are, however, other and more important advantages in favor of the adopted line, which will be briefly mentioned. On the river line two masonry dams would be needed between Minetto and Oswego. In addition to the expense of these structures the present Minetto dam and the dam midway between Oswego and Minetto would be submerged and their water powers destroyed. The hydraulic canal on the east side of the river in Oswego would be intersected by the canal and its water supply cut off. The line would cross two liigh- 1 The figures and quantities here given are for the final adopted line with lock No. 4 at Minetto. ' 2 Lock No. 4 at Minetto. 3 The discussion of the comparative cost of the two lines was written by Mr. Himes before the location of lock No. 4 was changed to Minetto and before the lock estimates were completed. The cost of the locks and dams on the river line is estimated at $4,500,000. The cost of the locks on the adopted line is $5,660,913, leaving a difference of $1,160,913 in favor of the river line in the cost of structures. This will offset the above difference in cost of excavation in favor of the adopted line, so that the costs of the construction of the channels themselves will be about the same on both lines. The river line, however, destroys many very valuable water rights, and the right of way would be much more expensive than for the adopted line. 464 DEEP WATERWAYS. way bridges and one railway bridge in the city, and the necessary draws would be a source of danger and delay to the shipping, as well as to the street and railroad traffic. The same number of bridges would be needed on the other line, but the traffic upon them would be much lighter, and they would be less troublesome. But most important of all is the reduction in lift on the locks. The maximum lift on the adopted line is 21.4 feet, while on the river line it is 30.6 feet. The locks at Fulton would have lifts of 22.5 feet, and since the water used for lockage is regulated by that lock below the source of supply which consumes the most water, the saving of water on the adopted line would be equal to the difference between the vol¬ umes required by 22.5-foot and 30.6-foot locks, respectively. This saving would be considerable, and would materially lessen any possi¬ ble chance of a shortage in the supply. With the above advantages offered by the line west of the city, it was considered unnecessary to go any further with the estimates on the river line. Fulton to Peter Scott’s swamp .—It was supposed at first that the line beyond Fulton would follow the river to Phoenix, and then strike eastward into Peter Scott’s swamp instead of leaving the river at Fulton and running across country to the swamp, as previously described. The survey showed the adopted line to be 1.8 miles shorter than the line through Phoenix and it is estimated to require about 2,700,000 cubic yards less excavation. The estimate of the Phoenix line was made with less detail than the other, it being considered unnecessary to make a very close compari¬ son of the two lines. For that reason, the rock excavation on the Phoenix line was not separated from the earth. The rock excavation in the river at Fulton alone, however, is greater than all the rock excavation on the adopted line, and there is much more rock between Fulton and Phoenix. Such portions of the Phoenix line as lie in the river were estimated for a prism base of 400 feet as already described for the alternative line at Oswego. In addition to the above differences in distance and excavation, the line through Phoenix would submerge the upper Fulton dam and the Phoenix dam, and the injury to the manufacturing plants which use their powers would add a large item to the cost. A new dam would be needed at Fulton, and the lock at that place would require a lift of about 30 feet. The disadvantage of such a high lift has already been stated and it is sufficient here to explain the necessity of such a lift. The crest of the Phoenix dam has an elevation of about 361. The country above the dam is very low and flat, there being very little fall from Syracuse or Baldwinsville down to Phoenix. The area of land that would be affected by raising the canal water surface higher than the dam is so great, and about Syracuse the property bordering Onon- DEEP WATERWAYS. 465 daga Lake is so valuable, that the crest of the dam was taken as a limit above which the water must not be raised. It was necessary, therefore, to raise the water surface below Fulton from 331 to 361, the crest of the Phoenix dam. The cheapest way to do it is by a single lock; the way which would best economize the water is by two locks. In either case, by the high-level plan, a lock would be needed at Phoenix to rise from 361 to 376, so that if the two locks were used at Fulton there would be needed on the Phoenix line three locks to rise 45 feet, while on the adopted line the elevation is overcome by two locks. These differences between the lines were deemed sufficient ground for discarding further estimates on the Phoenix line. It should be stated, however, that the river from Phoenix to Oswego was fully developed by the survey, and the maps will permit as complete a study of these alternative lines as of the line selected. Peter Scott’s Swamp .—In locating a line through Peter Scott’s Swamp it was considered that, should a more thorough examination of it be made with a view to construction, and any reason be found why it would be inadvisable to build a dike and Hood the swamp as already described, it would be well to know that the general plan for a canal would not be impaired. For that reason an alternative line has been located around the north side of the swamp on higher and firmer ground. No borings were made on this line, but it is so close to the adopted line that the information obtained is sufficient to make it very probable that no rock excavation would be necessary. The following estimate shows the comparative cost of the two lines: Table No. 8 .—Comparative cost of lines passing Peter Scott's Swamp Sv-foot channel. LINE THROUGH SWAMP. Earth excavation: Prism, 8,020,000 cubic yards, at 18 cents.. .$1,443, COO Dike, 383,000 cubic yards, at 18 cents ... 68,940 Embankment, dike, 588,000 cubic yards, at 25 cents... 147,000 Riprap, 71,000 cubic yards, at 90 cents. 63,900 Total ..... 1,723,440 LINE NORTH OF SWAMP. Earth excavation, 10,440,000 cubic yards, at 18 cents... $1,879,700 Cost of line through swamp... 1,723,440 Difference of cost in favor of line through swamp.. 155,760 Sand Ridge Crossing .—The line through Sand Ridge is believed to be as cheap as any that can be found. It was located in the field for the purpose of making borings, and having been thoroughly devel¬ oped, all estimates have been based upon it. If the canal should be built north of Peter Scott’s Swamp, on the alternative line just described, the alignment could be improved by H. Doc. 149-30 DEEP WATERWAYS. 466 crossing the ridge a little farther north. A profile of the northern crossing indicates a slightly greater amount of excavation, but no borings were made upon it and so the amount of rock which might be encountered can not be estimated. Sod us route .—There is said to be an excellent natural harbor at Lit¬ tle Sodus, 25 miles west of Oswego, and an attempt was made to find a route for the canal which would make that port the northern ter¬ minus. The country on the west side of the Oswego River, between Phoenix and Fulton, is quite high, and the only possible opening is through Lake Neatahwanta, near Fulton. But the reconnoissance mentioned elsewhere showed the route to be so long and tortuous as to be wholly impracticable, and no survey was made of it. Change in Jocks .—A rearrangement of locks on the proposed line north of Minetto is possible, by which the total lift from Lake Ontario to the river level, of 85.6 feet, could be made in three equal lifts, instead of four as proposed. The lift of each lock would then be 28.5 feet, instead of 21.4 feet, as before, and the principal results of the change would be an increased consumption of water and a saving of time in navigation equal, approximately, to that required to pass through one lock. The difference in cost would probably not be great, though no comparative estimate has been made, because it has been considered throughout that the plan which would require the least water to operate the canal would be preferable, at least until the adequacy of the water supply shall be finally demonstrated. A similar change can be made near Fulton, in the low-level plan, to be described later. In this case three locks, two of them having equal lifts of 18 feet, and the third a lift varying between 12 feet and 10 feet, according to the stage of the Oneida reservoir, can be replaced by two locks, the lower one to have a lift of, say, 26 feet and the upper one a lift to vary between 22 feet and 20 feet. The effect of this change would be precisely similar to the first, and its feasibility depends almost wholly upon the water supply. CHANGES IN RAILROADS AND HIGHWAYS. At Oswego, where the canal crosses the Delaware, Lackawanna and Western Railroad and the Rome, Watertown and Ogdensburg Rail¬ road, a slight change of alignment would permit both roads to use one double-track drawbridge. The Rome, Watertown and Ogdensburg freight yard would require some alteration and the side track running down to the lake front should be moved to the east side of the canal. The track of the Delaware, Lackawanna and Western Railroad would need to be raised a few feet for a short distance at Minetto. 1 The New York, Ontario and Western Railway would require a single-track drawbridge at the canal crossing in the northern part of 1 This will not be necessary with the change ol' location of lock No. 4. DEEP WATERWAYS. 467 Fulton and a change of alignment between the bridge and the station. Instead of providing another drawbridge for the same road near Ingalls, the estimate covers the relocation and construction of the road from Ingalls to the Oswego River, on the east side of the canal. The Fulton travel would be obliged to cross the highway bridge in the eastern part of the village and the freight traffic could enter the vil¬ lage over the bridge in the northern part of the village. The Rome, Watertown and Ogdensburg Railroad would require a single-track drawbridge where it crosses the canal at Brewerton. At Sylvan Beach the New York, Ontario and Western and Lehigh Valley railroads now use in common a single-track bridge over Wood Creek, and one single-track drawspan over the canal is all that would be needed at that place for both roads. Each road has several miles of track that would need to be either raised and the embankments protected with riprap or else rebuilt on higher ground. The estimate for the high-level plan provides for raising the track; for the low-level plan the water surface in the Oneida reservoir would be so much higher that it is cheaper to change the location and rebuild the roads, so the estimate is based on such a change. According to the latter plan, an additional 4 miles of New York, Ontario and Western Railway track, on the north side of Oneida Lake, would need to be raised. At Rome the estimate provides for a single-track drawbridge, to be used by the New York, Ontario and Western and Rome, Water- town and Ogdensburg railroads, and for building a four-track road entirely south of the canal, to be used by the New York Central and Hudson River Railroad. At present the immense passenger and freight traffic of that railroad passes through Rome and crosses the streets at grade. Many trains make no stop there, but go through the city at a high rate of speed, which is a constant menace to the public safety and a great expense to the company. If all trains were run over the line proposed the dangerous grade crossings would be abolished, a very sharp curve near the station would be eliminated, and through trains would not be obliged to slow up in passing the city. Rome passengers could cross the canal on the Rome, Water- town and Ogdensburg and New York, Ontario and Western trains, and the arrangement would be a great improvement for both the pub¬ lic and the railroads. The plan would, of course, meet with opposi¬ tion, but it is probably the most expensive one that would be proposed; so whether it should ever be constructed or not, it will serve the pur¬ pose of making the present estimate large enough to cover any other plans that may be devised. The advantage to the canal which justifies the plan requiring the largest first cost is the avoidance of two four-track drawbridges which would otherwise be necessary, and the operation of which would be no less vexatious and expensive to the railroad than to the canal. DEEP WATERWAYS. 468 The next railroad crossing is that of the Black River branch of the New York Central and Hudson River Railroad, near Utica. It would require a single-track drawbridge. The next and last crossing is that of the New York Central and Hudson River main line, 4.7 miles east of Utica. This crossing is unavoidable and requires a four-track drawspan. 1 The angle between the canal line and the railroad is very acute, and it will be necessary to rebuild the line far enough on each side of the canal to permit the angle of crossing to approximate 90 degrees. The above crossings would be approximately the same for 21-foot or 30-foot channels on either the high-level or low-level plan. Between Utica and Frankfort, according to the high-level plan, the water surface would have an elevation of 396, and several miles of the West Shore Railroad would need to be raised a few feet. With the low-level plan the high-water surface would be about 10 feet lower and no change would be needed. Of the highways affected by the canal, drawbridges have been esti¬ mated for those near cities and villages which have considerable travel. There are two in Oswego and one each at Minetto, Fulton, Rome, Utica, Frankfort, Ilion, and Herkimer. The single-track railroad bridges at Fulton, Brewerton, and Sylvan Beach may also be used for highway purposes, since the railroad traffic upon them is not heavy. Table No. 2 gives the location, kind, and cost of all bridges on the high-level plan for both the 30 and 21 foot channels. In all other cases where crossings are necessary steam ferries have been estimated for. In cases where two or more highways intersect the canal within a comparatively short distance the estimates provide for the construc¬ tion of new roads sufficient to bring them together at a single bridge or ferry. RIGHT OF WAY. Besides the ground required by the canal itself a considerable area must be provided for spoil banks, and in the Mohawk Valley those banks should be as far as practicable from the canal, so as not to cause its slopes to cave into the water. About Oneida Lake a large area of land would be submerged, and it also is included in the estimate. The submerged area is much greater for the low-level plan than it is for the high-level plan. The only water powers affected by either plan are those at the lower Fulton dam and at the Battle Island dam on the Oswego River. These two dams would be completely submerged, and their values for power purposes have been included in the estimate. LOW-LEVEL PLAN. The conversion of Oneida Lake into a storage reservoir and the excavation of a channel through the Rome divide, deep enough to 1 Estimated as two double-track drawspans. DEEP WATERWAYS. 469 extend the canal from the lake into the Mohawk Valley without locks would create a summit level 72.084 miles in length. The level would begin a short distance east of Fulton, and its eastern terminus would be at Frankfort. The surface elevation of the reser¬ voir when full would be 386; when empty it would be 379, thus making a storage depth of 7 feet. The summit level at Rome under the former plan has a surface elevation of 416, which is 30 feet higher than the reservoir surface when full, and this plan will therefore effect a saving in lockage of 60 feet or more, according to the stage of the reservoir. It will be observed that, according to the low-level plan, the surface of Oneida Lake is higher than under the high-level plan. No confu¬ sion will result from this fact if it is remembered that the terms are used only with reference to the summit levels. The raising of Oneida Lake will increase the area of its water surface from 77.3 square miles at its present normal stage of 371 to 148.7 square miles when the reservoir is full. The area to be flooded includes the whole of Sylvan Beach, small portions of Cleveland, Constantia, and Brewerton, about 12 square miles of swamp, and a considerable area of farming land. Several miles of railroad about the eastern end of the lake would also lie submerged and would need to be relocated on higher ground. The survey about the lake extended generally only as high as the 380 contour, and beyond that the estimates are based upon the maps of the United States Geological Survey. The weir at the crossing of the Oneida River west of Brewerton would be located as before, but its crest would be 10 feet higher, and a bulkhead with sluice gates would be needed to permit the discharge of water when needed for power purposes in the Oswego River. 1 The embankment at the east end of the weir would be required to retain water about 16 feet in depth. In passing Peter Scott’s swamp the line selected runs around the swamp on the north instead of through it as before, because the water being 10 feet higher the increased cost of the embankment necessary to raise the level of the swamp and the difficulty of maintaining so high an embankment would remove whatever advantage the line through the swamp may possess for the other plan. There is also a slight change of alignment in Wood Creek Valley. It is necessary, as has been explained before, that the high-level line should leave the creek and rise on higher and firmer ground so as to go over the divide. For the low-level line it is better to remain in the bottom of the valle}", both because the volume of excavation would be less and because the material to be handled is not so hard and could be excavated at a smaller cost per cubic yard. This change is, how¬ ever, so slight that the curvature and distance are nearly the same. 'Til's weir was estimated according to the plan adopted by the Board and explained in a lootnote describing the same for the high-level plan. 470 DEEP WATERWAYS. The two locks at Fulton will'have lifts of 18 feet instead of 22.5 feet as before, and a third lock, with a lift ranging from 12 feet to It) feet, according to the stage of the reservoir, is located near the crossing of the New York, Ontario and Western Railway, 2.5 miles southeast from Fulton. Since the lift of this lock will generally be less than that of the locks below, and, therefore, require less water for lockage, provision is made to pass water around it to the lower locks by means of a bulkhead with sluice gates and a by-pass. The two locks between Oneida Lake and Rome and the lock at Oriskany are not required in this plan. The next lock, and the last one in this division, is located at Frankfort. Its lift is 10 feet when the reservoir is full and 3 feet when it is empty. The water surface below the lock will have an elevation of 376, as before. 'The Frankfort dam in this case has a fall of 10 feet* instead of 20 feet, as under the high-level plan, and it has a bulkhead with gates to pass water on to the locks below whenever that may be necessary. 'Fhe streams tributary to the line of the canal in the Mohawk Val¬ ley are received into the canal as before, save that on account of the greater depth of the channel the intake works are more extensive and their cost is increased. It is a part of the low-level plan to store the water of Salmon River and allow it to flow down Fish Creek into Oneida Lake, thereby increasing the available water supply. An estimate for the necessary reservoir on Salmon River has been made in Appendix No. 16 and is added to the estimate for this plan. The volume of water which it is expected to receive from Salmon River is such that it may be carried down Fish Creek without any considerable improvement to its channel. The gauging of the stream at McConnellsville, made by the water supply division in the spring of 1898, showed a flood discharge of 4,100 cubic feet per second, while the amount which is expected to pass through it from Salmon River is only 350 cubic feet per second, and will be controlled by gates, so that in case of floods it may be shut off entirely. It is necessary to excavate a channel through the ridge separating Salmon River from Fish Creek. This channel is 3.5 miles long and has a slope of 0.00015. It is 10 feet deep, has a bottom width of 15 feet and side slopes of 1 on 2. Its cross section and slope are such that it will carry about twice the proposed discharge, but it may be desirable at times to increase that amount, and, moreover, the flow may be lessened at certain seasons by the growth of vegetation, so that a smaller channel would be found less satisfactory. In addition to the estimate of the reservoir there is a dam across the Salmon River at the entrance to the proposed feeder, and several smaller dams across creeks which intersect the feeder, for which esti¬ mates have been made, and they are included herein, being indicated in the tabulated estimate by a footnote. Table No. 9 gives the lengths of levels and lifts and costs of the DEEP WATERWAYS 471 locks; Table No. 10 gives the location, kind and cost of the bridges for both 30 and 2 1 foot channels; Table No. 11 shows ihe alignment; Table No. 12 gives the different classes of navigation, with the per¬ centages of each; and Tables Nos. 13 and 14 give the total estimated costs of 21-foot and 30-foot channels, respectively, all being based on the low-level plan. Table No. 9. Num¬ ber of Location of lock. Single or double Elevation of low water. Lift in feet. Length of level lock. lock. Below. Above. in miles. 1. Oswego, N. Y . Single.... Double... 245. 4 266. 8 21.4 Hand 3 .. do .. 266. 8 309.6 42.8 .890 4 . .. . Minetto, N. Y.. Siugie.... Hut*. 6 331.0 21.4 4.384 5 . Fulton, N. Y . _do_ 331.0 349.0 18.0 5 900 6. . do. .do... 349.0 367. 0 18.0 1. 193 7. _do. _do.... 367.0 379.0 1 la. 0- 19.o 1.648 8. Frankfort. N. Y .. ... __ _do.... 376.0 379.0 1 3.0- 10.0 72.084 Head gates lock 10 to end of division 4.529 Total .. 136.6-150.6 90.628 1 Reservoir full. COST. Number of lock. 30-foot channel. 21-foot channel. Operating machinery, 30 foot and 21 foot channels. *100.000 1 . .... $1,054,942 3, 456,9,"0 1,1(1.*. 991 1,001.2 1 1.004.381 1,034,907 913,106 *641.108 2 and 13......... 2.179.291 175.001) 4 . ... 681, 426 609,689 607.149 624,589 552,712 100.000 100,000 100,000 100. OIK) loo.ooo 6.. ... 8 .. . . Total..... 9,575,538 775,000 5.895,964 775,000 775,000 Operating machinery. Total... .. 10,350,538 6,670,964 Table No. 10 .—Bridges for luiv-level plan. Location. Station. Kind of bridge. Number of tracks. Fixed or swing. 30-foot channel. 21-foot channel. Number of spans. Total length. Estimated cost. Total length. Estimated cost. Oswego, N. Y. 15 Highway. Swing . 1 545.0 $68,240 525.0 $66, 796 Do.. 87 Railway . O ....do. . 1 550.0 221,429 530.0 196,161 Near Oswego .. . 110 Highway ... do ... 1 545.0 116,534 525.0 98,138 275 .do_ _do .. - 1 235 .0 19,986 195.0 Fulton... 554 Railway 1 _do ... 1 537.5 152,916 517.5 132,174 Do . 582 Highway. 1 235.0 19,986 19.5.0 16.650 Brewerton- 1403 Railway 1 _ do... 1 537.5 156,434 517.5 149,118 Near Sylvan Beach _ 2740 _do_ 1 _ do... 1 537.5 156,538 517.5 139,292 Highway . . do 1 545.0 93 764 525 0 92 320 Do. 3304 Railway . 1 ...do... 1 537.5 125,218 517.5 117,054 1 . do 1 152 800 517 5 130.810 Do 4038 Highway. _ do... 1 545.0 106,950 525.0 91,356 Near Utica_ .. — 4236 Railway *> ....do... 1 550.0 238.139 530.0 209,113 Do .. . 4242 do . . ‘) _ do . . 1 550.0 2: * 8 , 139 530.0 209,113 Frankfort . . 4555 Highway. _ do ... 1 235.0 19,986 195.0 16.650 Ilion . 4648 _ do _ . .do... 1 545.0 103.354 525.0 84,086 Herkimer . 4747 . do _ _ do ... 1 545.0 103,354 525.0 84,086 2,093,767 1,849,567 472 DEEP WATERWAYS. Table No. 11.— Low-level plan. Length of tangent. Length of curve. Minimum radius of curvature. Degrees of curvature. Percent¬ age of tangent. Percent¬ age of curvature. Miles. TO. 762 Miles. 19.866 Feet. 4,533.4 O / 705 40 78.08 21.92 Table No. 12.— Loir-level plan. Miles. Per cent. Open water .......... 20.928 4.962 64.738 23.09 5.48 71.43 Improved water wavs. . Canal prism.-...-... 90.628 100.00 Table No. 13. — Oswego-Mohawk Route, Western Division. LOW LEVEL PLAN—30-FOOT CHANNEL. Lake Ontari > section (station —26 + 00 to —9+20): Excavrtioa— Submerged rock. 319,670 cubic yards, at $2.. $699,340 Breakwater, Appendix No. 3 . ... .. 1,190,317 Crib work— Pine. 2,415,600 feet B. M., at $30 per M.. 72,468 Hemlock, 5,840,620 feet B. M., at $23 per M„. 134,334 Oak, 97,920 feet B. M., at $50 per M. 4,896 Iron, 742.870 pounds, at 3 cents... 22,286 Stone Mil, 102,034 cubic yards, at 60 cents. 61, 220 Total..... ... .. 2,184,861 Oswego-Minetto section (station —9+20 to 288+00:: Excavation— Earth, dry, 11,866,321 cubic yards, at 20 cents. 2,373,264 Rock, dry, 1,969,596 cubic yards, at 65 cents. 1,280,237 Retaining wall, 143,479 cubic yards, at $4 ... 573,916 Masonry in by-passes, 4,300 cubic yards, at $4.... 17,200 Slope wall, 68,448 s iuare yards, at $1.10 .. .. 75,293 Back fill, 449,336 cubic yards, at 25 cents. 112,334 Crib work— Pine. 2,374,400 feet B. M., at $30 per M. 71,232 Hemlock, 6,992,000 feet B. M., at $23 per M. 160,816 Oak, 96,920 feet B. M., at $25 per M... 2,423 Iron, 786,632 pounds, at 3 cents... ... 23,599 Stone fill, 125.856 cubic yards, at 60 cents. 75,514 Railroad and highway changes. 29,660 Locks— No. 1.... . 1,054,942 Nos. 2 and 3...... 3,456,980 No. 4.. 1,109,991 Lock-operating machinery— 2 sets, single, at $100,000. 200,000 1 set, double. 175,000 DEEP WATERWAYS. 