ILLINOIS AND ST. LOUIS BRIDGE COMPANY.
REPORT
OF THM
CHIEFI ENGINEER.
OCTOBER, 1870.
ST. LOUIS:
GEORGE KNAPP & CO., PRINTERS AND BINDERB.
1870.








CHIEF ENGINEER'S REPORT.
ST. Louis, October 1st, 1870.
To the President and Directors of the St. Louis and Ilinois Bridge
Company:
GENTLEMEN: I have the honor to submit the following report:
THE WEST ABUTMENT.
The masonry of the west abutment has been completed from
the bed-rock of the river to a point thirty-one feet above lowwater mark.   From  the bed-rock to the top of the masonry,
the height is now forty-four feet.  Eight courses of granite are
laid on it and two courses, or five feet, are required to complete it to the skewbacks against which the steel ribs of the
western span will rest. This mass of solid masonry stands upon
the lower edge of the wharf in St. Louis, and measures at its
base, in the direction of the river current, ninety-four feet, and
transversely sixty-two feet nine inches, and contains at this time
six thousand three hundred and eighty cubic yards of masonry.
When completed to the carriage-way, it will be one hundred
and fifteen feet high, above the bed-rock of the river, and will
then contain eleven thousand eight hundred and sixty cubic yards
of masonry.
The work on it is going steadily forward in a satisfactory
manner.
Although the bed-rock at the site of this abutment is seventythree and a half feet higher than at the east pier, the difficulties
encountered in building its foundation were of a much more perplexing and tedious character than those encountered at either of the
others. Its site had been for over sixty years a part of the steamboat wharf of the city, and as such had received every kind of
useless material thrown overboard from the various steamers




4                  THE WEST ABUTMENT.
lying over it during that time.  The old sheet-iron enveloping
their furnaces, worn-out grate-bars, old fire-bricks, parts of smokestacks, stone-coal cinders and clinker, and every manner of things
entering into the construction of a Mississippi steamer seemed to
have found a resting place at this spot, and constituted a deposit
averaging twelve feet in depth over the rock.  During the memorable fire of 1849,when twenty-nine steamers were destroyed at the
levee, the wrecks of two of them sunk upon the site of this abutment. One of these was partly covered by the hull of the other,
which probably sunk immediately afterwards.  The lower one
was but two or three feet above the bed-rock. After this terrible
conflagration, the city authorities determined to widen the wharf.
Its front was extended to a line inclosing about one-half of these
two wrecks, by filling in with stone and rubbish from the city. During this extension, several other vessels were burnt at the wharf,
and the wreck of one of these also sunk upon the site of the
abutment. The coffer-dam, constructed to enclose the site, had to
be put down through these three wrecks, the hulk of either of which
was not probably less than four hundred tons measurement. Their
bottom planking was all of oak, three or four inches in thickness.
To drive the sheet piling down through these hulks, an oak beam
six by ten inches square, armed with a huge steel chisel, was first
driven down as far as a steam pile-driver could force it. It was
then withdrawn and a sheet-pile, five by ten inches square, was
driven down in its place.  The coffer-dam was formed of two
courses of sheet-piling, six feet apart, which were filled in between with clay. When this was completed, the water pumped
out and the excavation prosecuted within it, the discovery was
made that from one-third to one-half of the length of each of these
three steamboat hulks was enclosed within the dam, and that some
of the sheet-piling had not been driven through the lower one, owing
to the great resistance of the hulk and the mass above it. Before the
space between the lower wreck and the bed-rock could be made
secure on the inner side of the dam, the water came through and
flooded the inclosure. A stream from a powerful Gwynne pump,
having an eight-inch diameter of jet, was then directed against the
material deposited over these wrecks on the outer side of the dam,
where the water was fifteen feet deep, and enough of the deposit was
washed away to enable another course of sheet-piling to be driven
down six feet beyond the dam, through all of the wrecks to the
rock. After this, that part of the wrecks enclosed between this




THE EAST PIER.                      5
last course of piling and the dam, was removed by a diver and the
space filled in with clay, and the inclosure again pumped out.
This portion of the dam, about fifty feet in length, was by this
construction, made double. As the excavation within progressed,
it revealed the fact that another portion of the dam had been built
and made water-tight through and over a water wheel of one of
the wrecks. The crank of an engine of seven feet stroke attached to the head of the shaft of the wheel, was just within
the inclosure, while the flanges, arms and braces of the wheel were
within the walls formed by the sheet-piling. From the inclosure
within the dam were taken parts of several old.and burnt steamboat engines, the iron parts of some of which had to be cut off at
the dam. Four wrecks of barges, some of them in use doubtless
before the era of steam, were also found within it; likewise several
oak sawlogs, some anchors, chains and a great variety of smaller
articles lost or thrown overboard from the river craft, or dumped
in from the city.
This incongruous deposit made it exceedingly difficult to maintain the integrity of the dam, which at times had to resist a
pressure of thirty feet of water. Frequent floodings consequently
occurred, which delayed and increased the cost of the work.
These difficulties were however, finally overcome, and the bed-rock
within was at last exposed to view.
On the 25th day of February, 186S, after thoroughly testing the
solidity of the rock by drilling, the first stone of your Bridge was
laid in this abutment fifty-five feet below high water mark, about
four months after commencing the construction of the dam.
THE EAST PIER.
The caisson for sinking the east pier described in my published
report, September 1st, 1869, having been completed, and the
requisite guide piles, air and sand pumps, hoisting machinery, etc.,
being made ready, it was launched from the ways at Carondelet, on
the 18th of October, 1869, and was towed up to its position in
the river and duly secured within the guide piles on the same day.
Some days more were occupied in securing the caisson to the
suspension screws that were to steady it until it reached the sandy
bed of the river, and in connecting the various air and water-pipes
with it. On the 25tb day of October, the first stone was laid upon




6                     THE EAST PIER.
it, from which moment until it reached the bed-rock of the river,
on the 28th of February last, (one hundred and twenty-eight feet
below high water mark, or one hundred and twenty feet below the
city directrix,) the progress of its descent and the working of the
machinery connected with it, were marked by an almost total
freedom from accident.*
During low water the normal depth of sand over the bed-rock at
this pier is about eighty feet. A rise in the river causes it to scour
down, whilst a subsidence of the flood permits the moving sands
from above to deposit rapidly and again raise the sand bed. At
the time of placing the caisson in position the water in the river
was thirty-five feet deep, and sixty-eight feet of sand were
then overlying the rock. As the caisson descended, the current
sweeping under its bottom  at the rate of about three and a
half miles per hour, caused a further scour of five feet, leaving
an irregular surface of sand, averaging about sixty-three feet
deep above the rock.    When the caisson had fairly entered
the sand a deposit was made rapidly around it, especially in the
eddy created below the caisson. The sides of the caisson were:iwept to some extent by the current, otherwise the derrick boats
on each side of it would have grounded on the deposit. This deposit
was increased by the discharge from the sand pumps, and by the
completion of the ice-breaker above the pier, so that for twenty or
thirty days before it reached the rock, the sand was visible above
the water, both above and below the caisson.  The iron walls
of the caisson had consequently a severe external pressure upon
them from this mass of sand, and as the sides of the pier were
inclined whilst the walls of the caisson were vertical, a space was
left between the two. This space was at first occupied with bracings
against the masonry to sustain the walls. As the pier descended
and the pressure increased