473 Oswego-Minetto section, etc.—Continued. Bridges.. .... $426,189 Right of way— City property, 277 acres, at $2,080 ...... 576,160 Farm laud, 729 acres, at $200. 145,800 Total_______ 11,940,550 Minetto-Fulton section (station 28S+00 to 550+00): Excavation- Earth, 1,911,170 cubic yards, at 18 cents. 344,011 Rock. dry. 117,268 cubic yards, at 65 cents. 76,224 Slope walls. 14.508 square yards, at $1.10... 15,959 Minetto dam (No. 1) — Excavation, 25,567 cubic yards, at 20 cents. .. 5,113 Masonry in dam and abutments, 44,913 cubic yards, at $6. 269,478 Cofferdams, estimated cost .. 55,000 Highway and railroad changes.. 25,670 Right of way— Farm land, 1,154 acres, at $100.. .... 115,400 Swamp land, 408 acres, at $12.50... 5,100 Water rights, Battle Island Dam .. 150,000 Total. 1,061.955 Fulton-Brewerton section (station 550+00 to 1410-1-00): Excavation— Earth, 29,876,982 cubic yards, at 18 cents.. 5,377,857 Rock, 2,033.872 cubic yards, at 65 cents.... 1,322.017 Masonry— Retaining wall, 198,210 cubic yards, at $4 .. 792,840 By-pass, 600 cubic yards, at $4 _ ..-... 2,400 Slope wall. 513,405 square yards, at $1.10 . 564, 746 Back fill, 723.338 cubic yards, at 25 cents.. 180,834 Crib work— Pine. 2.381.600 feet B. M., at $30 per M ... 71,448 Hemlock, 9.173,000 feet B. M.. at $23 per M ... 210,979 Oak, 77,000 feet B. M., at $50 per M. 3,850 Iron, 979,024 pounds, at 3 cents... 29,371 Stone fill, 154,800 cubic yards, at 60 cents.. 92,880 Oneida River dam— Excavation, 51,000 cubic yards, at 18 cents__ 9,180 Em' ankment, 21.000 cubic yards, at 15 cents.. 4,050 Masonry in dam and abutments, 21,270 cubic yards, at $6. 127,620 Stoney gates.. ..... 46,250 Railroad and high way changes... . ... 73,400 Steam ferries, 4, at $20,000.. 80,000 Locks— No. 5. 1,001,231 No. 6.. 1,004,381 No. 7. 1,034,907 Lock-operating machinery. 3 sets single, at $100,000 .. 300.000 Bridges. 329,336 474 DEEP WATERWAYS. Fulton-Brewerton section, etc.—Continued. Right of way— Village property, Fulton, N. Y., 55 acres, at $2,600.. $143,000 Farm land, 5.910 acres, at .$100... 591,000 Swamp land, 2,446 acres, at $12.50____ 30,575 Water rights.......... 806.000 Total_____- 14,230,152 Brewerton-Canada Creek section (station 1410+00 to 2970+00): Excavation— Earth. 21,464.652 cubic yards, at 8 cents... 1,717,172 Rock, wet, 505,500 cubic yards, at $2.. 1, Oil. 000 Slope wa.l, 259,231 square yards, at $1.10____285,187 Railroad and highway changes..... 264,675 Bridges ..-____- 156,538 Right i f way— Farm land, 36,714.65 acres, at $100 ...... 3,671.465 Swamp land, 4,800 acres, at $12.50 ..... 60,000 Salmon River feeder— Earth excavation, 1,207,300 cubic yards, at 15 cents.. 181,095 Reservoir 1 ....'...... 1,350,000 Dam No. 53 1 ...... 366,500 Dam No. 55 1 ........ 5,100 Dam No. 56 1 .... 1,000 Dam No. 57 C... ...... .57,000 Dam No. 58*._. 87,600 Guard gate, No. 7 1 .... 14,000 Total...... 9,228,332 Canada Creek-Fort Bull (station 2970+00 to 3165+00): E cavation— Earth, 12,387,183 cubic yards, at 20 cents... 2,477,437 Rock, dry, 2,789,719 cubic yards, at 70 cents.. 1,952,803 Masonry- Retaining wall, 219,402 cubic yards, at $4 . 877,608 Receivers. 5.500 cubic yards, at $4 .... 22,000 Slope wall, 28.713 square yards, at $1.10... 31,584 Back fill. 512.027 cubic yards, at 25 cents.. 128,007 Highway changes.......... 600 Steam ferry. 1 ..... 20.000 Right of way— Farm land, 943 acres, at $100... 94,300 Total...,. 5,604,339 Fort Bull-Herkimer section (station 3165+00 to 4789+91.5): Excavation— Station 3165 to 3280— Earth, 12,861,717 cubic yards, at 22 cents .. 2,829, 578 Rock, 102,500 cubic yards, at 70 cents . 71,750 Station 3280 to 3420- Earth, 9,245,002 cubic yards, at 18 cents. 1,664.100 Rock, 3,902,848 cubic yards, at 70 cents.. .. 2, 731,994 'From estimate for water supply, high level. Appendix No. 16. DEEP WATERWAYS. 475 Fort Bull-Herkimer section, etc.—Continued. Excavation—Continued. Station 8420 to 3500— Earth. 16,044,368 cubic yards, at 22 cents..$3,661,761 Rock, 133,235 cubic yards, at 70 cents._. 93,265 Station 3590 to 3780— Earth, 9,021,975 cubic yards, at 18 cents____ 1,623,956 Rock. 4,465,075 cubic yards, at 70 cents. 3,125,553 Station 3780 to 3980— Earth, 15.008,168 cubic yards, at 18 cents.. 2,701,470 Rock, 121,002 cubic yards, at70 cents.. 84, 701 Station 3980 to 4180— Earth, 12,390,681 cubic yards, at 22 cents .. . 2,725,950 Station 4180 to 4533— Earth, 18,295,204 cubic yards, at 22 cents___ 4,024,945 Station 4533 to 4789+91.5— Earth, 11,123,827 cubic yards, at 18 cents.. 2,002,289 Rock, 117,023 cubic yards, at 70 cents. ... 81.916 Excavation for receivers not in above, 520,300 cubic yards, at 18 cents 93.654 Excavation for drainage channels— Earth, 6,568,000 cubic yards, at 15 cents.... 985, 200 Retaining wall, 341,268 cubic yards, at $4 ..... 1,365,072 Masonry for receivers and by-passes, 41,100 cubic yards, at $4_ 164,400 S.ope wall, 991,540 square yards, at $1,10... .. . 1,090, 694 Back fill, 862,371 cubic yards, at 25 cents . ..... 215,593 Crib work— Pine, 1,431,600 feet B. M., at $30 per M .. 42, 948 Hemlock. 4,553,340 feet B. M., at $23 per M. 104, 734 Oak, 75,840 feet B. M., at $50 per M.. 3, 792 Iron. 523,344 pounds, at 3 cents.... 15. 700 Stone fill, 79,126 cubic yards, at 60 cents ..47,476 Frankfort dam— Excavation, 90,000 cubic yards, at 18 cents.. 16,200 Masonry in dam and abutments, 2,650 cubic yards, at $6. 15,900 Gates... . 6,000 Railroad and highway changes..... 412,320 Steam ferry, 1 . ... . 20,000 Lock, No. 8.... ..... 913,106 Lock-operating machinery— 1 set, single.... 100.000 Bridges......... 1,181.704 Right of way— 10,335 acres, at $100..... 1,033,500 Total........ 35,251,221 TOTAL COST, Lake Ontario section . ... $2, J84.861 Oswego-Minetto section.... 11,940,550 M netto-Fulton section ...... 1,061,955 Fulton-Brewefton section ..... 14,230,152 Brewerton-Canada Creek section.... 9,228,332 Canada Creek-Fort Bull section ...... 5,604,339 Fort Eu l-Herkimer section . . 35,251,221 Total.... 79,501,410 476 DEEP WATERWAYS. Table No. 14. — Oswego-Mohciwk route, western division. LOW-LEVEL PLAN, 21-FOOT CHANNEL. Lake Ontario section (station —20+00 to —9+20): Excavation— Submerged rock, 152,361 cubic yards, at $2 .... $304,722 Breakwater, Appendix No. 3...... 721,380 Crib work— Pine, 2,416,000 feet B. M., at $30 per M... 72,480 Hemlock, 3,096.800 feet B. M., at $23 per M... 71,226 Oak. 83,600 feet B. M.. at $50 per M..... 4,180 Iron. 490,585 pounds, at 3 cents .... 14,718 Stone fill, 63,325 cubic yards, at 60 cents. 37,995 Total.... 1,226,701 Oswego-Minetto section (station —9+20 to 288+00): Excavation— Earth, dry, 9,300,777 cubic yards, at 20 cents.. 1,860,155 Rock, dry, 843,884 cubic yards, at 65 cents ... 548, 525 Retaining wall, 97,906 cubic yards, at $4.... 391,624 Masonry in by-passes, 4,300 cubic yards, at $4. 17, 200 Slope wall, 88,288 square yards, at $1.10... 97,117 Back fill, 348,900 cubic yards, at 25 cents .... 87,225 Crib work— Pine, 2,374,400 feet B. M., at $30 per M.. 71,232 Hemlock, 4,894,400 feet B. M.. at $23 per M____ 112,571 Oak, 53,760 feet B. M., at $50 per M.... 2, 688 Iron, 593,600 pounds, at 3 cents ....... 17,808 Stone fill. 94,576 cubic yards, at 60 cents .. 56.746 Railroad and highway changes.•.. 29,660 Locks— No. 1..... 641,108 Nos. 2 and 3........ 2,179,291 No. 4............ 681,426 Lock-operating machinery— 2 sets single, at $100,000.... 200.000 1 set double .. 175,000 Bridges . 377,745 Right of way— City property, 277 acres, at $2,080 . 576,160 Farm land, 729 acres, at $200.. 145,800 Total..... 8,269,081 Minetto-Fulton section (station 288+00 to 550+00): Excavation— Earth, 914,740 cubic yards, at 18 cents. 164,653 Slope wall, 13,414 square yards, at $1.10_. 14,755 Minetto dam— Excavation, 25,567 cubic yards, at 20 cents. . 5,113 Masonry in dam and abutments, 44,913 cubic yards, at $6. 269. 478 Cofferdams _ .. ... 55,000 Railroad and highway changes. 25,670 DEEP WATERWAYS. 477 Minetto-Fulton section, etc.—Continued. Right of way— Water rights, Battle Island dam.... $150,000 Farm land, 1,154 acres, at $100.... 115,400 Swamp land, 408 acres, at $12.50... 5,100 Total_______ .. 805,169 Fulton-Brewerton section (station 550+00 t,o 1410+00): Excavation— Earth, 22,243,558 cubic yards, at 18 cents-... 4, 003,840 Rock, 905,227 cubic yards, at 65 cents .... 588,398 Masonry— Retaining wall, 128,268 cubic yards, at $4.. 513,072 By-pass, 600 cubic yards, at $4...„. 2,400 Slope wall, 530,906 square yards, at $1.10..... 583,997 Back fill, 465,160 cubic yards, at 25 cents.. 116,290 Crib work— Pine, 2,222,000 feet B. M., at $30 per M .. 66,660 Hemlock, 3,794,260 feet B. M., at $23 per M...... 87,268 Oak, 85,800 feet B. M., at $50 per M..__. 4,290 Iron, 918,168 pounds, at 3 cents ... 27,545 Stone fill, 230,243 cubic yards, at 60 cents ..... 138,146 One - da River dam— Excavation, 51,000 cubic yards, at 18 cents... 9,180 Embankment, 27,000 cubic yards, at 15 cents __ 4,050 Masonry in dam and abutments, 21,270 cubic yards, at $6. 127,620 Stoney gates...... 46,250 Railroad and highway changes....... 73,400 Steam ferries, 4, at $20,000..._. 80,000 Locks— No. 5..... 609,689 No. 6......... 607.149 No. 7... 624,589 Lock operating machinery— Three sets, single, at $100,000 .. 300,000 Bridges . _....... 297,942 Right of way— Village property, Fulton. N. Y., 55 acres, at $2,600. 143,000 Farm property, 5,910 acres, at $100... 591,000 Swamp land, 2,446 acres, at $12.50..... 30,575 Water rights, Fulton, N. Y.. 806,000 Total.. 10,482,350 Brewerton-Canada Creek section (station 1410+00 to 2970+00): Excavation— Earth, 14,079,204 cubic yards, at 8 cents.... 1,126,336 Rock, wet, 30,646 cubic yards, at $2. 61,292 Slope wall, 259,261 square yards, at $1.10.... 285,187 Railroad and highway changes...... 264,675 Bridges. . ...... 139,292 Right of way— Farm land. 36,714.65 acres, at $100.. 3, 671,465 Swamp land, 4,800 acres, at $12.50. 60.000 478 DEEP WATERWAYS. Salmon River feeder: Earth excavation. 1,207,300 cubic yards, at 15 cents. $181,095 Reservoir'.......... 1,350,000 Dam No. 53 1 ..._..... 366,500 Dam No. 55 1 ......._... 5,100 Dam No. 56 1 ...... 1,000 DamNo.57 1 ....... 57,000 Dam No. 58 1 ........,.- . 87,600 Guard gate No. 7 1 ....... 14,000 Total... ..... 7,670,542 Canada Creek-Fort Bull section (station 2970 + 00 to 3165+00): Excavation— Earth, 11,773.284 cubic yards, at 20 cents.... 2,354,657 Rock, 1,364,600 cubic yards, at 70 cents. 955,220 Masonry— Retaining wall, 96,670 cubic yards, at $4.. 386,680 By-pass, 5,50.) cubic yards, at $4. ..... 22,0.JO Slope wall, 73,683 square yards, at $1.10__ 81,051 Back fill, 193,550 cubic yards, at 25 cents..... 48, 388 Highway changes....... 600 Steam ferry, 1. _.. 20,000 Right of way— Farm land, 943 acres, at $100...... 94,300 Total......... 3.962,896 Fort Bull-Herkimer section (station 3165+00 to 4789+91.5): Excavation— Station 3165 to 3280— Earth, 11,421,882 cubic yards, at 22 cents. 2,512,814 Rock, 41,015 cubic yards, at 70 cents....,... 28, 710 Station 3280 to 3420- Earth, 9,005,048 cubic yards, at 18 cents.... 1,620,909 Rock, 2,587,541 cubic yards, at 70 cents..... 1,811,279 Station 3420 to 3590— Faith, 14,530,927 cubic yards, at 22 cents.. 3,196,804 Stat on 3590 to 3780— Earth, 8,528.606 cubic yards, at 18 cents. 1,535,149 Rock, 2,773,258 cubic yards, at 70 vents.. 1,941,281 Station 3780 to 3980— Earth, 12,649,373 cubic yards, at 18 cents... 2,276,887 Rock. 121,002 cubic yards, at 70 cents.. 84,701 Station 3980 to 4180— Earth, 10,285,965 cubic yards, at 22 cents.___ 2,262,912 Station 4180 to 4533— Earth, 14,605,749 cubic yards, at 22 cents.. 3,213,265 Station 4533 to 4789+91.5— Earth, 8,600,133 cubic yards, at 18 cents... 1,548,024 Rock, 25,190 cubic yards, at 70 cents..•.. 17,633 Excavation for receivers not in above— Earth, 524,046 cubic yards, at 18 cents... 94,328 1 From estimate for water supply, high level, Appendix No. 16. DEEP WATERWAYS. 479 Fort Bull-Herkimer section, etc.—Continued. Excavation for draining channels— Earth, 6.568,000 cubic yards, at 15 cents. $985, 200 Retaining wall, 200,096 cubic yards, at $4. .. 800,384 Masonry for receivers, etc., 41,100 cubic yards, at $4. 164,400 Slope wall, 1,020,253 square yards, at $1.10 ._... 1,122,278 Back fill, 484,170 cubic yards, at 25 cents .. 121,043 Crib work— Pine, 1,770,000 feet B. M., at $30 per M.... 53,100 Hemlock. 6,328,300 feet B. M., at $23 per M . 145, 551 Oak, 88,920 feet B. M., at $50 per M_. 4,446 Iron, 714.144 pounds, at 3 cents. 21, 424 Stone till, 101,872 cubic yards, at 60 cents.__ 61.123 Frankfort dam— Ex avation, 90,00 ) cubic yards, at 18 cents ... _ 16,200 Masonry in dam and abutments, 2,650 cubic yards, at $6_ 15, 900 Oates. .. . . .. 6,000 Railroad and highway changes.... 412,320 Steam ferry, 1 .. . 20,000 Lo/kNo. 8.... .. 552,712 Lock-operating machinery— 1 set single....... 100.000 Bridges ....... 1,034,588 Right of way— Farm land, 10,335 acres, at $100 .. 1.033, 500 Total.i....... 28,814,865 TOTAL COST. Lake Ontario section........$1,226,701 Oswego-Minetto section...... 8,269,081 Minetto-Fulton section .......... 805,169 Fulton-Brewerton section..._. 10,482,350 Brewerton-Canada Creek section... 7,670.542 Canada Creek-Fort Bull section ..... 3,962,896 Fort Bull-Herkimer section.._.. 28,814,865 Total..... . 61,231,604 WATER SUPPLY—LOW-LEVEL PLAN. For the purpose of this discussion the period front April to Novem¬ ber, inclusive, is called the navigation period, and that from Decem¬ ber to May, inclusive, the storage period. It will be observed that the first two months of the navigation period are the last two months of the storage period. The amount of water required for the summit level may be divided into two parts—that lost by evaporation and that consumed by lock¬ age. To determine the amount of water which may be lost by evap¬ oration, reference is made to the measurements of evaporation at the Rochester waterworks, given in Table No. 15, and to the record of pre¬ cipitation during the navigation period for those portions of the Oneida and Mohawk watersheds which will drain into the proposed summit level, given in Table No. 16. 480 DEEP WATERWAYS Table No. 15. —Showing evaporation during navigation periods at Rochester , N. Y., 1802 to 1896, inclusive . 1 Inches. 1892 1893 1894 1895 30.86 30.02 29.27 34.38 1896 32.23 Table No. 16. —Record of precipitation during naviga tion periods on portions of Oneida and Mohawk watersheds tributary to proposed summit level, 1S26-1S98, inclusiver [Depths given in inches.] 1826. 1827. 1828. 1829. 1830. 1831. 1832. 1833. 1834. 1835. Pomney.. Utica.. 20.81 27. 75 36. 70 28.94 30.88 23.90. 24.97 26. 75 34.28 31.00 26.33 29.75 22.40 34.21 27.94 28.59 28.94 30.24 26.85 25.89 17.72 24.26 30.42 26.61 18.24 Average_ 20.81 32.23 29.91 24.44 30.68 28.04 28.18 29.26 23.49 24.88 1836. 1837. 1838. 1839. 1840. 1841. 1842. 1843. 1844. 1845. Pompey.. Utica... Cazenovia.. 14.63 19.37 22.26 25. 71 30.35 17.53 20.97 35.54 21.01 20. 72 24. 92 25. 43 20.26 25.60 29.44 33.46 41.20 19.16 29.94 21.23 23.53 41.34 32.16 24.52 33. 95 34.20 23.02 29.62 28.65 26.90 Average. 18.75 24.53 25.84 22.83 32.43 23.44 32.34 30.89 26.32 27.78 1846. 1847. 1848. 1849. 1850. 1851. 1852. 1853. 1854. 1855. 40.59 28.38 31.59 29.44 31.39 28.73 23.13 23.63 Utica 28.08 29.71 25 22 28.53 29.51 30.07 33.20 31.83 30.38 24. 30 38.07 24.80 Palermo.. Average_ 28 90 25.22 28.53 29.51 40.59 29.23 31.41 30.65 25.94 28.83 1856. 1857. 1858. 1859. 1860. 1861. 1862. 1863. 1864. 1865. 28.69 21.19 32.00 28.57 25.21 33. 40 Clinton.. Palermo.. 23.00 25.90 48.90 30.30 32.29 28.10 35. 64 33.90 37. 05 31.40 32.20 27.12 33.80 33.51 32.05 Oneida 48.26 Average. 24.45 39.60 30.20 34. 77 34.23 32.48 27.29 29.06 31.48 40.16 1866. 1867. 1868. 1869. 1870. 1871. 1872. 1873. 1874. ! 1875. 1876. Palermo. 31.81 19.90 22.05 25. 60 15.25 17.70 23.60 23.73 28.40 | 20.40 22. 70 South Trenton ... 38.47 36.54 39. 78 48. 73 35.20 34.60 31.67 32.64 29.56 1 38.87 33.67 Oneida 48. 92 51.26 41.47 52.60 . 35.33 I Average .... 39. 73 35.90 39.69 43.45 32.43 31.26 27.64 28.19 36.85 29.64 30.57 1 From Report New York State Engineer and Surveyor, 1896. 3 From data furnished by the water-supply division. DEEP WATERWAYS. 481 Table No. 16. — Record of precipitation during navigation periods on portions of Oneida and Mohawk watersheds, etc. —Continued. 1877. 1878. 1879. 1880. 1881. 1882. 1883. 1884. 1885. 1886. 1887. TTt,u*a. _ 25.67 26.02 28.32 23.37 38.07 34.75 23.07 Palermo. Oneida .. 21.35 34.65 28.60 18.10 23.37 15.71 13.79 22.90 13.78 22.65 20.63 17.33 Average_ 28.00 28.60 18.10 23.37 20.69 19.91 25.61 18.58 30.36 27.69 20.20 1888. 1889. 1890. 1891. 1892. 1893. 1894. 1 1895. 1896. 1897. 1898. Utica .. 30.96 33.27 44.85 21.17 39.97 Palermo .. 22.20 25.29 30.67 18.13 34.95 25.00 26. 74 18.81 20.41 23.03 24.05 Rome__ 48.36 34.11 30. 07 30. 29 22.89 25. lo 28 91 38.47 Phoenix.. 30.59 27.82 18.68 22. 53 28.00 30.23 Average .... 26.58 29.28 41.29 19.65 36.34 28.55 28.28 20.13 22.68 26.65 30.92 The maximum evaporation recorded during the navigation periods for the years 1892 to 1-896, inclusive, occurred in 1895, and amounted to 34.38 inches. This was an excessively dry year throughout the lake region, and the evaporation is not likely to be exceeded. The year of least precipitation during the navigation period, shown in Table No. 16, is 1879, and the amount for that year is 18.1 inches. Subtracting this minimum precipitation from the maximum evapora¬ tion gives 16.28 inches for the maximum depth of water that will be taken from the surface of the reservoir during any navigation season by evaporation. The area of the surface of the reservoir when full is 148.7 square miles. This area reduced to square feet and multiplied by the depth of evaporation in feet gives 5,624,086,000 cubic feet, the volume of water required for evaporation. It is assumed, by the direction of your Board, that a flow of 1,500 cubic feet per second throughout the navigation period will afford an abundant supply for lockage, and that of this amount 350 cubic feet per second can be supplied from the Salmon River reservoir. .V flow of 1,150 cubic feet per second remains to be provided. This flow continued throughout the navigation period, two hundred and forty- four days, amounts to 24,243,840,000 cubic feet, and when added to the volume required for evaporation makes a total of 29,867,926,000 cubic feet which must be obtained from the area naturally tributary to the summit level. It is desirable that a quantity of water sufficient for canal purposes shall be supplied, in addition to that now in use for water powers. In the event of the construction of the canal, it would be necessary, in order to accomplish this purpose, hat the amount of water now used for power be carefully measured and provision made for the continual flow of that amount both at Little Falls and in the Oswego River. If we except the three summer months, there is generally a H. Doc. 149-31 482 DEEP WATERWAYS. large volume of water wasted over the dams on both the Oswego and Mohawk rivers, and it is here assumed that the Seneca River alone will furnish water for power purposes on the Oswego during the stor¬ age period, and that West Canada ( reek alone will furnish water for powers at Little Falls during that period. The portion of the Oswego watershed not made tributary to the summit level has an area of about 3,654 square miles, and the portion of the Mohawk watershed above Little Falls not made tributary to the summit level, and which includes West Canada Creek, has an area of 646 square miles. The natural flow from the area that would drain into the summit level for the months not included in the storage period is not consid¬ ered a part of the supply available for canal purposes and may be used for water powers as at present. While it is impossible without further data to say that by this arrangement the various water powers will continue to receive this present supply of water, it seems reasonable to expect that they will, and such is assumed to be the case in this discussion. The portion of the Oneida watershed to be made tributary to the summit level has an area of 1,348 square miles, the portion of the Mohawk watershed to be made tributary to the summit level has an area of 660 square miles, and the sum of these areas, 2,008 square miles, is the total from which the volume 20,867,926,000 cubic feet must be obtained during the storage period. It is assumed that during the above period 60 per cent of the pre¬ cipitation will find its way into the reservoir. This assumption was made after a study of Table No. 17, in which the precipitation and run-off statistics of several streams whose records were available have been arranged to show the percentage of precipitation which appears in the discharge of the streams during the storage period of years of average and minimum precipitation. Table No. 17. —Rainfall and run-off data for storage period. Stream. Period of observa¬ tion. Area of water¬ shed in square miles. For mean precipitation dur¬ ing observation period. For year of minimum precipitation. Per¬ cent¬ age of precip¬ itation. Precip¬ itation in inches. Run¬ off- in inches. Per¬ cent¬ age of precip¬ itation. Year. Precip¬ itation in inches. Run¬ off in inches. Muskingum. 1888-1895 5,828.0 18.82 9.57 50.85 1895 13.04 4.04 30.98 Upper Genesee. 1894-1896 1,070.0 19.58 10.20 52.09 1895 13.20 5.63 42.65 Upper Hudson. 1891-1896 4,500.0 120.80 15.95 76.68 1895 2 15.79 11.68 73.97 Croton . 1870-1896 338.0 23.44 18.00 76.79 1872 14.57 11.59 2 79.55 Passaic. 1888-1893 822.0 25.49 21.03 82. 50 1892 22.26 14.44 64.87 Sudbury. 1875-1895 75.0 23.28 17.58 75.52 1883 16. 78 9.70 57.81 Mystic.. 1878-1895 26.9 22.41 15.08 67.29 1883 16.24 7.41 45.63 Cochituate. 1863-1895 18.9 23.08 14.89 64.51 1872 14.51 8.88 61.20 Note.— This table was computed from data collected by G. W. Rafter and published in the Reports of the New York State Engineer for 1895 and 1896 and in Water Supply and Irrigation Papers of the United States Geological Survey, No. 24. 1 Precipitation on “northern plateau.” 2 A percentage of run-off greater than the average in a year of minimum precipitation seems very improbable and is probably due to an error. DEEP WATERWAYS. 483 The year of minimum precipitation on the Upper Hudson is 1895. In that year the percentage of precipitation which appeared in the run-off at Mechanicville is 73.97. The conditions affecting precipita¬ tion and run-off which prevail upon the watersheds of the Oneida and Upper Mohawk rivers probably resemble those of the Hudson more closely than the conditions which prevail upon any of the other streams. Geographically the summit level is nearer to the Hudson than to any other stream of which the records are available, and the climatic conditions of the two localities are similar. The streams draining into the summit level have channels which are generally stony and precipitous, though perhaps they are not so torrential in character as the tributaries of the Hudson. The Hudson River record of run-off has been kept under favorable conditions and should be quite reliable, and it includes the year 1895, which is noted for its general dryness. On the other hand, the precipitation record used in computing the percentage of run-off for the Hudson is the record of the “northern plateau,” an area which includes observation stations north and west of the Upper Hudson basin and only one or two stations within that basin. Much of the Hudson basin lies in the heart of the Adiron- dacks, where no observations of precipitation have been made, and the record of the “northern plateau” has been used as affording the nearest approximation to the precipitation of the Hudson. It may therefore be surmised that since in 1895, the year of minimum pre¬ cipitation, the percentage of run-off of the Hudson was so great, it being 73.97, the actual precipitation on the Hudson was probably greater than that of the “northern plateau.” If such is the case, the effect would be to reduce the percentage of run-off, and this proba¬ bility was considered in deciding upon the percentage to be used in calculating the run-off from the Oneida-Mohawk watershed. If 60 per cent of the precipitation during the storage period is to afford a volume of 29,867,926,000 cubic feet of water, then the total volume of precipitation, or 100 per cent, must be 49,779,877,000 cubic feet, which is equal to a depth of 10.67 inches over 2,008 square miles. This depth exceeds the recorded precipitation during the storage period in but two 3 r ears out of the seventy-three years’ record (this record is given in full in Appendix No. 16. The observation stations considered are the same as those in Table No. 16), once in 1829, when the precipitation was 9.34 inches, and once in 1837, when it was 8.95 inches. The average precipitation during the storage periods for the seventy-three years was 19 inches, or 78 per cent more than the amount required. The record of precipitation is not wholly satisfactory. In several years we have the record for but a single station. The record at Oneida of a mean annual precipitation of 64.83 inches is abnormally large, and the difference between the records of Utica and Whitesboro sug- 484 DEEP WATERWAYS. gests a difference in methods of measurement. The latter places are on the south side of the Mohawk Valley, not over 3 miles apart, and the mean annual precipitation for five years—1834, 1835, 1836, 1839, and 1840—at Utica is 36.98 inches, while for the same years at Whitesboro it is 31.83 inches. Such a discrepancy is unaccountable. In this discussion evaporation from water surfaces during the months December to March, inclusive, has been neglected, and also the fact that the area of the proposed reservoir is about 7 per cent of the watershed, and upon its surface the precipitation will fall directly. These factors will, in a measure, balance each other, the net result being too small for consideration in a problem involving the use of such rough data. For the same reason no account has been made of the present Erie Canal reservoirs and feeders, which, though small, will nevertheless assist a little to increase the water supply. We have next to inquire concerning the capacity of the reservoir. Since April and May are, by reason of the large run-off of water during those months, made a part of the storage period, it would seem unnecessary for the storage supply to be greater than is needed for the months of June to November, inclusive. The consumption of water for the months of April and May will be, in a year of maximum evaporation, as per Rochester data: Cubic feet. Lockage.. .. .. 6,060.960, 000 Evaporation, 8.44 inches on 148.7 square miles. 2,915.681.000 Total...... ... 8,976,641,000 This volume is equivalent to a run off of 1.86 inches on 2.008 square miles, or a precipitation of 1.92 divided by 0.6 = 3.2 inches. Table No. 18.— Precipitation on Oneida and Upper Mohawk watersheds during the months of April and May for the t (reive years of the record, showing the least pre¬ cipitation for those months: 1879. 1891. 1873. 1828. 1881. 1847. Inches. . 2.05 . 3.02 3.08 . 3.17 . 3.32 .. 3.33 1872. 1896. 1877. 1887. 1884. 1837. Inches. . 3.62 . 3.65 . 3.80 . 3.88 . 4.11 . 4.14 Table No. 18 gives the recorded precipitation in April and Maj" for the twelve years of the record showing the least precipitation during those months. The necessary quantity, 3.2 inches, was lacking in but four years—1879, 1891, 1873, and 1828. In those years the precipi¬ tations during the months of April and May were respectively 2.05 inches, 3.02 inches, 3.08 inches, and 3.17 inches. The only year in which the deficiency is at all serious is in 1829, and during the storage period of that year the precipitation was abundant. It may there¬ fore be considered that, since the months of April and May constitute DEEP WATERWAYS. 485 about one-fourth of the navigation period, a storage volume equal to three-fourths of the total volume of water to be furnished by the Oneida and Mohawk watersheds will be all that is required. This total volume, the Salmon River supply not included, is 29,867,926,000 cubic feet, and three-fourths of it is 22,400,945,000 cubic feet. The capacity of the reservoir between elevations 379 and 386 is 25,302,400,000 cubic feet, a surplus of 2,901,455,000 cubic feet above what is needed. In addition to this, the water surface could be raised 1 or 2 feet above elevation 386 without doing any harm, since the locks are to be built to elevation 391 and the embankment at the weir in Oneida River to elevation 396. Because of the uncertainty of calculations based upon rainfall data it may be well to inquire what conditions would develop in the event of a year so dry that a sufficient supply of water failed to be collected during the storage period. It is seen at once that there is at the beginning of the storage period a surplus of water in the reservoir of 2,901,455,000 cubic feet, the amount by which its capacity exceeds the demand. This amount alone will furnish water for lockage and evap¬ oration at the usual rate for nineteen days. Referring to the Salmon River reservoir, there would probably be many years when the whole or a part of its waters would not be needed. In such a case during the succeeding storage period only a fraction of the drainage from the Salmon River basin would be needed to fill that reservoir and the remainder could be turned directly into Oneida Lake. If it were not needed in the lake, it would run over the weir in Oneida River and thence into Lake Ontario, but if it should happen that the Oneida reservoir was not tilled by the drainage naturally tributary to it, then this overflow from Salmon River might help to till it. It may be well to state that in calculating the volume of the Oneida storage reservoir no account was made of the ground water which would drain from the shores of the lake when the water surface is lowered. The area bordering the lake which it is proposed to submerge contains about 12 square miles of swamp land, the drainage from which, when the water surface is lowered, would be con¬ siderable. The occurrence in succession of two years of maximum dryness is a rare contingency, but because of its possibility it may be profitable to ask what would happen if, after using the reserve supply above described, there came another year of equal dryness. In that case there would be a deficiency in the ordinary supply and the reserve described above would be wholly wanting. The case would be unfor¬ tunate, of course, but it would not cause the suspension of navigation. Suppose a 30-foot channel has been constructed, then, when the water supply begins to fail, the available depth in the summit level will diminish; but the volume of water lying between elevations 379 and 377 is 6,244,760,000 cubic feet and would furnish water for evaporation 486 DEEP WATERWAYS. and lockage at the usual rate for forty-one days. With the water surface at an elevation of 377 there would still be a depth in the channel of 28 feet, and it would only be necessary for the boats to carry smaller cargoes until the great drought had passed. The foregoing discussion has not been carried to that degree of refinement that would be desirable with more exact data. The pre¬ cipitation records have not been kept at a sufficient number of sta¬ tions, and those we have exhibit some vagaries which raise suspicion as to their value. The percentage of run-off used in the computations may prove too high and the provision for the noninterference with water powers may not prove satisfactory. But the conditions discussed include a maximum length of navigation season, a maximum lockage, maximum evaporation, and a minimum rainfall. The average con¬ ditions would be more favorable, and in the extreme case Oneida Lake would continue to act as a reservoir no matter how great the demand upon it. If drawn down 8 feet below proposed low water, it would still have an area of about 80 square miles, and a month’s traffic would lower it but little. So, if it be assumed that the average water supply is sufficient, navigation need never be suspended, though at times the maximum depth of water may not be available in the summit level. This condition, is in striking contrast to that which would exist with a short high-summit level dependent upon a feeder supply. With such a plan, if the storage supply were exhausted or if the banks of the feeder or reservoir should fail, the water would at once be cut off from the canal and navigation wholly suspended. In conclusion, it maybe stated that notwithstanding the unsatisfac¬ tory character of the data used, the results indicate a strong proba¬ bility that a sufficient water supply may be obtained by this plan, and it would be well to verify them by a series of discharge measure¬ ments on the Oneida and Mohawk rivers. Such measurements, in connection with gauge records covering years of maximum and mini¬ mum discharge, would permit a proper and final design for the water supply. A gauge was established at Brewerton in January, 1898, and was read under the direction of the undersigned until June 1, 1899, and the record preserved. Mr. F. II. Newell, hydrographer in charge, United States Geological Survey, has since continued the readings, and each year will add to the value of the record. The effect of floods on Oneida Lake has been discussed under the high-level plan with an area of water surface of about 96 square miles. Since by the low-level plan the area would be 148.7 square miles, or about 55 per cent greater, the fluctuation due to freshets would be much less, and so small as to be of no especial interest. The advantages shown by the low-level plan may now be summed up as follows: DEEP WATERWAYS. 487 If at any time the supply of stored water should prove insufficient, navigation need not be suspended, since the Oneida reservoir could be drawn below the low-water elevation, and thus furnish a continuous supply of water at the expense of a slightly decreased depth of water in the summit level. The difficulties and uncertainties of maintaining a feeder about 90 miles long and extending for a considerable portion of its length across the drainage lines of the country would be avoided. There would be a saving of two locks with their attendant delays to navigation, and if the three locks at Fulton should be combined in two, three locks would be avoided. In either case, the reduction in total lift would vary between GO feet and 74 feet, according to the stage of the reservoir. The loss of water by leakage, though very small in either case, would be somewhat less, since between Oneida Lake and Frankfort the water surface would be lower than the ground surface, whereas with the high level there would be a few places where the water might seep through the canal banks into Wood Creek or the Mohawk River. DESCRIPTION OF SURVEYS. The survey was made to cover the route described in the 1896 Report of the United States Deep Waterways Commission, and also such minor variations as were deemed worthy of investigation. Besides the usual base line, stadia work, and test boring described in Appen¬ dix No. 9, the survey included triangulation of Oswego Harbor and Oneida Lake and soundings in Lake Ontario, Oswego River, Oneida River, Oneida Lake, Wood Creek, and Mohawk River. The survey was begun at Oswego, N. Y., in October, 1897. The force consisted at first of a base-line party, in charge of Mi-. E. E. Hart, and two stadia parties, in charge of Messrs. C. PL Curtis and PL A. Bagg, respectively. A sounding party and a level party began* work early in November, and one boring machine was started in Peter Scott’s swamp in the latter part of December. Mr. J. C. Hoyt was in charge of the soundings, Mr. E. Hilborn, jr., was selected to run the levels, and Mi-. A. W. Saunders was employed to superintend the borings. An effort was made in the beginning to keep the force together as closely as possible for the purpose of getting started right. The survey from the Oswego light-house to the first lock in the Oswego Canal was controlled by triangulation which was made by the base-line party. Upon its completion, the measurement of the base line along the river was begun. The stadia work was begun at the lake shore with a party on each side of the Oswego River. When the base-line party got well ahead of the stadia work, it was used, together with the level party, to locate sounding ranges. These 488 DEEP WATERWAYS. parties carried along the base line and levels, and located the sound¬ ings until December 27, when the running ice caused a suspension of sounding, and the base line was then pushed forward to Oneida Lake as rapidly as possible. The work upon Oneida Lake could be done best in the winter time, when its surface was frozen, and plans were therefore made to do it all, if possible, during the first winter. That would necessarily cause a break in the continuity of the survey, but the importance of com¬ pleting the lake work during the ice season made it the controlling feature of the survey. As it was impossible to foretell either the duration of the ice or the amount of work which might be necessary in order to find the best channel through the lake, it was deemed advisable to concentrate the whole force on the lake and push the work as rapidly as possible. For this purpose the stadia work on the Oswego River was suspended and the parties moved to Brewerton about the middle of January. The sounding and triangulation were both begun January 17. Mr. Curtis’s party had been assigned to the triangulation, Mr. Bagg’s party to the soundings, and the force was considerably increased by the addition of a number of laborers. Fp to that date very little time had been lost because of bad weather, and there was so much topography to be taken about Oneida Lake that a new stadia party was organized and set at work on the north shore of the lake January 17. The borings in Peter Scott’s swamp were completed about the mid¬ dle of January, and as soon as the sounding party was out of the way two boring machines were started on the lake. In order to find the deepest channel a line of soundings was made across the lake at Constantia, and another at North Bay. A transit line was then run from the outlet of the lake at Brewerton to the deepest part of the lake opposite Constantia, and from there to the deepest point in the lake opposite North Bay. This line was used as a base from which to lay out the soundings. A parallel line 300 feet south was also staked and soundings made along both lines at inter¬ vals of 100 feet. At intervals of 500 feet lines were staked perpendic¬ ular to the base and sounded far enough on either side to show tliat- the deepest water lay within the sounded area. A sounding reel and an ice auger, such as are described in Appen¬ dix E E E of the Report of the Chief of Engineers, United States Army, for 1895, were used in the work. The reel was found very convenient and rapid, but the auger proved no more economical than axes. The sounding was completed February 7 and the laborers discharged. The stadia party resumed its work at Brewerton. The base-line party, after crossing the lake, extended its line up the Wood Creek Valley to the high ground near New London. The valley is subject to frequent floods in the spring, which rise above tlie DEEP WATERWAYS. 489 banks of the creek, and it was intended to run the line far enough to permit its continuance at any time that might be desired. The party next returned to the lake and measured a base near North Bay and one near Brewerton for nse in the triangulation. The base line being then far in advance of the rest of the survey the party was assigned to stadia work in Sand Ridge, where the thick woods and brush made it desirable to do the work before the leaves came out in the spring. While the base-line party was crossing the lake the level line had been carried around the lake on the north, and, when the base line was suspended, the level party was broken up temporarily and its members divided among the other parties. The material in the bed of the lake proved to be mostly thin mud, with occasional bars of sand and gravel, and a little rock near the outlet. The boring was therefore very easy and made rapid progress. It was completed February 9, and the party moved back to Sand Ridge and worked in the small ponds and marshes in that vicinity while the ice lasted. In the spring the borings were completed between Phoenix and Brewerton and during the first week of June two boring machines were started at Sylvan Beach and one on the Oswego River at Phoenix. The latter was mounted on a small flatboat, which had an open well near the center through which the drill and casing could be worked. The triangulation of Oneida Lake was completed March 15 and the party resumed its stadia work on the Oswego River. A description of the triangulation is given under a separate heading. As soon as the lake work was completed the efforts of the corps were directed toward the completion of the work between Brewerton and Oswego. The soundings of the river were completed in a short time, and in the latter part of May the stadia work was completed and the two original parties moved to the south shore of Oneida Lake, while the base-line and level parties resumed their work near New London. There was no further interruption of the base-line work until its completion at Herkimer, July 11. On May 24 a reconnoissance was made by your Board of a route from Fulton westward toward Sod us. The route was found to be so crooked and the work so heav 3 T that no survey was made of it. On the same day a proposed line from Fulton southeasterly through Ingalls Cross¬ ing into Peter Scott’s Swamp was examined. A preliminary stadia line had already been run over the ground, and it appeared so favorable that its complete development was ordered. When the base line had been completed to Herkimer the parties engaged upon it were there¬ fore moved back to Fulton to run this line. Soundings and borings were made for a harbor at Oswego in July, one stadia party and a sounding party being sent from Wood Creek and a boring machine from Phoenix for that purpose. The soundings 490 DEEP WATERWAYS. were made from a steam launch running on ranges, and the soundings located by two transits on the shore. The borings were made from a small flatboat, the same that began work on the river at Phoenix early in June. On the completion of the harbor work the stadia and sounding par¬ ties returned to Wood Creek and the boring machine was moved back up the river. There was no further interruption of the sounding and stadia work and they were both completed to Herkimer early in November. A second boring machine was started near Fulton August 9, and the two machines continued their work along the Oswego and over the line from Fulton to Peter Scott’s Swamp until its completion in the middle of January, 1899. The boring party east of Oneida Lake found very hard material near New London, and the progress was so slow that the force was gradually increased until there were six machines at work. The hard material was passed a short distance west of Rome and the progress down the Mohawk Valley was more rapid. The borings were com¬ pleted during the first Aveek in January, 1899. The plotting of the survey was continued at Rome until March 1, 1899, Avhen the force was moved to the office of the board at Detroit. Since the method pursued in the mapping varies in some respects from that pursued by the other corps it is described with some detail under a separate heading. In the spring of 1899 a few additional borings were made at the site of a proposed lock near Fulton, and in July a complete development was made of a line from Minetto to Lake Ontario, running west of OsAA T ego. Table No. 19 indicates the size of the force and the progress of the work from its beginning in October, 1897, to the closing of the field office, February 28, 1899. The men engaged upon the survey have taken great interest in the Avork and are worthy of credit for their earnest efforts to accomplish the best possible results. 1 am particularly indebted for the active sympathy and cooperation of the folloA\ing men who had charge of instrument parties in the field: F. A. Bagg, A. E. Broenniman, C. E. Curtis, E. E. Hart, J. Hayes, E. Hilborn, jr., and W. A. Miller. A. W. Sanders Avas superintendent of borings. W. J. Bergen, A. Haring, and II. II. Ross assisted in the Detroit office and deserve mention. DEEP WATERWAYS. 491 Table No. 19 .—Approximate statement of monthly progress of field work. Month. Transit line, distance run in miles. Levels, dis¬ tance run in miles. Triangulations, n u m - ber of angles ob¬ served. Topography. Soundings. Borings. Number of names on pay roll. Base line. Auxiliary line. a3 fl 0) 00 2 CQ Auxiliary line. Square miles surveyed. Days lost on account of storms. Miles of chan¬ nel sounded. 1 Area sounded in square miles. 2 N u in ber of holes. Total depth in feet. 3 Surveys. Borings. 1897. Oct_ 6.01 5.54 0.4 168 . 83 go Nov_ 5.34 8.33 16.38 12.9 3.5 7.06 28 Dec .... 14.40 8.34 32.01 13.0 . . 4.5 7.33 . 3 81 31 12 1898. Jan __ 30.34 69.58 4.6 t) 1.0 33 1,302 52 15 Feb .... 3.91 7.63 36 4.4 2.98 113 3,398 48 18 Mar 5.31 5.1 39 1.4 7.79 15 1 287 38 21 Apr . . 2 8 45 20 May.. 9.U3 15.18 4. 7 34 1.241 as 25 June . 31.35 47.97 1.2 23. 73 70 3,838 41 23 July 17.03 24.46 9. 46 46 1.435 44 28 Aug_ 3.2 09 cm 6t> 2,798 45 45 Sept. -. 1.5 3.03 54 3.166 47 Oct. 3.6 3.60 109 6,045 44 47 Nov_ \ 2 97 4.978 39 53 Dec . 2,931 32 48 1899. Jan_ 10 464 4 22 18 Feb ... 4 22 Total 106.89 32.5 308.65 lO N CO _ 249 132.30 cc Cn 88.77 15.80 750 33,711 1 Includes all other soundings. 3 Depth of water not included. 2 Includes Lakes Ontario, Onedia, and Pleasant. 4 Drafting. Note.— The small amount of field work done in the spring and summer of 1899 is not included. The time lost by storms while taking topography is given as the best available data of that character. It is the average number of days lost by the several stadia parties at work. TRI ANGULATION. The Oswego Harbor was covered by a net of simple triangles. The work was done by the base-line party, Mr. E. E. Hart in charge. The base was 461.24 feet long. The number of triangles was 47 and the longest side 1,435.3 feet. The area covered was yW square mile. The triangles were all small and no especial care was necessary to insure satisfactory results. Five sides were both measured and com¬ puted, and the results are as follows: Side. Measured length. Computed length. Discrep¬ ancy. A B... 564.59 564.58 +0.01 B C. 915.52 915. 68 — .16 H, I . ... 1,309.57 1,309.49 + .08 R, S . . 150. 77 150. 76 + .01 a g .. 491.43 491.43 .00 The only difficulty encountered was in dodging lumber piles and buildings. The triangulation controls the survey from the outer light-house to the first base-line station near Oswego Canal lock No. IS. 492 DEEP WATERWAYS. The triangulation of Oneida Lake was performed by Mr. C. E. Cur¬ tis and the members of his stadia party, aided, during the latter part of the work, by Mr. E. Hilborn, jr., and an additional rod man. The work consisted of the preliminary reconnoissance, selection and marking of stations, measuring base lines, and observing the angles. The adjustment of the angles and the computations were performed later in the office at such times as were convenient. Two Buff & Berger 64-inch transits were used on the work. The verniers of both instruments read to 30 inches. The one with which the work was begun had been in use about three months. It had an inverting telescope. When Mr. Hilborn was assigned to the work, the second transit was purchased new. It was just like the one already in use, save that it had an erecting telescope. The object of the triangulation was to extend the base line of the survey across Oneida Lake and to establish points along its shores with which to control the topography. The greater part of the lake had been mapped by the United States Geological Survey, the map of which was very useful in locating the various stations. The general scheme adopted was a chain of six quadrilaterals and a single triangle, the triangle being at the western end of the lake. The western vertex of the triangle was called A and the stations along the shores were lettered consecutively B, C, D, E, F, G, and H, the subscript n and s being used to indicate whether the station was on the north or south shore. The lines B„ B s and G s and H s were selected as bases, and their lengths measured on the ice. There were also three stations, l s , 2 n , 3„, located outside of the main system, and in connection with other stations they formed two additional triangles. The stations were first marked on the geological map, care being taken that no angle should scale less than 30°. Flags were then set up around the lake at the various stations and observed with a lield glass to see if the adjacent stations were intervisible, and the angles measured roughly with a pocket compass. Many of the stations had to be shifted several times before the desired conditions could be obtained. The map was not wholly accurate, and it was found that trees and buildings prohibited the use of some of the lines first planned. The line D s , E s had to be cleared for about 1,000 feet through a piece of thick woods. It was found that two angles were slightly less than 30°, and since they could not be increased without great difficulty they were allowed to stand. Most of the stations were located at the edge of the water in order to avoid trees. Two stations, F n and G n , were on quite high ground, but the view from each was obstructed by trees and buildings, and two wooden towers had to be erected. When it was found that F„ and G n could not be used without the towers, other plans were st udied. The only alternate locations avail- DEEP WATERWAYS. 493 able were such that if adopted they would make several of the angles considerably less than 30°, and they could not be used at all without a further reconhoissance. It seemed to be a question whether it were better to incur a further expense and delay for reconnoissance and accept angles of a size which were too small for the best results or to build the towers. The cost of the two towers was $106.40, and the expense for a further reconnoissance and setting the signals would have been not less than $25, so it is possible that $81.40 was paid for the sake of having the angles greater instead of less than 30°. The signals used were pine boards, 1 inch by 12 inches by 8 feet long. The edges of the boards were chamfered, so that they could not be seen when a signal was being observed. On the face of the board were painted a series of black 12-inch squares, with their diag¬ onals in the axis of the board. The body of the signal was painted white, so that in pointing the telescope the vertical wire would bisect the series of angles made by the squares. A tapered oaken plug was bolted to the foot of each signal. The axis of the taper plug was in line with the axis of the signal. The station was marked by a cedar post set in the ground, with its top just below the surface. A tapered hole was bored in the post to receive the signal. When the signal was thus set up, it was securely guyed to avoid motion or injury by the wind. When the signals were not in use, the holes in the posts were stopped by oak plugs made for the purpose, and well greased with tallow to keep out the water. No trouble was experienced at any time in removing the plugs and setting the signals. These signals were used on sides that were 5 miles long, and were plainly seen on clear days, but there was much hazy weather and light snow, which made the seeing bad, so it was finally decided to work at night. Six tubular lanterns, with parabolic reflectors and 1-inch flat wicks, were procured and mounted on stakes which set in the station posts just as the signal boards did. The center of the flame was plumbed over the center of the hole. These signals worked very well, and could be plainly seen 64 miles, which was the length of the longest line. The towers were each 36 feet high, and were built of rough hemlock fastened together with nails. Each tower consisted of a tripod to support the instrument and a staging for the observer and recorder. The staging had four posts placed at the corners of a 12-foot square. With the legs of the tripod there were in all 7 posts. Each post was built up of 2 by 4s nailed together so as to break joints. The posts were braced with boards, 1 by 6 inches, nailed horizontally and diag¬ onally in panels, excepting that the bottom horizontal braces of the tripod were 2 by 4s. The staging was wholly independent of the tripod, so that no jar occasioned by the wind or moving on the stag¬ ing could affect the instrument. 494 DEEP WATERWAYS. A box was built up in the center of the tower to protect the plumb bob string from the wind. The box was supported entirely by the staging. At one corner of the staging a ladder was built to give access to the platform at the top. The legs of the tripod were set into the ground about 18 inches, and rested on large flat stones. A platform was built on the bottom of the tripod and weighted with several wagonloads of stone. Angles were read successfully from the towers when the wind blew so hard that the transit box was blown to the ground, and the motion of the tower was never so great that the motion of the movable head of the transit was not sufficient to center the plumb bob. The board on which the transit rested, when in the box, was fastened to the tripod with wood screws, and a hole cut in the center for the plumb bob. ’J'he station posts were set in the ground after the towers were built so as to avoid any difficulty in centering the tower. The different stations were all referenced before reading the angles, so that if disturbed they could be relocated and also that they might be readily found later by the stadia parties. An azimuth was also measured to some local object or stake, so that the stadia parties could readily check on it and not be obliged to depend on a sight to another triangulation station. An azimuth observation was made at A and another at H„, for which the azimuths of the triangulation lines were computed, making due allowance for convergence. The bases B s B n and G s H a were measured by the base-line party. The conditions were favorable for good measurements. B s and B n were close to the lake at about the elevation of the water surface. G s was similarly located, but H s was about 20 feet from the shore and 3 feet higher. The weather was mild and cloudy. A path was shov¬ eled, in each case, along the ice, and the tape rested in snow water; so its temperature was practically constant. The tape lengths were marked with a knife blade on the top of little pine pegs driven in the ice, holes being bored with an auger for that purpose. Each base was measured twice in opposite directions and a new set of pegs used each time. A tension of 12 pounds was applied each time to the tape, and the pegs were lined with a transit. After correcting for tape error and temperature, the measured lengths of G S H S differed by T VoV The mean corrected length was 11,521.215, and the proportional errorwas 1 in 101,000. For B s B n the measured lengths differed by yf The mean corrected length was 6,429.480, and the proportional error was 1 in 213,300. At each corner of a quadrilateral there were measured the angle formed by its sides and the two angles formed by the sides and the diag¬ onal. There were, therefore, 12 angles measured in each quadrilateral, DEEP WATERWAYS. 495 and since there were 6 quadrilaterals and 3 triangles, 2 of them small and of secondary importance, there were, in all, 81 angles. Each angle in the quadrilaterals was repeated twenty-four times, and if for any reason a set of readings was interrupted before it was complete an entire new set of twenty-four was made. Two methods of pointing were used, the first being as follows: Let A be the signal to the left and B the signal to the right; then with the telescope direct point on A, clamp below, read, loosen above, point on B, clamp above, read, loosen below. Perform the same operation six times, save that the reading is not to be made again until after the sixth pointing on B. Then without disturbing the reading, reverse the telescope and make six pointings on A and B as before, save that no reading is to be made after pointing on A nor until after the sixth pointing on B. Next loosen the plates and shift them so that the readings will come on a different part of the limb. Then make twelve pointings as before, save that when the upper plate is loose, the motion should be from B to A. The operation is now, point on B, clamp below, read, loosen above, point on A, clamp above, read, loosen below, point on B, and so on. Do not read again until after the sixth pointing on A and again after the twelfth pointing on A. Both verniers were always used when making a reading. In this method, while a set of twelve repetitions are being made, the motion of the upper plate on the lower is always in the same direction, and any slipping has a cumulative effect. It was found that this method gave angles a trifle too small; so the second method was adopted. It differs from the first only in that the second six repeti¬ tions are made from B to A instead of from A to B. Likewise the third and fourth sets of six are made in opposite directions. This method eliminated the errors due to slipping and produced better results. The first method was used in triangle A B„ B* and the two quadrilaterals next east and gave values a little too small. The second method was used in the remaining quadrilaterals and gave half of the values too small and half of them too large, the mean being more nearly correct than in the former case. The quadrilaterals and triangles all closed within 10 seconds, save B n B s C n C s in which the error of closure was 15.8 seconds. It was not considered worth while to try to reduce the error, as its effect would occasion no practical inconvenience or injury to the work. When it became necessary to increase the force, the work was pros¬ ecuted by one party at night and by the other in the daytime. The night party occupied the stations having the longest sights and was able to see the signals without difficulty. The delays due to wind, snow, and mist were about the same as in the daytime, the advantage being wholly in the fact that the night signals could be seen a greater distance. 496 DEEP WATERWAYS. I) n was located on a small island which was so low that the ice cov¬ ered it. There was some motion of the ice during the triangulation; so to be sure of the position of the station the angles in the triangle D n 2 n 3 n were measured thirteen times, while 1) was in use. Fortunately no motion occurred. The quadrilaterals were adjusted according to the “Rigorous method,” given on page 514 of Johnson’s Theory and Practice of Sur¬ veying, thirteenth edition. The spherical excess, being less than 1 second, was not considered. MAPPING. The survey has been mapped on mounted sheets, the size of which, inside the working limit, is 25 by 37 inches. East of Sand Ridge the survey has been plotted to a scale of 1 in 5,000. West of and includ¬ ing a portion of Sand Ridge the surface of the country is more irreg¬ ular, and a larger scale was necessary to clearly show the country; so it was mostly plotted on a scale of 1 in 2,500. The base line, stadia circuits, and triangulation have been plotted by latitudes and departures and the courses all scaled after plotting. The work was simplified and the chances for error lessened by mak¬ ing the sides of the sheets parallel to the meridians and the top of the maps toward the north. This plan also permitted a sheet to be laid out at any time on any part of the survey, because the latitude and departure of its working limits could generally be fixed independently of the other sheets. Contours have been drawn having a vertical interval of 2 feet, save in some few cases where it seemed improbable that the country might be affected by any canal location. In those cases a 10-foot interval was used. The conventions furnished by the Board have been followed throughout the work, but while they were used sufficiently to show clearly the topography of the country, care has been exercised to avoid everything in that line not wholly necessary, to the end that the maps might be a useful medium for an engineering study rather than an exhibition of artistic skill. There has been some shrinkage of the paper, which amounted, on a few sheets, to one-tenth of an inch in each direction, but generally it has been much less. It has occurred generally after the plotting has been done, and scalings from the maps may be corrected so as to eliminate its effect. The method of laying out the sheets prevents any cumulative errors due to this cause. The process of mapping is naturally separated into two principal divisions, pencil work and pen work. The former includes plotting the base line, triangulations, and stadia circuits, plotting the topogra¬ phy and sketching contours. The latter includes inking the figures, making the topographical conventions, drawing contours and right DEEP WATERWAYS. 497 lines, and lettering-. In addition to these there is checking the work and cleaning the sheets. In doing the work the men assigned to each of these subdivisions have been continued long enough on one class of work to acquire con¬ siderable skill, so that a good degree of progress could be made. For instance, in sketching contours a continued improvement in skill and rapidity was observed for two or three months after beginning. Some parts of the mapping, such as sketching contours, lettering, and making the conventional signs, required especial skill and natural ability, and this arrangement permitted the execution of all such work by the men who were best fitted for it. It was sometimes necessary to shift the men from one part to another, in order to keep the work moving forward in an orderly manner, and in that way the monotony of the work was varied somewhat. This method of mapping has several advantages which were well suited to the situation. In the first place, it permits the use of a num¬ ber of engineers who are not skilled draftsmen, and many of the men who had been employed on the survey could be retained for work on the maps. Moreover, a sufficient number of skilled draftsmen to do the whole work was hard to find, and had they been found would have required larger salaries. By keeping a man on one piece of work for several weeks, as, for example, plotting the base line and stadia cir¬ cuits, he becomes thoroughly familiar with it and is much less liable to make errors than where he only plots the lines on a single sheet at a time and then takes up another part of the work. There is a natural tendency among men to interpret notes according to their memory, and while it is true that some men have excellent memories, it was desired to secure a set of notes which were not dependent upon such assistance and to know that the plotting was done in strict conformity thereto. With this method the work was so divided and distributed that no one had a chance to make any use of such familiarity with the topog¬ raphy as he may have gained in the field, and if the notes were not clear the fact was quickly discovered and steps taken to remedy the fault and guard against its recurrence. A memorandum book for each of the stadia parties was kept in the office, and in it were recorded all questions or doubts which arose concerning the notes taken by that party. During the progress of the survey, when all the notes in one field book had been plotted, these memoranda, together with the field book, were returned to the party in the field which had taken the notes, and the proper corrections and additions made. By this means the stadia parties were continually posted regarding their weak points and the quality of their work improved. The plan was so successful that during the latter half of the survey it became very unusual to find in the notes any important defects. H. Doc. 149-32 498 DEEP WATERWAYS. This method also secured a degree of uniformity in keeping the notes and making computations that could have been attained in no other way. The work, in fact, was reduced to a system and was, therefore, at all times under complete control. The principal errors guarded against in the mapping were such as might occur in plotting the stadia readings and in sketching the contours. In order to eliminate these, after the contours had been sketched the maps were carefully examined by men who gave nearly their whole time to that work, and all irregular or unusual features, such as crooked property lines or peculiar contours, were investigated and sometimes replotted. All boring notes were plotted twice, as their importance was much greater than that of single contour points and it was imperative that they should be correct. The maps were checked immediately after the contours were sketched in pencil, and then throughout the process of inking a close watch was kept for errors and omissions. Another part of the office work consisted of figuring the latitudes and departures and adjusting the circuits, checking the reduction of stadia readings, and extending the elevations of contour points. Much of the computing that had been done in the field was refig¬ ured in the office. It was found that checking which had been done in the field by the parties who took the notes was less reliable than that done in the office. This was probably because the members of the field parties were unable to do as good work in the evening after an active day in the field as were the office men who did no other work, and also because they were so familiar with the notes that any error was more readily passed over and repeated than where the checking was done by men who were using the notes for the first time. It has therefore been a general rule on the survey that all checking of numerical work, as well as drafting, should be done by parties who had nothing to do with the original work. In the beginning of the survey the field work was checked by hav¬ ing each party plot its own work roughly on protractor sheets. At that time there were only two stadia parties at work. They were located very near together, and the checking was fairly well done. It was found, however, that to do the work thoroughly would require the addition of a draftsman and a computer to each party. Even with this change it would have been difficult to keep the checking close up to date, because the book which was in the field during the day was always needed at night for reducing the readings and com¬ pleting the notes, so the draftsman either had to be a whole book behind in his work or else the field party must work first in one book and then in another, either of which methods was objectionable. After the survey had been in progress fora time it was decided that since working maps were needed they could as well be made by the field parties instead of the plotting of the protractor sheets, thus sav- DEEP WATERWAYS. 499 ing some labor and securing the necessary check on the survey and a complete set of working maps at the same time. It was desirable to plot the base line on these working maps by lati¬ tudes and departures, and, as they must be figured, it became con¬ venient to check the stadia circuits by figuring their latitudes and departures and comparing results with those of the base line at the closing station. The method pursued thereafter was, therefore, to check the stadia circuits by latitudes and departures immediately when closed; to send a book, when completed, to the headquarters of the survey, where the numerical work contained therein was first checked and the notes then plotted and errors or omissions reported back to the field for correction. The work was so divided among the stadia parties that each party could remain long enough in one locality to fill several books, and it was very seldom that a party had to go back to a pre¬ vious location to look up errors. After the winter work on Oneida Lake was completed the force was so large, numbering, as it did for several months, nearly 100 men, and the work was so scattered that it would have been impossible to maintain a thorough control of it if the mapping had not been cen¬ tralized and reduced to a system as above described. A record has been kept of the amount of time spent on the different portions of the drafting, from which lias been prepared the following statement showing the percentage of the total labor expended on each part: Per cent. Plotting base line, stadia circuits, and triangulation. ... 5.4 Plotting topography ... .. 39. 9 Sketching contours . 12.1 Checking work. ... 11.3 Inking elevations ....... 6.1 Inking right lines .-.... 5.4 Inki ng con tours ... 4.4 Inking names of property owners ... 5.1 Inking conventions, station numbers, and contour elevations _ 3.5 Lettering . 3.0 Cleaning . 3.2 Total........ 100.0 These percentages are based upon the actual t ime spent upon t he work, and have nothing to do with its cost, since the salaries of the men were varied according to skill, experience, and the position which had been held on the survey. Respectfully submitted. Albert J. Himes, Ass istant Engineer. The Board of Engineers on Deep Waterways. 500 DEEP WATERWAYS. Appendix No. 14. OSWEGO-MOHAWK ROUTE, EASTERN DIVISION. Detroit, Mich., October 17 , 1899. Gentlemen: 1 have the honor to respectfully submit the following report on the Oswego-Mohawk route, generally known as the Oswego route, eastern division, from Herkimer, N. Y., to the Hudson River at Troy, N. Y. INSTRUCTIONS. Reference is hereto made to Appendix No. 9, “Instructions for sur¬ vey parties,” as issued the United States Board of Engineers on Deep Waterways, as governing in general the methods of work pur¬ sued. As the result of experience obtained as the work advanced, I found certain modifications of detail connected with these instruc¬ tions advisable on this division, as better adapting themselves to the conditions existing and the work in hand; these, with the methods pursued and results obtained, are herewith referred to in this report. organization. The field work of this division was actively b gun at the Hudson River end of the Oswego route in October, 1897, the first one-half of the month being occupied in the organization of parties and general preparations incident to the beginning of the work. A level party began work on October 16, a transit party on October 19, one stadia party on October 21, and another on October 28. On December 1, 1897, it was deemed to the best interest of the work to combine the transit and level work under one party, and during the remainder of the survey this combined work was done by the former transit party. On December 1, 1897, a sounding party was organized. During the first part of December, 1897, the necessary outfit and plant for a boring party was collected, and one party began work on December 21, a second one on January 27, 1898, and a third one about March 1. This force of three boring parties was continued until the latter part of July, 1898, when two additional parties were added, and on September 19, 1898, a sixth party was organized. This force was continued until the completion of the borings to Herkimer and from Herkimer about 4 miles northwest to Frankfort, on the Oswego route, western division, to which point they were completed on November 18, 1898. METHODS OF WORK, WITH RESULTS OBTAINED. Transit work .—This party was engaged on days unfit for field work in the reduction of notes and computation of coordinates of transit DEEP WATERWAYS. 501 stations, which were sent in to the office of the assistant engineer as fast as completed. As a general check on the results of the transit work for long dis¬ tances connection was made between this work at station 0.0 at the Congress Street Bridge and a New York State triangulation station there. Connection was also made between the transit line and the following New York State triangulation stations along the line by intersections on the same: At Amsterdam, Roman Catholic Church spire; at Canajoharie, Dutch Reformed Church; at Little Falls, Meth¬ odist Episcopal Church; at Herkimer, Methodist Episcopal Church. The coordinates of transit stations were carried continuously through the work, corrections being applied as authorized in “Instructions for survey parties.” The “ running” error between observations taken ranged from 00' 35” to 06' 20”, and the distances between observations taken from 5.7 to 9.6 miles. 1 A transit line traverse along both banks of the Hud¬ son River from Troy to Waterford showed a “running” error in azimuth of 01' 30” and an error in latitude of 0.15 feet and in depar¬ ture of 0.59 feet. The general direction of this traverse was north and south. Another transit line traverse near here of about 34 miles showed an error in azimuth of 02' and in latitude of 3.75 feet and in departure of 3.82 feet. The following table shows the comparison between the transit work and that of the New York State triangulation. The latter is assumed to be correct, but it must be remembered, in reviewing this compari¬ son, that the coordinates of these triangulation stations are given to the third decimal place in seconds; also that the value of 1" of lati¬ tude, as measured on the meridional arc at 42° of latitude, is about 101 feet, and 1” of longitude, as measured along the parallel at 42° of latitude, is about 75 feet. Place. Difference in lati tude (New York State survey). Difference in longi¬ tude (New York State survey). Error in transit line (lati¬ tude-long- gitude). Propor¬ tion of From— To- error to distance. Congress Street Bridge. Amsterdam. Feet. 24 Feet. 29 Feet. 38 5832 Do. Herkimer. 40 00 40 The proportion of error to distance is obtained by finding the hypothenuse of a right-angled triangle from the latitude and longitude as base and height, respectively, divided into the total distance and expressed as a fraction. ‘The average running error for 14 observations was: “Running” error 02 30 , distance 6.3 miles. 502 DEEP WATERWAYS. Level work .—For the first 20 miles of this work from the Hudson River end Philadelphia level rods were used. For the remainder of the work, two rods specially made to order were used. These rods were made of white’pine, paraffined in about one-fourth of an inch, in one length of 10 feet 7f inches over all, read¬ ing to 10 feet. They were graduated on the face to feet, tenths, and hundredths, with vernier reading to thousandths. On the back of each rod a round folding plumbing level was attached. The target was' square, 4 inches high by 54 inches wide, with black enameled face, with a vertical white enamel strip one-fourth inch wide through the center, and two narrow, horizontal white enamel strips, each tapering from the outside edges of the target to the cen¬ ters, these white strips being one-fourth inch at outside and about one-sixteenth at center. The target was moved up and down by an endless steel tape one- fourth inch wide, attached to top and bottom of target with screws to regulate tension of the same and passing over pulleys set flush in the rod near the top and bottom. The target vernier was arranged so that it could be pressed down fiat on the face of the rod by a spring, enabling more accurate reading of the vernier. The accompanying sketch shows a front view and cross section of these rods, also the steel pins used for turning points. These were driven firmly in the ground by a wooden mall. The advantages of the rods and turning points used are, briefly stated, less liability of the rod to change of length through influences of moisture, no risk of rod slipping as in use of sliding rod, exact graduation of the same, no risk from rod being held at different eleva¬ tion for fore and back sight, and the target arrangement enabling a more exact setting and reading of the same. All of which are essential points for obtaining good results. Duplicate levels were run in opposite directions and the instrument shaded by an umbrella from the rays of the sun during hot weather. Immediately after the back sight to obtain height of instrument, the forward sight was taken on the next turning point, thus minimizing risk of instrument settling. The limit of error between duplicate level lines per “ Instructions for survey parties,” as issued by the United States Board of Engi¬ neers on Deep Waterways, was 0.05 \/distance in miles. The 0.05 in this formula will be referred to as C in the following remarks. The elevation of 68 bench marks were determined between Congress Street Bridge at Troy, and Herkimer, the distance by the transit line being about 88 miles. The mean elevation of the two lines, run in opposite directions, was used as the adopted elevation of the bench mark. The maximum value obtained of C for lines between consecutive board measurements was 0.035 for 1.66 miles; the minimum, 0.001 for 0.63 miles, and the average, 0.0115 for 88 miles. _ Level Rod_ Front View Side View JULIUS BIEN & CO PHOTO LITH H Doc 149 56 2 DEEP WATERWAYS. 503 The maximum value of C for liues from the origin was 0.022 for 18.29 miles; minimum, 0.001 for 5.86 miles; average, 0.009 for 88 miles. The following values of C were obtained at distances stated from origin. These values being selected at random: 0.018 at 20.81 miles, 0.003 at 39.53 miles, 0.004 at 60.96 miles, 0.002 at 87.86 miles. Buff & Berger level No. 2652 was used ou the work. It is but just to say this liue of level was run for the most part over level stretches, following the transit line along railroads, roads, and the Erie Canal towpath, the total difference in level between the origin and the end of line at Herkimer being about 367 feet. Stadia work .— The best and most satisfactory results were obtained on this division under the following organization of a stadia party: One in charge to sketch and direct party, one observer, one recorder, four rodmen or stadiamen. There are certain advocates of the plan that with a liberal amount of contour points taken and location by shots made of existing fea¬ tures, natural and artificial, the time taken in making sketches in the field is largely time wasted. This supposition, I believe, in actual practice will be found to be theoretically correct only. My experience convinces me that an absolutely indispensable requi¬ site to a correct representation of the topographic features of a section, when reduced to a drawing, is a full, clear, and complete sketch, not necessarily to scale, of the relative positions of the several points located and the topography intervening, accompanying the notes and made on the ground while the survey is in progress. This statement I mean to apply regardless of the number of shots taken or locations made. Whether the notes are to be platted by the party or parties taking the same, which I believe to be preferable, or by other parties unfamiliar with the locality, the sketch is of inestimable value to a correct interpretation of the notes, however clear they may be. In actual practice I have found that two intelligent persons engaged on the same survey will, with all the points taken in the field platted on a map, in the absence of sketches, misrepresent, each to a marked degree of difference, the facts as they exist. A sketch, however crude it may be, is a transmission to paper of the impression of the ground and its features as made on the mind of the observer with the features spread out before him, and is of more value than anything else in recalling these features and correctly drawing them on the maps at some future time. It also acts as a check on the accuracy of the readings for position and elevation. It enables not only those engaged on the survey, but others not familiar with the same, whose duties it may be to plat the notes, to more quickly and accurately draw the maps. While the governing conditions may not always be such that the notes taken in the field can be platted and drawn by the parties tak¬ ing the same, yet the nearest approach to this, and the sooner they 504 DEEP WATERWAYS. can be drawn after being taken in the field, is believed to be the best practice. The above observations are indulged in as influencing the methods of work as pursued on this division, and also for what use they may be to the engineering profession, and it is this last consideration that may prompt a brief discussion of certain methods, although not wholly pertinent to this report. In the above organization of a stadia party, the man in charge directed the work and made all sketches of the area being surveyed, being at liberty to follow the rodmen where he deemed it best to more accurately sketch at close range. He identified this sketch with the recorder’s notes by similar numbers for shots taken, checking up these numbers with him at intervals. The observer ran the instrument and took all observations, having a code of signals for each rodman. The recorder recorded all readings, noting, under remarks, the character of the shot taken—whether a contour point, a fence corner, stream, ditch, river bank, or corner of a building, etc., abbreviations for these being used and an index of the same recorded in the note¬ books. Jlist here I would state that no recorder, however quick he may be, can with a rapid observer do more than actually record for two or three good rodmen or stadiamen, and at times to do even this he is crowded for time, and his attention is wholly absorbed therewith, to the exclu¬ sion of any time to make sketches. He is compelled to remain at the instrument, and, even if time were available, sketches made by him at long range would be useless, as probably misleading instead of aiding. Three rodmen or stadiamen were generally employed in the field, and the fourth one when occasion required. These men briefly noted, when necessary, either in a book or on a slip of paper, on the back of their rods the character of the location held at, or, if covering a section of less importance as to detail or inaccessible to the man sketching, they made a sketch connecting the points taken. The fourth rodman was employed the greater portion of the time in the office where the temporary headquarters of the party were located. The map sheets were furnished the stadia parties from the office of the assistant engineer, with the coordinates laid off and the base line platted from these coordinates on the same. The stadia parties were engaged, on days when the weather pre¬ vented fieldwork, in the reduction of notes and platting all shots taken in pencil and drawing in all natural and artificial features, the rodman in the office prosecuting this work alone when the party was in the field. In this way, all office work was kept fairly close up with the fieldwork, and any corrections necessary or additional data needed could be at once supplied. As soon as these sheets were completed they were sent in to the office of the assistant engineer. DEEP WATERWAYS. 505 There are those who, generally speaking, have never used, to any great extent, improved stadia methods, who seriously question the accuracy of the same to the extent of its usefulness for surveys of this nature and general preliminary surveys. It is believed a thorough appreciation of the excellent results attain¬ able with properly graduated rods and ordinary refined methods would lead to its more general adoption, and enable preliminary surveys and examinations for contemplated works to be made at a large sav¬ ing of time and expense. It is with a view to this that the following results of stadia work, as found on this division, are given. In the early stages of this work, when the organization was being perfected and the parties trained to the adoption of the best methods for the prevention and elimination of errors, more likely to occur at the beginning of surveys of such an extended nature, the transit line as measured by steel tape and leveled with Y-level, was retraced with a stadia line, each transit station being occupied and a forward and back reading being taken in each case for distance and vertical angle, the mean of the two being taken as the correct result. These distances and elevations, as obtained by stadia, were carried continuously over the distance given, and offer an interesting and instructive lesson, by direct comparison, as to the excellence attain¬ able by stadia methods, compared with careful transit work with steel tape and Y-levels. About one-half of this line was along the tow- path of the Erie Canal, and the other half across country. The work was done in the month of December. I do not believe quite as favor¬ able results could be depended upon if the weather was warm and the air “boiling.” The table of comparison follows: Table No. 1. — Comparison of stadia and base line work deduced from the reloca¬ tion by stadia of the base line for a distance of 5.6 miles. Station as per transit line. Station as per stadia line. Error in distance of st adia line (+ or —). Azimuth by transit line. Azimuth by stadia line. Elevation by level. Elevation by stadia. Error of stadia levels (+ or —1. Feet. Feet. Feet. O / » O / * Feet. 'Feet. Feet. 162+37. Sit 165 +56.68 165+56.7 98 16 00 98 16 00 95.10 95.10 168+87.92 168 +89.1 + 1.2 108 34 GO 108 35 15 98. 42 98.38 — 0.04 173 +26.60 173+26.3 — 0.3 167 10 (XI 167 11 15 90.32 90.31 — 0.01 182+37.06 182 + 36.3 — 0.8 158 39 30 158 40 15 104.95 104.99 + 0.04 185 +77.10 185 + 76.2 — 0.9 128 12 (K 128 12 00 123.86 123. 87 + 0.01 189+53.40 189+51.7 — 1.7 155 52 3<-/ 155 52 15 136.52 136.60 + 0.08 200+29.18 200 +25.6 — 3.6 140 50 30 140 50 CXI 142. 70 142.84 + 14 207 + 88.34 207+84.9 — 3.4 139 ii (XI 139 10 30 165.41 165. Ii4 4- 0.23 211+64.40 211+60.1 — 4.3 123 12 30 12-3 12 00 155.63 155.84 + 0. 21 217+74. 67 217+70.7 — 4.0 150 03 30 150 03 30 154.73 154.82 + 0.09 224 +08.30 224+03.2 — 5.1 153 43 30 153 43 30 164.44 164. 52 + 0.08 224 + 80.99 224 + 75.7 — 5.3 1 142 13 (X) 142 09 45 226 -f*66.60 226 + 62.8 — 3.8 2 186 58 1X1 186 56 00 154.56 154.57 + 0.01 230+17.29 230+13.8 — 3 5 174 47 30 174 46 30 154.37 154.32 — 0.05 236+52.67 236 + 47. 4 — 5.3 161 46 30 161 45 15 154.86 154.69 — 0.17 243+06.16 243 + 00.9 — 5.3 170 13 30 170 12 00 156.02 155.78 — 0.24 253 +34.99 253 +22.0 — 13.0 150 18 30 150 16 45 155.95 155. 78 — 0.17 257+58.35 257 + 47.6 — 10.7 120 39 30 120 37 15 190.21 190.23 + 0.02 267 + 13.73 267 +00.7 —13.0 159 20 (XI 159 17 30 191.85 191.69 — 0.16 270 +06.33 269+92.2 — 14.1 3 139 25 30 139 23 15 19.2.86 192.31 —0.55 1 Azimuth start together again from here. 2 These azimuths are cut out from the base line. 3 Difference here is due to the base line point being on a snubbing post. Height of instrument estimated. DEEP WATERWAYS 506 Table No. 1.— Comparison of stadia and base line work reduced from the reloca¬ tion by stadia of the base line for a distance of 5.6 miles —Continued. Station as per transit line. Station as per stadia line. Error in distance of stadia line (+ or —). Azimuth by transit line. Azimuth by stadia line. Elevation by level. Elevation by stadia. Error of stadia levels (+ or -). Feet. Feet. Feet. O / « O / - Feet. Feet. Feet. 276+16.07 276 +01. 7 — 14.4 120 31 00 120 29 00 190.99 190.90 — 0.09 284+41.93 284 +29.1 — 12.8 146 01 30 145 59 30 190.68 190.67 —0.01 289+08.63 289 +57.9 — 10.9 163 59 30 163 58 15 190.39 190.57 + 0.12 292 + 65.18 292 + 54.4 — 10.8 136 31 00 136 30 30 190.33 190.26 —0.07 297+69.65 297 + 60.2 — 9.4 125 35 30 125 34 45 190. 31 190.11 —0.20 301+99.41 301+90.7 -8.7 152 07 00 152 06 15 191.28 191.08 -0.20 313+10.28 313+03.7 — 8.6 183 00 30 182 59 45 191.22 191.25 +0.03 322 +57.21 322+50.9 — 6.3 178 44 00 178 43 00 191.46 191.58 +0.12 326 + 71.22 326 + 63.4 - 7.8 165 37 00 165 36 00 189.85 190.02 +0.17 332 +04.44 331+94.9 - 9.5 144 31 00 144 29 45 191.81 191.95 +0.14 342+11.26 342+00.0 -11.3 137 27 00 137 25 00 189.74 189. 79 +0.05 345 +84.25 345+72.1 _ 2 172 32 30 172 30 30 190.60 190.59 —0.01 356 H-46. 47 356 +32.6 -13.9 179 05 30 179 03 no 192.10 192.07 -0.03 361+80.69 361+66.3 -13.4 163 35 30 163 32 45 195.65 195. 75 +0.10 3664-33.37 3664-17.8 —15.6 63 57 00 63 54 15 183.60 183 74 +0.14 375+32.27 375 + 17.4 -14.7 78 49 30 78 46 45 178.56 178.50 —0.06 378+32.63 378 + 17. 5 -15.1 85 52 30 85 49 30 183.49 183.56 +0.07 388 + 44.34 388 +31.5 -12.8 31 36 30 31 33 30 192.53 192. 79 +0.26 392 +82.07 392 + 68.8 -13.3 26 35 30 26 32 15 181.46 181.72 +0.26 404+77.12 404 +65.1 -12.0 15 01 00 14 58 00 175.74 176.09 +0.35 410+88.65 410+75.8 -12. 8 352 04 00 352 01 00 179.17 179.46 +0.29 423 +62.13 423 +47.5 -14.6 28 49 30 28 46 45 180.20 180.59 +0.39 427+47.82 427+32.3 —15.5 38 18 30 38 15 45 186.70 187.11 +0. 41 431+88.25 431+72.0 — 16.2 30 36 30 30 34 15 178.26 178.54 +0.28 437+69.49 437+52.3 -17.2 44 42 00 44 40 15 177.05 177.36 +0.31 411 + 89.92 441 + 72.6 -17.3 46 49 00 46 47 30 193.62 193.96 +0.34 445 + 82.97 445 + 65.3 -17.7 37 56 30 37 55 45 215.91 216.22 +0.31 448 + 65.65 448+47.6 -18.0 43 53 30 43 53 15 190. 60 190.95 +0.35 454+17.49 453 + 98. 7 -18.8 14 02 00 14 02 30 187.29 187.67 +0.38 458+17.33 457+98.4 -18.9 15 »>•> 15 15 23 15 178.57 179.06 +0.49 A number of stadia circuits were run during the progress of the work along the entire line under many varying conditions as to tem¬ perature, wind, and character of ground passed over. To attempt to give a complete list of them, with the varying conditions under which they were run, would consume more time and space than the object in view would justify. Some of these circuits of various lengths began with and closed back on transit or stadia stations, others beginning with and closing on stadia stations of previous circuits. The general results following are selected at random as samples of the work of this kind. Sixteen circuits as run by one of the stadia parties during the period from May to October, 1898, gave the following results: Average length of circuits.. 12,366 feet. Average number of stations occupied....17. Average running error in azimuth......01' 49". Average error in elevation .... 0.20 foot. Average error in circuit. ... 1 in 1,783 feet. Maximum error in circuit._.lin 790 feet. Minimum error in circuit. 1 in 3,400 feet. Thirty-five circuits as run by another stadia party between October, 1897, and April, 1898, show: Average length of circuits _ „.. 5,356 feet. Average number of stations occupied....11. Average running error in azimuth. 01' 32 ". DEEP WATERWAYS. 507 Average error in elevation. 0.47 foot. Average error in circuit...... tin 1,408 feet. Maximum error in circuit... 1 in 801 feet. Minimum error in circuit..._... lin 4,09G feet. Forty-four circuits as run between April and July, 1898, show: Average length of circuits. __... 7.384 feet. Average number of stations occupied... 12. Average running error in azimuth...01 32". Average error in elevation ........0.35 foot. Average error in circuit.-... 1 in 1,832 feet. Maximum error in circuit .....lin 760 feet. Minimum error in circuit... Im4.244feet. The above are results of average work under usual conditions met with and can be duplicated under same conditions. Results largely in excess of these as to accuracy of distance were accomplished, but are not given here, as they are not considered fair averages of what can be accomplished. The error in the circuit, as given, is found by dividing the length of the hypothenuse of a right-angle triangle, as obtained from the closing error in latitude and departure, into the length of the stadia portion only of the circuit, the length of transit line, if any, not being included in the length of circuit. In stadia work it will be found many observers have a “personal error;” that is, a tendency to read all distances either somewhat longer or shorter than they really are. Very long sights, work done when the air is “boiling” or during high wind, or lines run over areas in which high vertical angles occur, all tend to decrease the accuracy of stadia work, but as all these con¬ ditions are met with the results given embody them. Soundings .—Soundings were generally made by stretching a line, tagged at 25-foot intervals, across the river between range stakes set opposite on each bank. These range stakes were set either by the sounding party in advance of the soundings, or by the stadia parties, as found most convenient. Soundings were taken with a line or rod, depending upon the depth of the water. Where the river was of such width as to render the use of the line impracticable, time soundings or intersections, or a combi¬ nation of both, were used. Borings .—Three different styles of boring machines were used on this division during the progress of the work, and a fourth one, dif¬ ferent from each of the others, for a short time during its last stages. The first machine used was a “Pierce well-boring machine.” Ref¬ erence is made to the maker’s catalogue fora description. The addi¬ tional machines used were built on this work, except the Sullivan machine. They were in two styles, both of wood, and of cheap con¬ struction, but rendered excellent service. One style was a simple tripod, with pulley at the top and a rope passing over the same, to one end of which either a large wooden 508 DEEP WATERWAYS. maul was attached for driving the casing, or else the drill rods for washing out the same. With the other end of the rope one or two men raised the maul or drill rods. The other style was in construc¬ tion similar to a small pile driver, with a base of about 4 feet between foot of leads and brace and extending out about 24 feet in front of the leads. It was about 15 feet in height, with leads protected by thin iron strips, in which an iron hammer, with a hollow in bottom for wooden cushion, was moved up and down by a rope attached to and passing over an 8-inch pulley at the top. Another pulley set in a bracket just in front of the leads at the top was used to pass a rope over, to which the drill rods were attached. The iron hammer weighed on one machine about 160 pounds and on the other about 180 pounds, and was used for driving the casing, the machine being slipped back about 18 inches by a bar when drill¬ ing was in progress. The machine was guyed by two ropes attached to the top and fastened to iron pins driven in the ground. This machine, with all the tools, could be loaded on a two-horse wagon in about fifteen minutes, and could be set up ready for work in about the same time. It rendered effective service, especially when the casing was hard to drive and the holes deep. The casing was 2£ inches in diameter, extra heavy, in lengths of 24 to 5 feet, with couplings. Generally, hollow drill rods, three-fourths inch in diameter, in 5 and 10 foot lengths, were used. Where rock was penetrated or tested, solid steel drill rods 1£ inches in diameter, in 10-foot lengths, with male and female screw couplings, were used. Either flat-pointed hydraulic bits with 2-inch blade or X chopping bits were attached to drill rods. Douglas A Gould force pumps were both used on the work for forcing water from barrels or streams near by to the bottom of the borings through the hollow drill rods as they were churned up and down, the overflow from the casing pipe being caught in buckets and the material, after settlement, preserved in sample bottles for that purpose. A sample was preserved at each change of material and samples taken dry, when necessary, by forcing down a hollow pipe into the material, the elevation above datum and the depth of each material being recorded on labels on the bottles. Where extra hard materials or bowlders or cobblestones were encoun¬ tered and the casing could not be driven, it was raised up 3 or 4 feet, and from one to four sticks of powder let down and fired with a bat¬ tery to loosen up the same. In severe winter weather portable houses, about 8 by 12 feet, were used, in which a stove was placed to heat water in times of freezing. DEEP WATERWAYS. 509 River borings were made by placing either the “Pierce ” or a tripod machine on a catamaran anchored in the river. The casing pipe was pulled either by a long wooden lever and chain, or by screw jacks, depending upon the difficulties encountered. The moves between borings were made by teams when practicable, or as near to the point as possible. In numerous instances the machines and outfit had to be carried by hand to places inaccessible to teams. In this case two gangs were doubled up for the moving when necessary. Two Sullivan machines were used for about one month during the last part of the work, one on land and one on river borings. The method used with these machines is, briefly speaking, the working down of 2^-inch flush-joint casing pipe by a constant twist¬ ing of the same with pipe tongs, and at the same time a liberal use of water forced down through a 1^-inch hollow drill rod, which is either pulled up and down by rope over a pulley at the top of a tripod or else by twisting the drill rod with pipe tongs, explosives being used to loosen the material or remove cobblestones or bowlders in the path of the pipe. One marked advantage in this method is that the casing is kept con¬ stantly loose and is easily raised to blast in the hole or pulled when boring is finished. Two wagon wheels attached to two legs of the tripod are used to wind the rope attached to drill rod on a drum or in moving the machine from place to place. The force employed with each machine was from 3 to 4 laborers and a foreman, a double team and driver being employed with each ma¬ chine, or for two machines, as conditions warranted, to haul water where not at hand or to move the plant. A superintendent of borings had charge of all parties, recording the results of the several borings from the books kept by the foreman, laying out and locating holes where directed, when not done by other parties, and general direction of the work under the assistant engineer. The varying conditions met with and different kinds of material encountered in a work of this kind render comparisons difficult between the results obtained from different machines and methods, a machine and method well adapted to one locality being comparatively useless in another, and results obtained as to progress made and cost of work on this division can not be taken as a criterion for others. The following tabulated statement showing depth penetrated in different kinds of material, with total penetration and cost of same per foot, maybe of interest. The borings varied in depth from a few feet to a maximum of from 100 to 190 feet. The price per foot includes all items of cost, including plant. 510 DEEP WATERWAYS. Table No. 2.— Shotring number of borings made . character of material penetrated, with total penetration and average cost per foot. Test soundings by hand with steel rod in Mohawk River. Test bor¬ ings with machine on land and river in Mohawk Valley.a Test bor¬ ings with machine on cross-coun¬ try lines and addi¬ tional bor¬ ings in Mohawk Valley. 6 Total. Number of land borings..-.. 894 271 1,165 397 Number of river borings... 397 Number of soundings.-_ 290 290 7,139 472 7.611 20.706 53 17.097 36 3,762 6.082 9,880 177 177 Gravel.-.. 118 2,678 19 2,815 161 Shale.-.-. 44 117 Hard pan....... 40 60 100 664 1,529 Sand and gravel._._ _. 290 1,637 726 861 2,728 Sand and clay...... 2,450 760 3.176 Clay and gravel..... 760 Sand and shale... 35 227 262 Clay and shale... 900 2 902 Gravel and stone.... .... \ 105 105 Gravel and bowlder... 177 177 Hardpan and bowlder ..... 87 87 Hardpan and stone.... 36 Sand and cobble..... 63 63 Gravel and shale....... 292 292 Sand,gravel.and stone.. 22 Sand, loam, and mud.... 173 727 900 Sand. clav. and gravel.. ... 1.843 1,843 91 Gravel and cobble .... 33 58 Mud. 417 417 Rock.. 413 213 626 Total penetration. . 1.087 38,052 16,382 55,521 a Including pai t of western division b Including Niskayuna-Albany line via Shakers: Niskayuna-Albany line via Town House Cor¬ ners: Scheneetady-Cedar Hill and Normans Kill lines. Mohawk Valley (boring for lock and dam sites). Average penetration of 290 soundings in Mohawk River (exclusive of depth of water), feet... ... 3.7 Average penetration of 1,291 borings in Mohawk Valley.. feet.. 29.5 Average penetration of 271 borings on cross-country lines and for dam and lock sites, feet. ..... 60.4 Average penetration of 1,562 borings, total on all lines......feet.. 34.8 Average cost of 55,521 linear feet, total penetration of all borings and soundings.. per foot.. SO. 54 OFFICE WORK IN ASSISTANT ENGINEER’S OFFICE. The office of the assistant engineer was located so as to be conven¬ ient and accessible to the several field parties, and was moved from point to point as the work progressed. The transit line was plotted on the final map sheets from the coor¬ dinates as computed and sent in by the transit party, and these sheets then sent out to the stadia parties, by whom they were returned with all points plotted in pencil, with elevations and all data except contours as taken by this party put on them. They were then inked, the borings and soundings put on, and all data checked in the assistant engineer’s office. They were then con¬ toured in pencil and carefully looked over by the head of the stadia party, who made the field sketches before inking and lettering. They DEEP WATERWAYS. 511 were then gone over by the head draftsman with the notebook in hand and any data in the stadia, boring, or sounding books not on the maps supplied, when they were sent in to the office of the United States Board of Engineers on Deep Waterways in Detroit, completed except titles. The above method of work was pursued on the Oswego route proper. On the surveys of the Normans Kill route all notes were worked up in the field and the remainder of the office work completed in the Detroit office, as the main portion of the office force was then located there and this arrangement being deemed best under the circumstances. OFFICE WORK IX DETROIT OFFICE. The first work in this office was the projecting of a center line of location for the canal on the completed map sheets by the United States Board of Engineers on Deep Waterways, and the plotting of a profile of the same on which the character of the material along the line and the elevation of the rock surface as indicated by the borings was shown. Final estimates of quantities were based upon cross-sections plotted generally at 100-foot, intervals from the map sheets showing the sur¬ face of the ground and elevation of the rock, with the canal prism drawn on the cross-section sheets, and the grade of the bottom of the canal and elevation of the water surface shown thereon. Preliminary comparative estimates, necessary to determine the values of different routes, were made either from a center-line profile on the basis of level cuttings or else by average surface elevation as taken from the map sheets. ROUTES SURVEYED. The work of this division embraced one main route, known as the Oswego route, eastern division, and three subsidiary or alternative shorter routes from the Mohawk River near Niskayuna and near Schenectady across country to the Hudson River, known respectively as Niskayuna-Albany route, via Shakers; Niskayuna-Albany route via Townhouse Corners and Schenectady-Normans Kill route. The several routes mentioned above will be considered under their respective headings. Oswego Route Proper. The Oswego route, eastern division proper, begins at Herkimer, N. V., and follows the general lines of tlie Mohawk River and its valley to the Hudson River above Troy, N. Y., the only deviations from the general line of the river and valley worthy of note being at Little Falls, N. Y., where the canal location passes along the foot of the bluff to the south of the river; at Rexford Flats, where tlie sharp rocky point of land between the New York Central Railroad and the 512 DEEP WATERWAYS. Mohawk River is cut through; at Crescent, where the line cuts across the sharp bend in the river and skirts along the foot of the hill to the south of the Mohawk; at Cohoes Falls, where the line passes along the top of the bluff to the east of the river, and where a flight of six locks is proposed. From the above point the line cuts across the north end of Simmons Island and through about the center of Van Schaicks Island, entering the Hudson about one-fourth of a mile below the Twelfth street bridge, connecting this island with Lansingburg. Field work .—The survey parties, consisting of a transit, level, and two stadia parties, began work, as authorized by the United States Board of Engineers on Deep Waterways, with the mapping of the shore lines of the Hudson River from the Congress street bridge, over the Hudson River at Troy, N. Y., to the bridge connecting Waterford and Lansingburg, X. Y., about 4^ miles above the Congress street bridge; also the several branches of the Mohawk River entering the Hudson, with the immediate country adjacent thereto, including all of Green Island, a portion of Van Schaicks and Peobles Island, and of Cohoes and West Troy, lying between the Erie Canal and the Mohawk River. The soundings proper began at the Cohoes Company’s dam above Cohoes Falls, the work below this point being done by the stadia parties. The boring parties began at the entrance of the South Branch of the Mohawk into the Hudson. The transit work was begun in October, 1897, and completed in September, 1898. The level work was begun in October, 1897, and completed m October, 1898. The stadia work was begun in October, 1897, one party completing its work in October, 1898, and the other party early in November, 1898. The work of soundings was begun in December, 1897, and com¬ pleted in October, 1898. The borings were begun in the latter part of December, 1897, and completed to Herkimer about the 1st of November, 1898, and to Frank¬ fort about the middle of November, 1898. All work was prosecuted continuously from its commencement to its final completion. The distance alone the Canal Center line from junction with Hudson River line to Washington street, Herkimer, is 83.67 miles. Number of square miles of mainland and islands mapped on this route, 54; number of square miles of river, 9.30; total number of square miles mapped, 63.30. Area of Normans Kill line is 11.95 square miles, making total area 75.25 square miles for the eastern division. Reduction of field notes and map work .—The reduction of all field notes of this route was kept close up with the work by the field par¬ ties, and also the plotting in pencil on the final map sheets of all data, ready for inking and contouring in the office of the assistant engi- DEEP WATERWAYS. 513 neer. The scale of the maps of this route was 1 to 2500 (except through Little Falls, when the scale was 1 to 1000) and contour inter¬ vals 2 feet. The 90 map sheets, embracing the total area mapped on this route, were completed, excepting titles, in the field office at Little Falls, N. Y., on February 14, and the necessary office force for the work of plotting profiles and cross sections and the making of estimates of cost of the proposed ship canal were transferred to the main office at Detroit, Mich. ARTIFICIAL FEATURES. The Mohawk Valley, from Herkimer to the Hudson, is a narrow, fertile valley, ranging in width from one-half to 3 miles, the average width, however, not exceeding three-fourths of a mile. Railroads .—The four-track New York Central Railroad follows along the north side of the valley from Herkimer to about 14 miles below Hoffmanns Ferry, where it leaves the immediate valley, passing to the north of the same, and crosses the Mohawk at Schenectady, going southeast to Albany, N. Y. The West Shore Railroad, double track, follows along the south side of the valley from Herkimer to about 3 miles above Schenectady, whence it bears off southward across country toward the Hudson River. The Fitchburg Railroad, single track, crosses the Mohawk about 5 miles above Schenectad} 7 , forming a junction with the West Shore Railroad at Rotterdam Junction. The Schenectady Branch of the Delaware and Hudson Railroad, single track, crosses the Mohawk about 1^ miles below Schenectady, going west toward South Schenectady. The Troy and Schenectady Branch of the New York Central Railroad, single track, from Schenectady, passes along the side hills and the top of the high bluffs to the south of the Mohawk and well above the same to about 3 miles below Rexford Flats, where it descends into the valley of the river, and passes through the same along the foothills to the south to about 2 miles below Niskayuua, where it again leaves the valley and rises to an elevation of about 160 feet above the river at Crescent station; then descends toward Cohoes, and crosses the South Branch of the Mohawk and the Hudson, entering Troy, N. Y., about one-half mile above the Congress street bridge. The Saratoga and Champlain Branch of the Delaware and Hudson from Albany, single track, crosses the Mohawk just above the State dam in the Mohawk, where the river separates into its several branches before its confluence with the Hudson River, and extends northward up the Hudson Valley. The Saratoga and Champlain Branch of the Delaware and Hudson from Troy, single track, crosses the canal location on the east side of Van Schaicks Island just before its entrance into the Hudson River. Existing canals .—The Erie Canal from Herkimer eastward passes along the south side of the valley between the Mohawk River and the H. Doc. 149-33 514 DEEP WATERWAYS. West Shore Railroad, descending with the river by a series of locks. It passes along the foot of the Rock Cliffs to the south of the Mohawk at Little Falls and crosses beneath the West Shore Railroad about 1 mile above Fort Plain, thence through Fort Plain to the south of the West Shore Railroad, and crosses the same again about one-half mile below. From here it follows along the south side of the valley, passing through Canajoharie and Fultonville and crossing Schoharie Creek by an aqueduct at Fort Hunter, thence through that part of the city of Amsterdam to the south of the river, and through the bottom land above Schenectady, and through this city to Rexford Flats. Here the canal crosses from the south to the north side of the Mohawk by an aqueduct, thence continuing along the north side of the valley to Crescent, where the canal again crosses the Mohawk by an aqueduct to the north side, and follows down this side and through the city of Cohoes, descending into the Hudson at West Troy, just above the Congress street bridge. Dams .—The first dams encountered in the Mohawk on this division are at Little Falls, N. Y., where three dams cross the river. The first of these, known as the New York State Dam, is at the west end of the city of Little Falls. This dam is in two parts, separated by Lock Island. It is of stone, about 6 feet in height, and has a total crest length of about 700 feet. Elevation of crest above datum, 363. On the south side of the river this dam diverts water for hydraulic power to a portion of the manufacturing industries located there, and also for the canal feeder, some 2,500 feet in length, entering the Erie Canal just below Lock No. 39. On the north side of the river, power is supplied to the large 3-story knitting mill facing on the river bank. The second or middle dam, located about one-fourth mile below the above-mentioned one, and known as the Mill Owners’ Dam, is a curved dam of cut stone, about 10 feet in height and with a crest length of about 370 feet. Elevation of crest, 356. On the south side of the river, power for hydraulic purposes is diverted by this dam for the manufactories located along the river to the South Ann Street Bridge, which crosses the Mohawk about 1,000 feet below the dam. On the north side, power is supplied to the large number of indus¬ tries as far east as the William Street Bridge, which is some 2,300 feet below the dam. The third and last dam here is about one-half mile below the Mill Owners’ Dam, and is known as the Gilbert Dam. It is a stone dam, about 9 feet in height, slightly curved, and with a crest of about 170 feet. Elevatiop of crest, 333.7. It diverts power for hydraulic purposes for the Astoronga knitting mill, on the south side of the river, and for the Little Falls Paper Company, on the north side of the river. The fourth dam in the Mohawk, just below Indian Castle, on the West Shore Railroad, and about 4 miles below Little Falls, is a DEEP WATERWAYS. 515 stone dam about 5 feet in height, and with a crest length of about 360 feet; elevation of crest, 319.2. Elevation of toj) of flash-boards, 321.2. It serves to impound water in the Mohawk and supply the Erie Canal by a feeder known as the Rocky Rift feeder, some 4 miles in length, which enters the Erie Canal just below Lock No. 34, at Mindenville. The fifth dam in the Mohawk is about 34 miles below Schenectady and about 800 feet above the aqueduct that carries the Erie Canal over the Mohawk at Rexford Flats. This is a stone dam about 5 feet in height, and a crest of about 670 feet. Elevation of crest, 209.7. It serves to supply the Erie Canal feeder at this place, some 1,900 feet in length, which enters the canal just below Lock No. 21. The sixth darn is the West Troy Water Company’s dam, about 3 miles above Crescent. This is a stone dam about 4 feet in height, separated into two parts by a rookj^ island. The upper or north dam, at the head of the island, has a crest length of about 380 feet, and an elevation of 174.2; the lower or south dam, at the foot of the island, has a crest length of about 290 feet and an elevation of 173.5. It serves to impound the river, from which a punt ping station, located at this dam, pumps water to the reservoirs supplying West Troy, N. Y. The seventh dam is the Cohoes Company’s dam, located about one- lialf mile above Cohoes Falls. This is a stone dam about 8 feet high, and with a crest from the gate-house to its north end of about 1,330 feet. Elevation of crest of masonry is 152.4, and of the flash-boards 154.2. The present dam was built in 1865, and its construction enables the converting, at low water, of the entire supply of the Mohawk into the hydraulic canal of the company, leading from the dam to the city of Cohoes, and from which the greater portion, if not all, the large manufactories in Cohoes are supplied with power. About one-lialf mile below the above dam is situated Cohoes Falls, where there is an almost perpendicular fall in the Mohawk of some 60 feet, the total fall from the foot of the Cohoes Company’s dam to the foot of Cohoes Falls being about 84 feet. The eighth and last dam on the Oswego route, eastern division, before the proposed canal location enters the Hudson, is the State dam opposite the city of Cohoes. This is a stone dam about 9 feet in height and with a crest of about 1,530 feet. Elevation of crest, 48.9. It was constructed to enable the boats on the Champlain Canal to cross the Mohawk, and also serves to divert water for hydraulic purposes for some industries on the south side of the Mohawk. Bridges, aqueducts, and ferries .—From Herkimer to where the Nor¬ mans Kill location leaves the Mohawk at Rotterdam Junction, 12 highway and 2 railroad bridges span the river. One of these railroad bridges, below Hoffmans Ferry, now in course of construction, is pro- DEEP WATERWAYS. 516 posed to carry the tracks of the West Shore Railroad to a connection with the New York Central; and the other, at Rotterdam Junction, carries the single track of the Fitchburg Railroad. Three of the above highway bridges at Little Falls and the railroad bridge at Rotterdam Junction will be undisturbed by the canal location as planned. The old stone aqueduct at Little Falls will also be untouched. It was formerly used to convey water across the river as a feeder for the Erie Canal. It is now unused and partly destroyed. From Schenec¬ tady to the Hudson River via the Mohawk River there are 7 highway bridges, 3 railroad bridges, and 2 stone aqueducts over the river. The railroad bridge at Schenectady carries the four tracks of the New York Central Railroad. The one at Mohawk, and also at Cohoes, carries the single track of the Delaware and Hudson Railroad. The aqueducts at Rexford Flats and Crescent carry the Erie Canal over the Mohawk. The only one of the above structures undisturbed by the canal location is the Crescent Aqueduct. Table No. 3. —Existing bridges, aqueducts, and ferries over Mohawle River from Herkimer to divergence of Normans Kill line. DEEP WATERWAYS. 517 2^2 to © m c5 o2o=Ss- H ^ a * ti S a ® z ©.2 £ c W © J4I f—• --- ss E CO S 3 CO coco SSs a »C 05 88 to t- 01 8 8 *1 P o Jig S 'P p p • — <0 03 -*-j 2 © AS O P ■§■3 3 I'M , OD 2.33 £jp§ S°S 3 ft .2 S5,P A CO CO 00 £8 cO 05 COro s I Is 88 rH -4 i-H sis 3 g 3 2 3 CO f* *3 ©,£ C 2 S- eg ,p p > eg © o >> *« P O 02 3 rP 03 ◄ >> u p o 03 p © Sfl 5 £ >> u o 03 Cg 02 t- CP © © 2 -2 CO co ® 2 43 43 r-S p p © OO © o 4-3 CO 43 p O P 02 O P ® o P *e P CO 3 43 >> qQ p o 03 P eg © © T- p £ o P 43 o 6Q o © p o -£ O CO 43 P o >» u P o $ a 'p © 02 h P o O OlOl .P 03 &c«m 2 ®°g ffi ■» ^^OOCC g3 g 8 Ol 8 8 3®£ & ■ 43 H © © ,fi « 8 8 8 cs a ■£ °3 O Q, Eh b ~3S3 S h ? S «Hr r-*. +3 5 © 33 EQO, p C CU o A ^3 50 P o - X3 e* 1 1 1 1 • 1 • 1 • • • t p >> • 1 • I • I 1 1 : : >> : ■ • >> 1 1 • CP . P 1 • ; : p ! • » p p o . fn bi © p 43 .S O CO P n t-4 O Cg ri O P 2) £ o r3 O 'O 25 o £ o o 22 o 'O r >|0 t.rP'P fc- tlO © *H o H3 h © eg O bp •r-« w p .2 p j © 3 © > P Pi © a 2 © ffl © © Ppp P,Pi 'H ©43 ©43^ a m 3 ® ^ fcsoooa « ® i a a affl a < 43 © © h 43 CO p p T. 2 2 © CO © Ol 03 03 2 2 P © > P . ™ *3 ^ 5?r « bo 2 ® ■“S al o.o .2 S 5 <-> a S aS 20 &| s-fe . © I H corp P % > P •2 3 m P 43 43 ^ -H U d P O •XClh^h a p ?-4 © >> 3 'S $ -s s © © p 43 © 43 H o ’g « m a Road. Table No. 3. —Existing bridges, aqueducts, and ferries over Mohawk River from Herkimer to divergence of Normans Kill line —Continued. 518 DEEP WATERWAYS. JAP c5 g a ©•« o.2 £ o o,a O W d o tl >hO A §o 'd q 2 © -p> o q q 53 t- o o; c! ?; 1 A' p 05 oi C'i ?s s r © ^ q q o O GO « 0 I | Hi to . 3 1 > q =3 j u 1 1 1 • o CT ' r P 1 1 'd a .8® 5 m © f ® : s < q ; o 0 ? 0 » ® -g * « h iH O o . H^h-P . -P i 2 0 2 O co > -P "i "I +3 ’ 1 j 3 • O co i : o oo_, ;© ;o ; io *,00 Q .. H CD 1—1 S3 CO rH h* 05 tO c o 40 tO to . +5 r G 3 0 o o m p« L ^ Cl (5 0-2 2 ^5,0 ce g M « N N 0 i>- N £ \ ~ ^ pp &G, J-< pd ,00 i -p> q > . i >0 o q 0 0 cC o £ a Q X 0 tf Q •g|S •S o s 0 CO 0 P o c ^3 O O Q oA •g “ O -P> c3 u di ■o >7 8510 350 225 50 .do_ _do... 150 8548 550 205 50 .do_ ...do... 120 Above Normansville 8931 6(H) 75 _do_ _do.. 110 Above Kenwood. 9027 400 1031 48 .do- — do... 70 a Pile foundation. The dam at Fultonville (station 0676) in order to obtain a rock foot¬ ing - , would necessitate the carrying of the foundation about 60 feet below the river bed. The material here on the south end of the dam and south side of the river is a belt of yellow clay for 9 feet below the surface, underlaid by fine sand and sand and clay mixed for a depth of about 24 feet, and beneath this about 10 feet of blue clay, overlying gravel and bowlders down to elevation 230, at which deptli no rock was struck. On the north shore of the river rock was encountered at elevation 211, the material passed through being sand and sand and clay mixed, with about 6 feet of gravel on top of the rock. 547 DEEP WATERWAYS. The elevation of the rock next to the New York Central Railroad is 245. The borings for the dam at Amsterdam (station 7210) developed rock about halfway across the dam from the north end at elevation about 200. For the south half no rock was found at elevation 174. The material is for the most part sand, gravel, and bowlders. The borings for the dam at Cranesville (station 7453) developed rock in only one boring at elevation 170. The material here is sand, gravel, and bowlders, with thin strata of blue clay and hardpan. The dam at Rotterdam (station 7753) will have to be founded on sand and sand and gravel mixed down to elevation about 210, and beneath this blue clay and sand mixed for about 50 feet, at least. Table No. 6. —Locks for 30-foot and 31-foot channels. Location. Kind. Length of level. Character of foundation N umber. Place. Station at cen¬ ter of lock. Single or double. Individual or in flight. Lift. Total lift. 30-foot chan¬ nel. 21-foot chan¬ nel. 11_ Jacksonburg ... 4940 Single.. Individual. Feet. 11 Feet. 11 Miles. 4.2 Rock . Rock. 12,13_ Little Falls_ 5165 Double. Flight_ 21.5 43 7. ti ... do ... Do. 14. Mindenville_ 5563 Single... Individual 22 22 9 ...do... Do. 15. Palatine Bridge Fulton ville. 6037 _do... .... do. ... 15 12 2 Do 16. 6680 _do... .. do .. 15 15 10 1 do 17_ Amsterdam_ 7215 ... do... _.do. 15 15 4. i) .. do... Rock. 18_ Cranesville _ 7459 _do_ .do. 15 15 Pile ... do _ Pile 19_ Rotterdam 7751 _do... .do. Guar< 1 lock. Do. 20.. Junction. French Mills ... 8406 _do .do. 15 15 17.9 Rock.. Rock. 21.... Normans Kill 8511 _do... _do. 20 20 9 ...do... Pile. 22 23 line. 8555 Double Flight 20 20.5 40 61.5 0.8 7.5 ...do... .. .do ... Rock. Do. 24,25. 26 . .do.. 8955 -do ... _do. 27,28. 29, .do. 9040 _do... _do. a 20. 7 103.5 1.4 .. .do... Do. 30,31. Total. 376 a This lift is taken at 20.7 to provide for a possible minimum low water at the lower end of lock. COST OF LOCKS. No. of lock. 11. 12,13_. 14..-. 15-.. 15.... 17.. 13.. 19 .. 20 .... 21.... 22,23... 24,25.26.... 27,28,29,30,31.. Total_ . Operating machinery Total. . 30-foot channel. 21-foot channel. Operating macliin ery. §912,926 $560,968 $100,000 3,571,594 2,434,589 175,000 1,123,687 684,348 100,000 1,010,095 615,861 100,000 1,015,283 721,484 100,000 1,062,977 648,475 100,000 1,245.321 765,818 100,000 965.777 629,650 100,000 884,446 577,209 100,000 966,258 728,626 loo, i hx) 3,114,687 2,172,165 175,000 4,720,697 3,065,295 225,000 7,445.609 4,533.769 ,350.000 28,039,357 18, LIS, 257 1,825,000 1,825,000 1,825,000 29,864,357 19,963,257 Note. —Lift of locks as given is for low-water level. 548 DEEP WATERWAYS. The elevation of the rock above the lock at Cranesville (station 7459) is about 170, and below the same about 208, as indicated by the borings. The material is of about the same nature as for the dam here. Then* is no rock for the lock at Rotterdam Junction (station 7751). The material is tlie same as that described for the dam here. Table No. 7. — Bridges. OSWEGO-MOHAWK ROUTE, EASTERN DIVISION. Sta¬ tion. Location. Kind of bridge. 1 Number | of tracks. Swing or fixed. & k c S a 5 rJ} £ c 30-foot channel. 21-foot channel. Total length. Esti¬ mated cost. Total length. Esti¬ mated cost. :»i: ;:> Highway- Swing i 545 §67,862 525 $66,382 do . .do_ ....do... i 545 105,868 525 90.666 5651) St. Johnsville.. __ .do_ _do... i 545 72,572 525 71.092 5948 Fort Plain .. .do_ _do.. i 545 92,250 525 90,770 6115 Canojarliarie. .do_ _do... i 545 71,834 525 70.354 Fonda . _ .do_ _do.. i 545 108,220 525 90.344 6982 Fort Hunter . .do _. _do . i 545 113i442 525 111.962 7260 Amsterdam.._ ...do .. _do_ i 545 149.962 525 148.482 7661 Hoffmans Ferry. Railway . 2 _do... i 550 275,539 530 263,249 7804 Rotterdam J unction .do_ i _do i 463 87. ail 423 75,512 8020 High wav - _do ... i 545 97.106 525 81,090 8071 South Schenectady . Railway . 9 Fixed .. i 600 307,342 580 287,712 8085 ...do .. ._ do .. - 1 . do i 375 81.078 363 SI (88 .. .do. Highway. _do_ i 315 40.784 305 39.940 8195 .do_ Swing. i 545 97,106 525 81,090 8302 _do_ _do... i 545 64,104 525 59.320 8413 French Mills. Railway . 1 Fixed .. i 260 39,412 250 37.057 8414 _do.. Highway. _do_ i 260 23,394 250 22.711 8485 _do ... Swing.. i 545 97,106 525 81,090 _do_ _do.. i 545 97,106 525 81,090 _ do_ _do_ i 545 97,106 525 81.090 8813 _do_ _do_ i • 545 97,106 525 81,090 8960 Normans ville. _do_ _do ... 2 335 20,476 296 17,140 9011 Railway . l _do... 1 5371 141,938 5174 120,300 9(141 Kenwood.. Highway. .. do... 1 235 15,946 196 12,610 9058 .do__ Railway . l -.. do... 1 228 21,773 197 17.215 Normans Kill a.. Highway Fixed .. 1 100 4,810 100 4.810 4275 Kenwood a .. Railway . i _do .. 1 150 12.234 150 12,234 Hoffmans Ferry« 20,000 20,000 (steam ferry). Total_ 2,520,807 2,293.040 a Bridges not over canal. Feet clear opening. Highway bridges... 22 Single-track railway bridges.. . 14 Double-track railway bridges. ......26 21-FOOT ADOPTED CHANNEL. The location as adopted for this channel is the same as that described for the 30-foot channel, and its total length is the same. 'I'lie same statements made for the 30-foot channel, as to the size of prism, locks, and estimated costs for change of railroads and land damages, apply also to this channel. The alignment is the same for this channel as for the 30-foot channel. The two channels are the same for depth and sectional area from the “Little Nose” (station 6389 + 92) to the head of the lock at Rot¬ terdam Junction, which is the beginning of the Normans Kill line, for the reasons already stated in this report. DEEP WATERWAYS. 549 Table No. 8. —Canal prism 21-foot channel. Location. Earth. Rock. From— To— Sec¬ tion. Bottom width. Sec¬ tion. Bottom width. Herkimer (station 4789+92)_ West Canada Creek (station Sq.feet. 5,397 Feet. 215 Sq.feet. 5.040 Feat. 240 West Canada Creek (station 4849+92). Jacksonburg Creek (station 7,014 292 7,098 338 4849+92). Jacksonburg (station 49.28+92)-- 4938 +92). Little Falls Creek (station 6,489 267 6,552 312 Little Falls (station 5119+92).. 5119+92). Little Falls Creek i station 5,397 215 5,040 240 Little Falls (station 5161+92)... 5161+92). East Canada Creek (station 6,489 267 6,552 312 East Canada Creek (station .5479 + 92). Mindenville (station 5561+92) .. 8,043 341 8,190 390 5479+92). Mindenville (station 5561+92) .. Garoga Creek (station 5819+92). 7,623 321 7,665 365 Garoga Creek (station5819 +92). Palatine Bridge (station 6035+ 8,526 364 8,526 406 Palatine Bridge (station 6085+ 92). Little Nose (station 6389 +92)... Little Nose (station 6389 +92) ... 7,791 329 7.812 372 Fultonville (station 6680+92) . .. 10,500 290 10,620 354 Fnltonville (station6680 +92)_ Schoharie Creek (station 6989+ 10,500 290 9.750 325 Schoharie Creek (station 6989+ 92). Amsterdam (station 7214+92) .. Amsterdam (station 7214+92)... 13,980 406 14,400 480 Cranesville (station 7453 +45)... 15,600 460 15,300 510 Cranesville (station 7453 + 45) ... Rotterdam Junction (station 15,600 460 15,420 514 Rotterdam Junction (station 7732 + 30). Kenwood (station 9052 +53)_ 5,397 215 5,040 240 i i 32-b30). Kenwood (station 9052+53). (Station 9106+19). Irregular shaped basin enter¬ ing Hudson River. Ill the above table change of section at approach to and leaving locks, and for curves where necessary, is not noted. The canal prism is the same for this channel as for the 30-foot channel from station 6389 + 92 to station 7732+30. Dams .—Table No. 5, as given for 30-foot channel, also applies to this channel. Larks .—Table No. 6 includes the locks on the 21-foot channel. At Fultonville (station 6680) the rock for the lock for this channel is about 15 feet below the bottom of the foundation. The material is clay and fine sand, loosely mixed for about 30 feet below the surface, overlying about 10 to 20 feet of soft, blue clay, and beneath this sand, gravel, and bowlders extending to the rock surface at elevation about 229 There is no rock for the lock at Cranesville (station 7458). The material is sand, gravel, and bowlders, with thin strata of blue clay and hardpan. The material at Rotterdam Junction (station 7751) for the lock is the same as that described for the dam there. The borings indicate rock at about the elevation of the bottom of the foundation in the center of the lock at station 8511, but the rock is about 17 feet below the same at its west end and about 60 feet below at its east end. The material is for the most part blue clay and gravel mixed, and is hard. Bridges , railroad and highway .—The table for the above, as given for 30-foot channel, also applies to this channel. 550 DEEP WATERWAYS, Table No. 9.— Oswego-Mohairk route , eastern division—Estimate of construction of 30-foot channel. SECTION NO. 1 (STATION 4789 + 91.5 TO 4848 + 00), FROM HERKIMER TO 1} MILES EAST OF HERKIMER. Quantity, j Total. Excavation: Gravel..cubic vards.. Sand......do Earth ....do Right of wav, farm land.....acres.. Entrance of streams, submerged weirs... 1,211,024 725,897 725,896 189 $0.18 .18 .18 100.00 $217,984 130,661 130,661 18,900 2.352 34,873 Slope wall.square yards. Total........ 31,703 1.10 535,431 SECTION NO. 2 (STATION 4848 TO STATION 5118), FROM H MILES EAST OF HERKLMER TO LITTLE FALLS. Excavation: Rock, quartzite, wet.. Rock, quartzite, dry.. Rock, quartzite, artificially dry Clay... Gravel. Sand... Earth. Diversion and levees.... Right of way: Farm land.. Town land... Railroad changes. Retaining wall..... Slope wall... Back fill.... Timber cribs: Oak.... Hemlock.. Pine... Stone fill. Iron bolts. Lock No. 11..... Operating machinery... Dams: Concrete, dam and wing. Concrete, core wall.. Excavation, earth. Embankment.. Cofferdam. Total. ... cubic yards.. 295,951 $2.50 854,600 ..do_ 709,818 ...do_ 988,104 .. do_ 861,281 ...do_ 3,504,267 --.do_ 1,630,330 .75 1.00 .18 .18 .18 .18 8729,878 640,950 709,818 177,859 155,031 630, 768 293,459 85,000 .acres.. .do_ .miles.. ..cubic yards., square yards.. . cubic yards. . 868 5 3.05 270,271 42,584 759,282 100. 00 14,286.00 4.00 1.10 .25 86,800 71,430 23,851 1,081,084 46,842 189,821 ... feet B. M._ .do_ .do_ cubic yards.. .pounds.. 103,440 7,267,067 3,060,093 122.640 852,013 a 50.00 a 23. 00 a 30.00 .60 .03 5,172 167.143 91,803 73,584 25,560 912,926 100,000 cubic yards.. _do_ .do_ ....do_ 21,606 10,000 53,630 45,750 129.636 45,000 9.653 6,863 6.00 4.50 .18 .15 12,800 6,512,731 SECTION NO. 3 (STATION 5118 TO STATION 5175), LITTLE FALLS. Excavation, rock quartsite, dry___cubic yards.. Right of way: Farmland...acres_ Town land.do_ Retaining wall.....cubic yards.. Timber crib: Oak.feet B. M.. Hemlock. do_ Pine. do_ Stone fill... cubic yards.. Iron bolts... pounds.. Locks Nos. 12 and 13...... 2,324,504 47 37 44,811 19,200 1,000,000 378, 400 16,380 137,660 $0.75 100.00 22,124.00 4.00 a 50.00 a 23.50 a 30.00 .60 .03 $'1,743,378 4,700 818,588 179,244 960 23,000 11,,352 9,828 4,130 3,571,594 175,000 30,900 1.190 27.030 67,862 Operating machinery. .... Dams: Concrete, dam and wing walls.cubic yards.. Excavation, earth....do_ Cofferdams... 5,150 6,610 6.00 .18 Bridge.... 1 Total... 6,668,726 a Per 1,000 feet. DEEP WATERWAYS 551 Table No. 9. — Oswego-MohawJc route, eastern division—Estimate of construction of 30-foot channel —Continued. SECTION' NO. 4 (STATION 5175 TO STATION 5260), FROM LITTLE FALLS TO THREE- FOURTHS OF A MILE BELOW SUSPENSION BRIDC4E. Quantity. Cost per unit. Total. Excavation: Rock, hard, artificially dry.cubic yards.. Sand.......do_ Diversions and levees... 128,721 2,013,232 $1.00 .18 $128,721 362,382 10,000 29,300 56,244 9,288 1,752 79,954 29,065 33,504 11,781 105,868 Right of way, farm land...acres.. Slope wall..... ..square yards.. Back fill...cubic yards.. Timber crib: Oak-.’. .feet B. M.. Hemlock.do_ Pine. do_ Stone fill..cubic yards . Iron bolts...pounds.. Bridge. 293 51,131 37,152 35,040 3,476,260 968, 820 55,840 392,695 1 100.00 1.10 .25 a 50.00 a 23.00 a30.00 .60 .03 Total __________ 857,859 SECTION NO. 5 (STATION 5260 TO STATION 5651), FROM THREE-FOURTHS OF A MILE BELOW SUSPENSION BRIDGE TO ST. JOHNSVILLE. Excavation: Rock, dry. Rock, artificially dry. Clay. Gravel. Sand.. Earth... For lock at additional price: Earth . Rock .. Diversion and levees.. Right of way, farm land. Railroad changes... Entrance of streams, submerged weirs Retaining wall. . Slope wall. Back fill. Timber cribs: Oak.. Hemlock.. Pine. Stone fill... Iron bolts. Lock No. 14. Operating machinery. Dam: Concrete. Excavation, earth.-. Cofferdam. Bridge.. Total.. cubic yards .do_ .do_ .do_ .do_ .do_ 1,585,022 2,109,283 251,776 1,815,080 4,622,108 1,519,296 $0.65 $1,030,264 . 75 1,581,962 . 18 45,320 .18 323,714 .18 831.979 .18 273,473 do.... do.... acres.. miles.. ..cubic yards., square yards.. ..cubic yards.. 255. 707 .32 185,593 .50 2,200 100.(Ml 4.22 327, 737 4.09 81.492 1.10 537,086 .2b 81,862 92, 797 200,000 220,000 73,795 2,006 1,310,948 89,641 134,272 ... feet B.M.. ..do_ ..do_ cubic yards.. _pounds. . 83,040 1 a 50.00 j 5,970,360 «23.90 1,763,200 a 30.00 100,350 I .60 i 668,886 | .03 I 4,152 137,318 52.896 60,210 20,067 1,123,687 100,000 cubic yards.. .. do ... 35,960 1 6.00 68,944 i . 50 1 8 , 215,760 34, 472 20.090 72,572 136,131 SECTION NO. 6 (STATION 5651 TO STATION 6103), FROM ST. JOHNSVILLE TO CANAJOHARIE. Excavation: Rock, wet...- 591,155 $2.00 $1,182,310 Rock, drv. .do_ 595,185 . 65 386,870 Rock, artificially drv.. .do — 1,144,.352 . 75 1,083,264 Clay... ..— do_ 1,128,556 .18 203,140 Gravel... ...do_ 1,875,048 . 18 337,599 Sand. . .do — 7,192,600 .18 1,294,668 Earth... 2,175,193 . 18 391,535 Extra for lock: Earth ..... 204,577 .32 65,465 Rock.. 167,970 .60 100,782 Diversion and levees. 310,306 a Per 1,000 feet. DEEP WATERWAYS 552 Table No. 9. — Oswego-Mohawk route , eastern division—Estimate of construction of 30-foot channel— Continued. SECTION NO. 6 (STATION 5651 TO STATION 6103), FROM ST. JOHNSVILLE TO CANA.JOHARIE—Continued. Quantity. Cost per unit. Total. Right of way: Farmland...acres.. Town land ....l.do- 1,544 30 1.51 $106.00 2,800.00 §154,400 84,000 26.304 2.569 443,252 273,571 57,394 2,544 77,062 31,344 35,100 9,526 1,010,095 100,000 134,964 25,105 14,500 92.250 Entranpfi of streams, submerged weirs_ Retaining wall.cubic yards.. Slope wall. square yards.. Back fill... cubic yards.. Timber crib: Oak.feet B. M.. Hemlock... ...do- Pine ...-.-. -do — Stone fill. cubic yards.. Iron bolts..pounds.. Lock No. 15______ 110,813 248,701 229,575 50,880 3,350,514 1,044,806 58,500 317,535 4.00 1.10 .25 a 50.00 a 23.00 a 30.00 .60 .03 Dams: Concrete.....cubic yards.. Excavation, earth........do- Cofferdam ........ 22,494 50,210 6.00 .50 Bridge.-___ 1 Total ... 7,929,829 1 SECTION NO. 7 (STATION 6103 TO STATION 6300), FROM CANAJOHARIE TO FULTON VILLE. Excavation: Rock, wet. Rock, artificially dry. Clay and bowlders, dry. Clay... Gravel... Sand .. Earth .. Extra for lock: Earth.-. Rock.. Diversion and levees... Riglit of way: Farm land. Town land. Railroad changes... Entrance of streams, submerged weirs Retaining wall.. Slope wall... Back fill. Timber crib: Oak.... Hemlock.-... Pine... Stone fill.... Iron bolts.. Lock No. 16. Operating machinery. Dams: Concrete, dam and wing. Timber in foundation. Piles in foundation .. Sheet piling in foundation. Iron in foundation. Excavation, earth... Embankment. Cofferdam .. Bridges. cubic yards.. .do_ .do_ ..do_ .do_ .do_ .do... ...do_ ..do.... acres.. ..do_ miles.. ..cubic yards., square yards. . ..cubic yards.. 32,970 §2.00 399,964 . 75 475,667 .20 2.892.628 .18 1,163,980 . 18 16,488.434 .18 4,188,200 .18 496,970 .32 10,842 .50 2,422 100.00 96 3,193.00 3.64 81,674 4.00 443.335 1.10 547,683 .25 _feet B. M-. .do_ .do_ cubic j T ards.. _pounds.. 122,880 10,715,304 3,522,566 181,590 1,286,180 a50.00 a 23.00 a30.00 .60 .03 cubic yards.. .. .feet B. M-. ..linear feet.. _feet B. M.. _pounds.. cubic yards . .do_ 19,902 1,191,000 163,320 262,800 102,000 53,400 3,700 .. 2 I 6.00 a 22.00 .20 a 33.00 .03 .50 .15 §65,940 299,973 95,133 520,673 209,516 2,967,918 753,876 159,030 5,421 60,000 242,200 306,528 40,560 4,252 326,696 487,669 136.921 6,144 247,144 105,677 108,954 38,585 1,015,283 100,006 119,412 26,202 32,664 8,672 3,060 26,700 555 25.000 180,054 Total. 8,726,412 a Per 1,000 feet. DEEP WATERWAYS 553 Table No. 9. — Oswego-Mohawk route , eastern division—Estimate of construction of 30-foot channel —Continued. SECTION NO. 8 (STATION 6800 TO STATION 7732), FULTONVILLE TO ROTTERDAM JUNCTION. Excavation: Rock, dry. Rock, wet. Rock, artificially dry. Clay. Gravel. Sand.. . Quicksand. Earth-;. Extra for locks: Earth. Rock . Diversion and levees. Right of way: Farm land ..... Town land... Railroad changes.... Entrance of streams, submerged weirs Retaining wall. Slope wall... Back fill.. Timber crib: Oak.. Hemlock.. Pine . . Stone fill. Iron bolts.. Lock No. 17.. Operating machinery.... Lock No. 18 ... Operating machinery... Dam at Amsterdam: Concrete... Excavation, earth... Cofferdam.. Dam at Cranesville: Concrete.... Timber in foundation... Piles in foundation. Sheet piling. Iron. Excavation, earth .... Cofferdam.:. Bridges... Steam ferry... Total..... Quantity. Cost per unit. cubic yards.. .do_ .do_ .do_ .do_ .do- .do_ .do_ 398,163 612,927 610,427 6,293,037 13,755,344 15,912,153 176,790 6,316,712 Si). 65 2.00 . 75 .18 .18 .18 .18 .18 Total. §258,806 1,225,854 457,820 1,132,747 2,475,962 2,864.188 31,822 1,137,008 cubic yards. . ..do_ 463,430 96,521 .32 .60 acres.. ..do — miles.. 3,168 100.00 61 7,486.00 3.99. ..cubic yards., square yards.. ..cubic yards.. 90,244 4.00 623,737 1.10 421,272 .25 148,298 57,913 47,000 316,800 456,646 105,893 2,643 360,976 688,111 105,318 ...feet B. M.. .do_ .do_ cubic yards.. .pounds.. 191,040 16,746.154 5,657,946 282.590 2,032,610 a 50.00 «23.00 a 30.00 .60 .03 9,552 385,162 169,738 169,554 60,978 1,062.977 100,0110 1,245,321 100,000 cubic yards.. .do.... 35,812 72,800 6.00 .50 214,872 36,400 17,800 cubic yards.. _feet B. M.. . .linear feet.. ...feet B.M.. .pounds. . cubic yards.. 17,325 1,116,720 154,140 249,600 100,000 40,800 6.00 a 22.00 .20 «33.01) .03 .50 103,950 24,568 30,828 8,237 3,000 20,400 11,000 538,943 20,000 16,205,085 SECTION NO. 9 (STATION 7732 TO STATION 8043), FROM ROTTERDAM JUNCTION TO 1 MILE NORTH OF SOUTH SCHENECTADY. Excavation: Rock, dry Hard material, dry Clay. Gravel .. Sand.. Earth.. Extra for lock, earth Right of way: Farm land.. Town land. Retaining wall. Slope wall. Back fill.. Core wall. Embankment. Timber crib: Hemlock. Pine.. cubic yards.. .do_ ..do_ .do_ .do_ _do — cubic yards.. .acres.. ,.do_ ..cubic yards. . square yards.. ..cubic yards.. .do_ .do_ _feet B. M_. .do_ 560,525 1,338,299 1,240,789 1,106,294 3,835,337 5,727,104 366,656 1,593 36 9,340 190,763 280, 761 15.417 1,046,767 6,990,900 1,568,140 $0.60 .30 .18 .18 .18 .18 .32 100.00 2,600.00 4.00 1.10 .25 4.50 .15 a 23.00 a 30.00 $336,315 401,490 223.342 199,133 690,361 1,030,879 117,330 159.300 93,600 37,360 209,839 70.190 69,377 157,015 160,791 47,044 a Per 1,000 feet. 554 DEEP WATERWAYS Table No. 9.— Oswego-Mohawk route , eastern division—Estimate of construction of 30-foot channel —Continued. SECTION NO. 9 (STATION 7732 TO STATION 8043), FROM ROTTERDAM JUNCTION TO 1 MILE NORTH OF SOUTH SCHENECTADY—Continued. Quantity. Cost per unit. Total. Timber crib: Oak....feet B. M.. Iron........pounds.. Stone fill....cubic yards.. Lock No. 19 . _ ......... 88,776 738,680 112,984 a 850. TO .03 .60 84.439 22,160 67,790 965,777 100,000 192,240 40,500 30,945 43,722 9,346 3.960 37,913 4.935 14,000 184,437 Operating machinery............_ Dams: Concrete...cubic yards.. Concrete core wall . do ... Timber in foundation..feet B. M_. Piles in foundation......linear feet.. Sheet piling in foundation.feet B. M.. Iron .. pounds.. Excavation . ....cubic yards.. Embankment. do_ Cofferdam...... 32,040 9, TOO 1.406,600 218,610 283,200 132, TOO 75,825 32,900 6.00 4.50 a 22.00 .20 a 33. TO .03 .50 . 15 Bridges.... 2 Total...... 5,725,530 SECTION NO. 10 (STATION 8043 TO STATION 842!)), FROM 1 MILE NORTH OF SOUTH SCHENECTADY TO FRENCH MILLS. Excavation: Rock, dry. Hardpan. Clay..... Gravel. Sand .. Earth . Long haul, shale. Long haul, quicksand Long haul, clay.. Long haul, gravel_ Long haul, sand.. Long haul, earth. Right of way: Farm land.. Town land --- Entrance of streams. Retaining wall.. Slope walls. Back fill. Timber crib: Oak. Hemlock. Pine. Stone fill. Iron bolts.. Lock No. 20... Operating machinery ... Dam: Concrete .. Cofferdam.. Bridges. Total... cubic yards.. ..do_ ..do_ .do_ .do_ .do_ ..do_ .do_ ..do_ ..do_ .do_ .do_ 603,252 4,654.025 679, 757 339,879 1,168,794 253.788 7,334,260 4,007,249 3,259,155 3,259,155 15,604,521 2,351,06 1 $0.60 8361,951 .30 1,396.208 .18 122,356 .18 61.178 .18 210,383 .18 45,682 .60 4,400,556 .17 681,232 .17 554,056 .17 554,056 .17 2,652,769 .17 400,786 acres. . ..do_ ..cubic yards., square yards.. ..cubic yards.. 1,516 | 100.00 7 | 500.00 78,252.4.50 92,175 1.45 616,210 .25 151,600 3,500 30,294 352,134 133,654 154,053 ...feet B.M.. .do_ _do.... cubic yards.. _pounds.. 87,576 7,428,272 1,068,988 113,519 730,958 a 50.00 a 23.00 a 30.00 .60 .03 | 4,379 170,850 32,070 68,111 21,929 884,446 100,000 cubic yard.. 1.548 6.00 8 9,288 6, Oi 10 658,030 14,221,551 SECTION NO. 11 (STATION 8423 TO STATION 8663), FROM FRENCH MILLS TO ROAD TO VOORHEESVILLE. Excavation: Rock, dry. 318,913 SO. 60 $191,348 Hardpan, dry. 5,368,895 .30 1,610,669 Right of way, farm land. ...acres.. 1.019 100.00 101,900 Slope wall. 9,771 1.45 14,168 Back fill... 378,981 .25 94,745 Timber crib: Oak. 152.400 a 50. TO 7,620 Hemlock... a Per 1,000 feet. 15,550,642 a 23. TO 357,665 DEEP WATERWAYS 555 Table No. 9. — Oswego-Mohawk route, eastern division—Estimate of construction of 30-foot channel —Continued. SECTION NO. 11 (STATION 8423 TO STATION 8603), FROM FRENCH MILLS TO ROAD TO VOORHEESVILLE—Continued. Quantity. Cost per unit. Total. Timber crib—Continued. Pine .. Stone fill. Iron bolts... T.ock No. 21 _ _ ... _ ... .feet B. M.. .pounds.. 1,949,308 197.644 1,430,686 a §30.00 .60 .03 §58,479 118,586 43,921 966,258 Operating machinery... 100,600 3,114,687 175,000 Locks Nos. 22 and 23... Oneratinsr maoliinerv. _ _ _ ... _ Dam: Concrete.. .. Excavation, earth.. Concrete . Excavation, earth... Bridges . .cubic yards.. .do... ...do_ .do_ 9,601 29,800 60,696 99,000 ,> 6.00 .15 6.00 .15 57,606 4,470 364,176 14,850 194,212 Total_ 7,589,360 SECTION NO. 12 (STATION 8663 TO STATION 8923), FROM ROAD TO VORHEESYIDLE TO 1 MILE WEST OF NORMANSVILLE. Excavation: Rock,dry .cubicyards.. Clay. . ..do_ Sand..do — Right of way, farm land... acres.. Slope wall.. square yards. Bridges ........ 5,281 4,037,641 1,345,880 4. *34 30.771 §0.60 .15 .15 100.00 1.45 §3,169 605,646 2)1,882 433,400 53.318 194,212 Total. .. . . 1,491,627 SECTION NO. 13 (STATION 8933 TO STATION 9106+18.5), FROM 1 MILE WEST OF NOR¬ MANSVILLE TO HUDSON RIVER. Excavation: Rock.hard,dry. Rock, dry_ Clay.... Sand .. Earth.. Right of way: Farm land. Town land .. Railroad changes. Retaining wall. Slope wall. Back fill . Timber crib: Oak... Hemlock. Pine. Stone fill.. Iron bolts. Locks Nos. 24,25, and 26. Operating machinery.. Locks Nos. 27,28,29,30, and 31 Operating machinery. Dam: Concrete. Excavation, earth. Concrete. Bridges... Total... ..cubic yards.. .do_ __do_ .do_ ...do_ .acres.. .do_ ..miles . ..cubic yards., square yards.. ..cubic yards.. 2,627,030 §0.65 $1,707,570 537,516 .60 322.510 6,954,046 .15 1,043.107 661,638 . 15 99.246 451,681 .15 67,752 585 1(H). 00 58,500 13 3,846. 00 50,000 o 119,!H)2 7,504 4.50 33, 768 28. 7*1 1.45 41,663 547,705 .25 136,941 _feet B. M-. .do.... .do_ cubic yards.. .pounds.. 161,784 18,789.595 1,567,981 273, 784 1,667,437 a 50.00 a 23.00 a 30.00 .60 .03 8,089 432,161 47,039 164.270 50.023 4,720,697 225.000 7,445.609 350,000 cubic yards.. ..do_ ....do_ 44,495 6.00 266,970 34.900 .15 5,235 6,750 j 6.00 40,500 5 1 . 212,367 17,648,919 a Per 1,000 feet. DEEP WATERWAYS 550 Table No. 0.— Oswego-Moliawk route , eastern division—Estimate of construction of 30-foot channel —Continued. SUMMARY. Section. Station to station. Total cost. 1 .... 4789+91.5 to 4848 . $535,431 6,512,731 6.668,726 857,859 8,136.131 7,929,829 8,726,412 16,205,085 5,725,530 14.221,551 7,589,360 1,491,627 17,648,919 *> 4848 to 5118. 3 ...... 5118 to 5175.. 5 ... 5260 to 5651. I) ..... 5051 to 0103.. 7 ... 6103 to 6800... 8 ... 6800 to 7732... u . ..... 7732 to 8043__ 10 .... 8043 to 8423... 11 . ... 8423 to 8663. 12 .... 8663 to 8923_ 13 .. 8923 to9106+18.5. —. 102,249,191 Table No. 10.— Oswego-Moliawk route, eastern division—Estimate of construction of 21-foot channel. SECTION NO. 1 (STATION 4789 +91.5 TO STATION 4848 +00), FROM HERKIMER TO 1* MILES EAST OF HERKIMER. Quantity. Cost per unit. Total. Excavation: Gravel*!. Sand. Earth.. ... Right of way, farm land.. Entrance of streams. .cubic yards.. . do.... _ __do_ ......acres.. 838,919 601,863 601,862 189 $0.18 . 18 .18 100.00 $151,005 108,335 108,335 18,900 2,352 Slope wall. .square yards.. 29,814 1.10 32,795 Total_ 421,722 SECTION NO. 2 (STATION 4848 TO STATION 5118), FROM U MILES EAST OF HERKI¬ MER TO LITTLE FALLS. Excavation: Rock, dry (quartzite)... Rock, wet (quartzite). Rock, artificially dry (quartzite) Clay.--- Gravel.... Sand... Earth.... Diversion... Right of way: Farmland... Town land... Railroad changes..... Retaining wall... Slope wall..... Back fill.... Timber crib: Oak.. Hemlock...... Pine... Stone fill.... Iron bolts. Lock No. 11. Operating machinery.. Dam: Concrete, dam and wing wall_ Concrete, core wall ... Excavation, earth. Embankment. Cofferdam. Total. cubic yards.. _do_ ..do.. . .do_ -do_ .do_ .do_ 342,786 137,078 111,445 750,024 504,467 3,263,024 1,481,095 SO 75 »> 50 1 00 18 18 18 18 $257,090 342,695 111,445 135,004 90,804 581,344 266,597 60,000 .acres.. .do_ ..miles.. .eubic yards., square yards.. ..cubic yards.. 668 5 3.05 50.355 116,720 149,071 100.00 14,286.00 4.00 1.10 2fi 86,800 71.4:10 23,851 201,420 128.392 37,268 ...feet B.M-. .do_ ....do ... cubic yards.. .pounds.. 103,440 7,293,355 1,616,445 110,210 785,255 a 50.00 a 23.00 a 30.00 .60 .03 5,172 167, 747 48,493 66,126 23,558 560,968 100,000 .cubic yards.. .do_ ..do_ ..do_ 21,606 10,000 53,630 45. 750 6.00 4.50 .18 .15 129,636 45,000 9,053 6,863 12,800 3,576,156 a Per 1,000 feet. DEEP WATERWAYS. 557 Table No. 10. — Oswego-Mohawk route, eastern division—Estimate of construction of 21-foot channel —Continued. SECTION NO. 3 (STATION 5118 TO STATION 5175), LITTLE FALLS. Quantity. ' ^P 01 ' Total. Excavation, rock, dry (quartzite)...cubic yards.. Right of way: Farm land...acres.. Town land.....do Retaining wall..cubic yards.. Timber crib: Oak.feet B M.. Hemlock.....do Pine.... ...do Stone fill...cubic yards.. Iron bolts.pounds. Locks Nos. 12 and 13....... 1,837,425 47 37 54,405 19,200 920,400 367,200 9,660 79,500 $0.75 100.00 22.124.00 4.00 a 50.00 a 23.00 a 30.00 .60 .03 $1,378,069 4,700 818,588 217,620 960 21,169 11,016 5,796 2,385 2,434,589 175,000 30,900 1,190 27,000 66,382 Operating machinery. Dam: Concrete, dam and wing walls.cubic yards.. Excavation,earth... .do_ Cofferdam. 5,150 6,610 6.00 .18 Bridge... 1 Total...... 5,195,364 SECTION NO. 4 (STATION 5175 TO STATION 5260), FROM LITTLE FALLS TO THREE- FOURTHS OF A MILE BELOW SUSPENSION BRIDGE. Excavation: Rock, hard, artificially dry.cubic yards.. Sand.....do ... Diversion . .... 147,717 1,674,046 $1.00 .18 $147,717 301,328 10,000 29,3< H) 81,168 1,752 61,865 27,189 26,175 9,436 90,666 Right of way, farm land.acres.. Slope wall.square yards.. Timber crib: Oak.feet B. M.. Hemlock. ...do... Pine.do Stone fill......cubic yards.. Iron bolts.pounds.. Bridge... 293 73,789 35,040 2,689,740 906,300 43 625 314,525 1 100. (X) 1. 10 a 50.00 «23.00 a 30.00 .60 .03 Total...... 786,596 SECTION NO. 5 (STATION 5260 TO STATION 5651), FROM THREE-FOURTHS OF A MILE BELOW SUSPENSION BRIDGE TO ST. JOHNSYILLE. Excavation: Rock, dry. Extra for lock, rock, dry_ Rock, artificially dry. Clay. Gravel. Sand. Earth . Earth, extra for lock. Diversion. Right of way, farmland.. Railroad changes. Entrance of streams.. Retaining wall. Slope wall.-.. Back fill.. Timber crib: Oak. . Hemlock. Pine. Stone fill. Iron. Lock No. 14. Operating machinery. Dam: Concrete, dam and wing wall Excavation, earth. Cofferdam. Bridge. Total.. cubic yards.. .do_ .do_ .do_ .do_ .do_ .do_ .do_ acres.. miles.. ..cubic yards., square yards.. ..cubic yards.. 861,759 $0.65 82,804 . 60 906,112 . 75 242.450 .18 1,708,542 .18 4,298,096 .18 1,523,982 .18 93,793 .32 2,200 100.00 4.22 135,696 4.00 128,745 1.10 228,510 .25 $500,143 f!*, 682 679,584 43,641 307,538 773.657 274,317 30,014 200, (XXI 220,000 73.795 2,006 542,784 141,620 57,128 _feet B. M.. .do_ .do_ cubic yards.. .pounds.. 102,240 5,292,820 2,747,520 102,050 688,192 a 50.00 a 28.00 a 30.00 .60 .03 5,112 121,735 82,426 61,230 20.646 684,348 100,000 cubic yards.. .do_ 35,960 6.00 68,944 . 50 1 215,760 84.472 20,000 71,092 5,372,730 a Per 1,000 feet. 558 DEEP WATERWAYS Table No. 10.— Oswego-Mohawk route, eastern division—Estimate of construction of 21-foot channel —Continued. SECTION NO. 6 (STATION 5651 TO STATION 61(8), FROM ST. JOHNSVILLE TO CAN AJOHARIE. Excavation: Rock, dry. Rock, dry, extra for lock- Rock, wet..... .. Rock, artificially dry... Clay.—. Gravel. Sand.. Earth.-. Earth, extra for lock.. Diversion. Right of way: Farm land. Town land ..... Railroad changes.. Entrance of streams ___ Retaining wall... Slope wall.... Back fill...... Timber crib: Oak .... Hemlock.. Pine. Stone fill.... Iron bolts.. Lock No. 15....... Operating machinery.. Dam: Concrete, dam and wing wall Excavation, earth. Cofferdam... Bridge.... Total Quantity. I Cost per unit. Total. cubic yards. . ...do_ .do_ ..do.... ..do_ .do_ ..do_ ..do_ ..do_ acres.. —do.. 96.214 73.814 88,429 567,710 1,122,065 2,057,292 7,180,540 2,304,549 96,979 1,544 30 I 2 miles.. 1.51 ..cubic yards., square yards.. ..cubic yards.. 33,560 270,089 120,784 $0.65 .60 2. IK) .75 .18 .18 .18 .18 .32 100.00 ,800.00 4.00 1.10 .25 $62,539 44,288 176,858 425,783 201,972 370.313 1,292,497 414,819 31,033 185,000 154,400 84,000 26,304 2,569 1:34,240 297,098 30,196 ...feet B. M.. .do_ .do_ cubic yards.. .pounds.. 122,880 5,388,910 3,094,800 106,080 732.467 a 50.00 a 23. (K) a 30.00 .60 .03 6,144 123,945 92,844 63,648 21,974 615.861 100,000 cubic yards.. 22,494 .do.... 50,210 1 6.00 .50 134,964 25,105 14,500 90,770 5,223,664 SECTION NO. 7 (STATION 6103 TO STATION 6800), FROM CANAJOHARIE TO FULTON YILLE. Excavation: Rock, wet.. Rock, artificially dry.. Clay and bowlders, dry. Clay.. Gravel.. Sand..... Earth. Earth, extra for lock .. Diversion. Right of way: Farm land.... Town land. Railroad changes. Entrance of streams .. Retaining wall. . Slope wall..... Back fill__ Timber crib: Oak... Hemlock___ Pine.. Stone fill.... Iron bolts... Lock No. 16.... Operating machinery.. Dam: Concrete, dam and wing wall Timber in foundation. Piles in foundation.. Sheet piling in foundation... Iron... .. Excavation, earth. Embankment. Cofferdam. Bridges.. Total... cubic yards.. _do_ ..—do_ .....do- .do_ .do_ ..do_ ..do... 15,199 181,404 475,667 2,892,628 1,115,445 15, t)00,066 4,166,681 280.120 S2. 00 . 75 .20 .18 .18 . 18 .18 .32 330,398 136,053 85.133 520,673 200,780 2,700,012 750,003 89.638 55,000 acres.. 2 422 ..do_ miles.. 96 3.64 3, ..cubic yards., square yards.. ..cubic yards.. 49,215 479,311 330,371 100.00 193.00 4.00 1.10 242,200 306,528 40,560 4,252 196,860 527,242 82,593 ...feet B. M. .do_ _ _do_ cubic yards.. .pounds.. 158,400 9,900,000 3,980,800 171,000 1,223,760 a 50.00 a 23.00 a 30.00 .60 .03 7.920 227,700 119,424 102,600 36,713 721,484 100,000 cubic yards.. ...feet B. M_. ..linear feet.. _feet B. M — _pounds.. cubic yards.. .do_ 19,902 1,191,000 163,320 262,800 102.000 53,400 3,700 6.00 a 22.00 20 a 33! 00 .03 .50 . 15 119,412 26,202 32,664 8,673 3,060 26, 7 0 555 25, IKK) 160,698 7,696,730 a Per 1,000 feet DEEP WATERWAYS 559 Table No. 10.— Oswego-Mohawk route, eastern division—Estimate of construction of 21-foot channel —Continued. SECTION NO. 8 (STATION G800 TO STATION 7732), FROM FULTONVILLE TO ROTTERDAM JUNCTION. Excavation: Rock,extra for lock... Rock, dry.. Rock, wet..... Rock, artificially dry.._ Clay... Gravel.-.... Sand. Quicksand.. Earth... Earth, extra for locks. Diversion.. Right of way: Farm land. Town land.-. Railroad changes... Entrance of streams..__ Retaining wall....... Slope wall.... Back fill. Timber crib: Oak. Hemlock____ Pine. Stone fill..... Iron bolts.-... Lock No. 17... Operating machinery. Lock No. 18. .. Operating machinery. Dam at Amsterdam:" Concrete, dam and wing wall Excavation, earth ___ Cofferdam. Dam at Cranesville: Concrete. Timber in foundation. Piles in foundation.. Sheet piling foundation. Iron.. Excavation, earth. Cofferdam. Bridges .. Steam ferry... Total Quantity. Cost per unit. cubic yards.. .do — ....do_ .do_ ..do_ .do_ .do_ ..do_ ..do_ ...do_ 29,139 $0.60 398,163 . 65 612,927 2.00 395.271 .75 6,293.037 .18 13, 755,344 .18 15,251,930 .18 176,790 .18 6,316,712 .18 245.871 .32 Total. §17,483 258,806 1,225,854 296,453 1,112,747 2,475,962 2,745,347 31,822 1,137,008 78,679 47,000 acres. . ..do ... miles.. ..cubic yards.. square yards.. ..cubic yards.. 3,168 61 3.99 100.00 7,486.00 66,717 4.00 623,737 1.10 279,672 .25 _feet B. M.. .do_ .do_ .cubic yards.. .pounds.. 272,980 17,238,000 6,626,740 271.010 1,984,060 .cubic yards. .do... .cubic yards. -feet B. M. ..linear feet. _feet B. M. .pounds. .cubic yards. 35,812 72,800 1 i , 325 1,116,720 154,140 249,600 100,000 40,800 a 50.00 a 23.00 a 30.0G .60 .03 6.00 .50 6.00 a 22.00 .20 a 33.00 .03 .50 316,800 456,646 105,893 2,643 266,868 686,111 69.918 13,649 396.474 198,802 162,606 59,522 648.475 100,009 765,818 100,000 214,872 36.400 17.800 103,950 24,567 30,828 8,237 3,00 i 20.400 11,000 523,693 20,000 14,812,133 SECTION NO. 9 (STATION 7732 TO STATION 8043), FROM ROTTERDAM JUNCTION TO 1 MILE NORTH OF SOUTH SCHENECTADY. Excavation: Rock, dry. . Hardpan... Clay.. Gravel. Sand.. Earth.. Earth, extra for lock. Right of way: Farm land. Town land. Retaining wall .. Core wall in embankmen .s_ Slope wall. Back fill.... Embankment. . Timber crib: Oak.... Hemlock.... Pine... Stone fill.. Iron bolts. . Lock No. 19.... Operating machinery. Dam: Concrete, dam and wing wall Concrete, core wall. .cubic yards. ..do .. .do... . do... .do... .do... .do... .acres. ....do... ...cubic yards. ..".do .. .square yards ...cubic yards. .do... _feet B. M. .do... .do... .cubic yards. _pounds. _cubic yards. ..do... « Per 1,000 feet. 333,953 $0.60 §200,372 488,537 .30 146,561 994,497 . 18 179,069 1,325,996 .18 238,679 1,657,495 .18 298,349 1,988,989 .18 358,018 206,095 .32 65,950 1,593 100.00 159,300 36 2,600.00 93,600 3,818 4. 00 15,272 15,417 4.50 377 197,85!) 1.10 217,645 158,169 . 25 39,542 1,033,426 .15 155,014 93,336 a 50.00 4,667 5,354,740 a 23.00 123.159 1,568,140 a 30.00 47,014 605,394 .60 363,236 89,246 .03 2,677 629,650 100,000 32,040 6.00 192,240 9,000 4.50 40,500 5(50 DEEP WATERWAYS Table No. 10. — Osicego-Mohciwk route, eastern division—Estimate of construction of 21-foot channel —Continued. STATION NO. 9 (STATION 7732 TO STATION 8043), FROM ROTTERDAM JUNCTION TO 1 MILE NORTH OF SOUTH SCHENECTADY—Continued. Quantity. |°^5» Total. Dam—Continued. Timber in foundation..feet B. M.. Piles in foundation.linear feet.. Sheet piling in foundation.....feet B. M_. Iron.....-...pounds.. Excavation, earth..cubic yards.. Embankment.....do — 1,406,600 218,610 283,200 132,000 75,825 32,900 a$22.00 .20 «33.00 .03 . 50 .15 $30,945 43,722 9.346 3,960 37,913 4.935 14,090 156,602 2 Total ...... 4,041,284 SECTTON NO, 10 (STATION 8043 TO STATION 8423), FROM 1 MILE NORTH OF SOUTH SCHENECTADY TO FRENCH MILLS. Excavation: Rock.— Hardpan . Clay.-- Gravel. Sand. Earth . Rock, long haul. Clay, long haul. Gravel,long haul.... Sand, long haul- Earth, long haul. Quicksand, long haul Right of way: Farm land. Town land. Entrance of streams_ Retaining walls_ Slope walls.. Back fill.. Timber cribs: Oak. Hemlock. Pine .. Stone fill.. Iron bolts. Lock No. 20. Operating machinery Dam: Concrete dam. Cofferdam. Bridges. Total. cubic yards.. .do_ ..do_ ..do_ _do_ _do.... .do_ ..do_ ..do. .. ..do_ .do.... .do_ .acres.. .do_ ..cubic yards., square yards.. ..cubic yards.. .feet B. M-. .do_ .do_ ..cubic yards.. .pounds.. cubic yards.. 238,237 $0.60 8142,942 3,645,077 .30 1,093,523 470,334 .18 84,660 135.167 .18 24,330 954, 660 .18 171.839 149.159 .18 26,849 5,935,815 . 60 3,561,489 2.954,860 .17 502,326 3,357,607 .17 570.793 13,986.595 .17 2,377,721 1,816,949 .17 308,881 3,123, 702 .17 531,029 1,516 UK). 00 151,600 7 500.00 3,500 30,294 18,682 4.50 84,069 118,512 1.45 171,842 284,666 .25 71,167 86,160 a 50.00 4,308 5,382.449 a 23.00 123.796 1,060,9! tO a 30.00 31,827 81,750 .60 49,050 557,880 .03 16.736 577,209 100,000 1,548 6.00 9,288 6,000 8 609,278 11,436,346 SECTION NO. 11 (STATION 8423 TO STATION 8663), FROM FRENCH MILLS TO ROAD TO YOORHEESYILLE., Excavation: Rock, dry. Hardpan, dry. Right of way, farm land Slope wall.. Back fill. Timber crib: Oak. Hemlock. Pine. Stone fill.. Iron bolts... Lock No. 21. Operating machinery... Locks Nos. 22 and 23. Operating machinery ... Dam: Concrete dam. Excavation, earth ... Concrete dam. Excavation, earth .. Bridges. Total... ..cubic yards.. ..do_ .acres.. square yards.. ..cubic yards.. 160,883 4,151,268 1,019 14,658 212,725 §0.60 .30 100.00 1.45 .25 S96,530 1,245,380 110,900 21,254 53,181 ...feet B. M.. ..do ... ..do_ cubic yards.. _pounds.. 149,520 9,862,660 1,891,000 142,010 1,051,620 a 50.00 a 23.00 a 30.00 .60 .03 7,476 226,811 56,730 85,206 31,549 728,626 100,000 2,172,165 175,000 cubic yards.. .do_ ..do_ .do_ 9,601 29,800 60,696 99,000 2 6.00 .15 6.00 . 15 57,606 4.470 364.176 14,850 162,180 714,120 a Per 1,000 feet. DEEP WATERWAYS 5G1 T able No. 10.— Oswego-Mohawk route, eastern division—Estimate of construction of 21-foot channel —Continued. SECTION NO. 12 (STATION 8063 TO STATION 8923), FROM ROAD TO VOORHEESVILLE TO 1 MILE WEST OF NORMANSVILLE. Quantity, j Total. Excavation: Clay. cubic yards.. Sand .. .......do — Right of way, farm land ... acres.. Slope wall.....square yards.. Bridges . .. ... .. . 3,292,919 1.097,639 4,334 44,479 o SO. 15 . 15 100.00 1. 45 $493,938 104,640 433,4. l( i 64 495 102,180 Total ... .. 1,318,659 SECTION NO. 13 (STATION 8923 TO STATION 9106 + 18.5), FROM ROAD TO VOORHEES¬ VILLE TO HUDSON RIVER. Excavation: Rock, hard,dry. Do . Clay . Sand. Earth . Right of way: Farm land .. Town land .. Railroad changes. .. Retaining walls.. . Slope wall. . . Back fill. .. . Timber crib: Oak . Hemlock .. Pine . Stone filling . Iron bolts .. . Locks Nos. 24, 25, and 20 . Operating machinery . Locks Nos. 27,28,29,30, and 31 Opei'ating machinery _ Dam: Concrete dam . Excavation, earth . Concrete . Bridges . Total . cubic yards.. .do.... .do_ .do_ .do- 2,049,245 $0.05 $1,332,009 222,083 . 00 133.250 5.532,532 .15 829,880 446 . 15 57.907 278,737 .15 41,811 .acres.. .do_ .miles.. ..cubic yards.. square yards.. ..cubic yards.. 585 100. (X) 13 3,846.00 2,630 . 4.50 39,234 1.45 311.823 .25 58,500 50,000 119,902 11,802 50,889 77,956 ...feet B.M.. . .do.... . do _ cubic yards. . pounds.. 188,880 17,527,041 2,275.805 260,472 1,593,970 a 50.00 a 23.00 a 30.00 .00 .03 9,444 403,122 08,274 150,283 47,819 3,065,295 225,000 4,533,7(9 350, (M10 cubic yards.. .. do _ .do _ 44,495 6.00 34,800 .15 0, 750 6. (X) 266.970 5,235 40,500 179,499 121,236 a Per 1,000 feet. SUMMARY. Section. Station. Total cost. 4789+91.5 to 4848 $421,722 3,576,150 5,195,364 708,590 5,372,730 5,223,064 7.096,730 14. 812,133 4,041.284 11,436,346 5,714.120 1,318,659 12.121,236 ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 4848 to 5118. 5118 to 5175 . 5175 to 5260 . ....... 5260 to 5051 .. ... 0103 to 0800 .. 6800 to 7732 7732 to 8043 8043 to 8423 .. 8423 to 8003 8003 to 8923 8923 to 9100+18.5 Grand total. .. . . . 77,716,740 Ill closing this report I desire to express to each member of the United States Board of Engineers on Deep Waterways my full appre- H. Doc. 149-36 DEEP WATERWAYS. 562 ciation of the universal courtesy at all times extended to me during the progress of this work, and the full liberty given in the selection of my assistants, and discretion allowed in obtaining the information desired. I also desire to take this occasion to state my indorsement and hearty cooperation in the wisdom of the suggestions and instructions as issued from time to time by this Board, as so greatly facilitating the obtaining of the results desired. In view of the large number of men employed at different times, special mention can not be made of each, but I desire to thank each one and state my full appreciation of the value of the services ren¬ dered and the care, zeal, energy, and interest in the work as displayed by each employee, and which so largely contributed to, I think I can state, its successful completion. The names of those who tilled the more important positions in the field and office are: John McCornb, James J. Overn, J. W. Paxton, E. M. Durham, jr., C. E. Pelz, and E. C. Reynolds, instrument men; A. D. Raymond, level man; W. H. Breen, superintendent of borings; Oliver B. Harden and H. Tracy Fisher, draftsmen, and Heningham Gordon, recorder. Respectfully, D. J. Howell, Assistant Engineer. The Board of Engineers on Deep Waterways. Appendix No. 15. TIDAL HUDSON. Detroit, Mich., August 4, 1899. Gentlemen : I herewith very respectfully submit to you this report pertaining to surveys and borings of the Hudson River and approxi¬ mate estimate of the cost of improvement from the State dam at Troy, N. Y., to the Atlantic Ocean. First. For a channel varying from 300 to 000 feet in width, with a depth of 21 feet below low water. Second. For a channel varying from 300 to 600 feet in width, with a depth of 30 feet below low water. The Hudson River is an arm of the sea, extending inland 150 miles, and the general direction is north and south. Extensive improvements have been made in the past by the State of New York and also by the Federal Government, and at present a 12-foot channel is maintained between New Baltimore and Broadway, Troy, N. Y. Between New Baltimore and Hudson City no improve¬ ments were necessary to maintain the 12-foot channel, but a dike has been constructed by the Federal Government closing up the channel DEEP WATERWAYS. 563 west of Bronk Island, and the heads of Light-house and Coxsackie islands have been protected from abrasion by pile dikes. During the dry season the stage of the river is largely dependent upon the tidal flow, and the ebb and flood currents are well defined in their proper directions. The State dam at Troy, N. Y., is a division between the tidal and nontidal sections of the river, and during the dry season of 1897 the mean rise and fall of tides at that point was 1.4!) feet. During ordi¬ nary stages of the river above the dam, when the river discharges an average quantity of water, the flood currents are not so well defined, but the ebb currents are still plainly marked. During freshets the tidal action is lost sight of entirely, and during ordinary spring fresh¬ ets the river rises from fi to 8 feet above mean low water. The highest recorded stage of water at Albany, X. Y., caused by rainfall alone, occurred in October, 1869, the rise in the river being 19 feet above mean low water. The highest known stage of the river at Albany occurred in February, 1857, and was caused by an ice gorge at Van Wies Point, and is recorded at 22 feet above mean low water. The greatest known depth of water in recent years on the crest of the dam at Troy, N. Y., occurred March 1, 1896, and is recorded as 9.92 feet, making an elevation of high water of 23.4 feet. It will not be necessary, however, to dwell upon the high-water problem below the State dam, since the deepening of the channel would relieve that condition to a great extent. LOW WATER. It is problematical how much the increased area of channel will decrease the stage of mean low water at the State dam in Troy, X. Y. During an extreme low stage of the Hudson River no water runs over the dam at all, and the only supply which the tidal section of the river receives during that time is water supplied by lockage from the canal, leakage, and through the water wheels. During this time the discharge is approximately between 3,000 and 4,000 cubic feet per second. The plane of mean low water of 1897 at the State dam is 0.37 foot lower than the mean low-water plane of 1876. At Green Island a difference of 0.59 foot is recorded between the two planes in the same direction, and at Albany the mean low-water plane of 1897 is 0.05 foot lower than the plane of mean low water of 1876. In establishing the grade line upon which this estimate is based I assume that the elevation of extreme low water at the State dam, after the improvements are made, will be +1.23 feet and at Green- bush + 0.24 foot. The distance between these two points is 8.24 miles, making the slope 0.12 foot per mile, approximately. At Stuy- vesant, the elevation of low water is —0.18 foot, and the low-water stage there will not be affected by the improvements contemplated. DEEP WATERWAYS. 564 The distance between Greenlmsh and Stuyvesant is 17.22 miles, and the total slope is 0.42 foot between these two points. Between Stuy- vesant and Hudson City the stage of low water will not be affected by any improvements, and quantities were computed for a 21 and 50 foot depth of channel, respectively, below this plane. Very reliable elevations of low water could not be obtained below Coxsackie during the time the survey was made, on account of the continuous freshets which affected the river in the autumn of 1808 and spring of 1899. In May, 1899, elevations of —0.77 foot were obtained by staff-gauge leadings at Hudson City during ebb tide, and approximate quantities were based on this elevation and the elevation, —0.18 foot, at Stuyvesant. It will be necessary, however, to investigate the stage of low water more thoroughly, and if necessary modify the grade line upon which this approximate estimate is based before the contemplated improve¬ ments are undertaken. TIDES. [Deduced from information obtained from Mr. R. H. Talcott, United States assistant engineer, Albany, N. Y. J Location. Mean high tide. Mean tide. Mean low tide. Mean rise and fall of tide. Year. State Dam.- +4.64 +3.90 +3.15 1.49 1897 Covill Folly Light. +4.29 +2. S3 +1.36 2.93 1897 Albany..... +4.09 +2. 67 +1.24 2.85 1897 Castleton ....... +3. 61 +2. 30 +0.99 2.62 1876 New Baltimore _ . . . +3. 27 + 1.67 +0.07 3.20 1876 Stuyvesant Light.._ ___ __ +3.02 + 1.43 -0.18 3.20 1876 Hudson ......... +3 25 +1.25 -0. 75 4.00 1899 Germantown .. .. - -.-. +3.20 + 1.20 —0.80 4.00 1899 SURVEYS. In general, the method employed in making the survey was in accordance with the instructions to field parties, Appendix No. 9. Before starting on the survey, observations were made on Polaris and a true azimuth established. A base line was then run from the Congress Street Bridge at Troy, N. V., starting with station 0, at the origin of the base line for the Oswego route, eastern division. This line was carried across the Hudson River to the Troy side and extended south to a point about 14 miles below Hudson City, and connected with triangulation point 273, commonly known as “ Wisnall,” of the New York State triangulation system, which is equal to base line sta¬ tion 18854-94.03 of the Hudson River survey. As the base line was advanced observations for azimuth were made, whenever possible, every 5 miles, corrections for easting and westing were applied, and instrumental error found and distributed. The latter, however, never exceeded 3 feet and generally fell within 1 foot. After the base line had been extended for some distance the levels were carried along by the same party, benches being established DEEP WATERWAYS. 505 every mile, approximately, and then the shore-line survey was taken up. The latter consisted entirely of a stadia survey, the closing error for distance being limited to 1 foot in 500 feet and error in elevation 0.5 foot per circuit. No difficulty was found in keeping well within these limits. All stadia circuits were connected with the base line, and coordinates of stadia hubs were computed and referred to the origin of the circuit. The shore-line survey extended from the State dam at Troy, N. Y., south to a point about 14 miles below the city of Hudson, the total distance being approximately 37 miles. From the city of Hudson to Livingston Creek, a distance of 5.1 miles, a 30-foot channel already exists, as indicated on the Coast Survey charts, and no detailed sur¬ veys have been made covering this stretch. The survey was taken up again, however, at Livingston Creek, and extended south a dis¬ tance of 4.4 miles to a point below Germantown. Covering this stretch soundings were taken on ranges 500 feet apart, but no exten¬ sive borings were taken. Occasionally the bed of the river was tested by using a half-inch gas pipe, pointed on one end, and forcing it down by hand to a distance several feet below the bottom of the proposed channel. The material encountered was a tine sand. The river was also examined near Barrytown and Rhinebeck. At Barrytown an estimate was made for dredging a channel about 3,(500 feet long across the bar between the east and west channels for the 30-foot canal. No work is required for the 21-foot canal. At Rhinebeck the required depth was found, and no improvements will be neces¬ sary. An estimate was made for widening and straightening the 30-foot channel near Sycamore Point. The data for this purpose was taken from the Coast Survey charts. LEVELS. All elevations in connection with this survey are referred to the plane of mean tide at Sandy Hook, and depend upon the elevation + 14.73 of the Greenbush bench mark. The method employed in transferring levels from this bench was as follows: Duplicate lines of levels were run, backsights and foresights never exceeding 250 feet in length, and if the error exceeded 0.05 feet x V distance in miles between bench marks, or from the origin of levels to any bench mark, the levels were rerun to bring the error within the prescribed limits. SOUNDINGS. Soundings were taken on ranges approximately 300 feet apart, established by the sounding party, and subsequently located by the stadia party. 566 DEEP WATERWAYS. The method employed in taking soundings was as follows: A flagman was placed at one end of the range, and it was his duty to keep the boat on range, as nearly as possible. Soundings were taken at from ten to thirty seconds intervals, depending upon the depth of the water, and were located by azimuths taken at minute intervals. Only one instrument was used, the location of the sound¬ ing being the intersection of the azimuth line with the range line. Generally a 10-pound lead was employed in taking soundings. This method was very satisfactory. The boat used for that pur¬ pose, being supplied with a rudder, could be kept on range very closely, and the error in locating soundings would not exceed the limits of the plotting. BORINGS. In order to ascertain the character of material to be excavated, extensive borings have been made at intervals of 1,000 feet or less, and in every case where borings were located within the limits of the proposed channel they are carried below the depth of excavation or to the rock surface. The rock found in the Hudson River is what is known as Hudson River shale. The other material met with in the river consisted of bowlders, gravel, coarse and fine sand, clay, and silt, I think that all the material to be removed, with the exception of the rock and bowlders, can be handled by pumps. In all, 1,385 borings were made, with a total of 28,965 feet of driving, at a cost of 25.11 cents per linear foot, including the cost of plant. Borings were taken on the sounding ranges and were located by intersections, stadia dis¬ tances, or by direct measurements from the ends of the ranges. Nearly all the borings relating to the investigation of the Hudson River below the State dam at Troy, N. Y., were taken from a cata¬ maran and scows constructed for that purpose, and the method employed was as follows: The crew for each boring outfit generally consisted of 1 foreman and 3 laborers. Two and one-half inch pipe, commercially known as “ 2^-inch R. H. flush-joint casing,” manufactured in lengths from 1 to 10 feet, was lowered to the bed of the river, and hollow rods, com¬ monly known as “I> drill rods,” with a cross bit attached, were inserted into the casing, the upper end of the rod being connected with a hand force pump, which was kept on the scow. It was the duty of one man to work the pump and force water through the drill rods, while the other worked the drill by hand. The drill loosened the material, which was then forced upward through the casing. As the drill advanced into the bed of the river it was followed by the casing. Quite frequently very formidable obstructions were encountered by the drills, such as large cobblestones and bowlders, and more forcible measures had to be resorted to in order to pass them. When that DEEP WATERWAYS. 5(37 was the case the drill rods were removed from the casing and one-half to two sticks of 40 per cent Atlas powder were lowered into the hole, having previously been connected with a battery on the scow. The casing was then hoisted from 4 to G feel above the bottom of the hole and the powder exploded. This plan would generally remove the obstruction and the work could be continued until another bowlder was met with. In some of the bore holes between Troy and Albany it would require from ten to thirty shots to advance below the depth of exca¬ vation necessary for a 30-foot channel, or to the surface of bed rock. Samples were taken when the flow through the casing indicated that a different stratum was encountered, or when it was considered necessary. The total number of samples collected is 348. The rock found in the Hudson River is a hard shale and dark in color, and in nearly every case when this rock was encountered small fragments would be forced through the casing; or, if the flow after shooting was lost, small pieces or fragments of the rock would be brought to the surface on the end of the drill. PLOTTING. All notes were plotted in the field office as the work progressed. The scale of the maps was 1: 5000. On the completion of the field work, a force sufficient to estimate the work necessary to provide 21 and 30 foot channels was transferred to Detroit. ALIGNMENT. In locating the center line of the proposed channel it was the object to confine it as close as possible within the channel line as it now exists, to avoid heavy rock cutting, whenever possible, and limit the degree of curvature. Following is a tabulated statement of alignment, including total length of curves and tangents, total curvature and maximum degree of curvature: Length. Maximum degree of curvature. Total cur¬ vature. Curve. Tangent. Feet. 69,523 Feet. 1H,9S2 o / 1 16 O / 485 50 WIDTH OF CHANNELS. The proposed width of the channel upon which these estimates are based is as follows: Feet. From station 5241-1-18 to station 5539+40..... 800 From station 5539+40 to station 5905+80 .......400 From station 5905+80 to station 7109 (end). .... 600 568 DEEP WATERWAYS. CROSS SECTION. The cross sections are of the standard form adopted by your Board for river improvement. The side slopes in rock cuts are 10 on 1 and in earth cuts 1 on 2. BRIDGES. The river between the State dam at Troy, N. Y., and the ocean is obstructed by six bridges. DESCRIPTION OF BRIDGES. - Draw- span. Width in clear in each draw. Height of bottom chord abo ve mean high water. (a) Delaware and Hudson Railroad bridge at Bridge avenue, Troy. Ft. in. Ft. in. Ft. in. N. Y ... 198 0 01 6 23 0 (b) Congress street bridge, Troy, N.Y.. (c) New York Central and Hudson River Railroad, upper bridge, 258 0 104 0 29 6 Albanv, N. Y...— (tf) New York Central and Hudson River Railroad passenger 275 0 110 7 33 5 bridge, Albany, N.Y. a .... 275 0 115 4 28 1 (e ) Lower bridge, or Green bush bridge. Albanv, N.Y-—- (/ ) Poughkeepsie bridge at Poughkeepsie, N. Y. 400 0 109 8 22 7 100 0 a Since this report was written this bridge has been replaced by a new one. (a) Railroad, wagon, and passenger bridge. Built of iron, on stone piers, drawpier on land at east shore. The only opening through which boats ply in 36 feet 6 inches in clear from dock line, east shore. The draw is opened by steam power. There are four piers in the main channel of the river east of Starbucks Island, the respective widths in clear between piers being 170 feet 6 inches, 173 feet 6 inches, 173 feet 6 inches, and 160 feet to east shore line of Starbucks Island. In channel of the river west shore line of Starbucks Island to first pier, 150 feet 5 inches; between first and second piers, 177 feet 5 inches, and from second pier to Green Island dock line, 68 feet 6 inches, making in all six piers in the river. The bridge stands at right angles to the direction of the stream. It was built in 1876 by the Delaware Bridge Company. (b) A wagon and passenger bridge built of iron on stone piers, and slightly oblique to the direction of the stream, with double draw opened by hand power. There are two piers and the drawpier in the river. Width in clear from east pier to dock, 205 feet 3 inches; width in clear from west pier to West Troy shore line, 210 feet. Built in 1874. (c) A railroad freight bridge, built of iron on stone piers, and at right angles to the direction of stream. It has a double draw, opened by steam power. There are four piers and drawpier in the river; respective widths in clear between bulkhead lines, 146, 170, and 170 feet, commencing at Columbia Pier dock line. Built in 1870. DEEP WATERWAYS. 569 {d) A railroad and passenger bridge, built of iron on stone piers, at right angles to the direction of the stream. It has a double draw, opened by steam power. There are four piers and a drawpier in the river; respective widths in clear, 146 feet 5 inches, 175 feet, 175 feet, and 119 feet, commencing at Columbia Pier dock line. Built in 1875. (e) A wagon and passenger bridge, built of iron on stone piers, at right angles to the direction of the stream. It has a double draw, operated by steam power. There are two piers and drawpier in the river; width in clear from west pier to Quay street dock line, 186 feet 3 inches; width in clear from east pier to shore line pier, 236 feet 3 inches. Built in 1886. (/) A railroad bridge on iron piers, with stone foundations, at right angles to the stream. There are five spans, but no drawspan. and four piers in the river. Respective lengths of spans, 548, 525, 546, 525, and 548 feet, cantilever type. DAM AT TROY, N. Y. The legal height of crest of dam is 12.07 feet above mean low tide of Hudson River at Albany, N. Y. Crest of dam is 8 feet below top of masonry of sloop lock. The total length of weir is 1,100 feet. DEPOSIT OF WASTE MATERIAL. Generally all material excavated will have to be deposited either on the shore or behind established lines, and so protected that during freshets it is not carried into the channel; or deposited in the river at such points as may be selected by the engineer in charge of the improvements. This matter can be decided upon, however, when the contemplated improvement is undertaken. CHARACTER OF MATERIALS TO BE EXCAVATED. As previously stated, the rock on this division consists entirely of Hudson River shale and must be excavated under water. From the Troy dam to station 5659 + 50 the material above the rock consists of bowlders, stones, sand, gravel, and some clay. Bowlders form nearly the entire covering of the rock for the first 6,000 feet below the dam. For the remainder of the distance to station 5659 + 50 the finer materials predominate. From station 5659 + 50 to the end of the work the excavation above the rock will consist entirely of sand, fine gravel, and some clay, and can be readily handled by a pump dredge. QUANTITIES. The following tables give the estimated quantities and cost for both the 30 and 21-foot channels. The quantities from the State dam to H. Doc. 149-36i 570 DEEP WATERWAYS the lower end of the approach to lo£k No. 1-, station 5235, are included with those for the Hudson River division, Appendix No. 10. THIRTY-FOOT CHANNEL. [Troy to Albany, station 5235 to station 5659 + 50.] Quantity. Cost per unit. Total. Excavation: Earth, wet.......cubic yards.. Rock, wet....—do- f 762,117 \ 8,201,433 2.429.161 $0.30 .25 2.00 S228,635 2,050,358 4,858,322 7,137,315 [Albany to junction with Normans Kill line, station 5659 + 50 to station 5757.] Excavation, earth, wet.cubic yards. Excavation: Earth, wi Dikes: Total. .cubic yards.. 3,439,411 $0.15 $515,912 •ater, station 5757 to below Sycamore Point.] cubic yards.. 48,942,073 945.896 $0.15 2.00 $7,341,311 1.891,792 ..linear feet.. .cubic yards. . 886,667 494,400 261,208 67,660 .15 a 30.00 .03 . 75 133,000 14,832 7,836 50,745 S. 439,516 SUMMARY. For Champlain route only: Troy to Albany. ..... $7,137,315 Albany to Normans Kill... 515,912 Total. .'... 7,653,227 Common to both Champlain and Oswego-Mohawk routes, Normans Kill to deep water. 9,439,516 TWENTY-ONE FOOT CHANNEL. [Troy to Albany, station 5235 to station 5659 +50.] Quantity. Cost per unit. Total. Excavation: Earth, wet.cubic yards.. Rock, wet ........do.... Total....... / 715,768 \ 4,697,864 744.709 $0.30 .25 2.00 $214,730 1,174,486 1,489,418 2,878,014 [Albany to junction with Normans Kill, station 5659+50 to station 5757.] Excavation, earth, wet...cubic yards.. 1,920,318 $0.15 $288,048 [Junction with Normans Kill line to deep water, station 5757 to below Sycamore Point.] Excavation: Earth, wet..cubic yards.. Rock, wet........do_ Dikes: Piles. ...linear feet.. Pine...feetB. M._ Iron.......pounds.. Stone...cubic yards.. Total... 22,353,773 111,126 886,667 494,400 261,208 67,660 $0.15 2.00 .15 a 30.00 .03 .75 $3, &53,066 222,252 133,000 14,832 7,836 50,745 3 ,4 1 31 a Per 1,000 feet. DEEP WATERWAYS. . 571* SUMMARY. • For Champlain route only: Troy to Albany....... «■. ... $2,878,614 Albany to Normans Kill.-... .. 238,048 Total......... 3,166,662 Common to both Champlain and Oswego-Mohawx routes, Normans Kill to deep water. 3,781,731 Iii conclusion, I wish to acknowledge the faithful services performed by my assistants, among whom should be mentioned E. J. Thomas, instrumentman; A. L. Harris and A. N. Dunaway, draftsmen; A. W. Clark, recorder, and Paul Beer; also the many courtesies received from Mr. Frederick W. Orr, of Troy, and li. II. Talcott, United States assistant engineer at Albany, N. Y. Respectfully submitted. II. F. Dose, Assistant Engineer. The Board of Engineers on Deep Waterways, 4 r ■% ♦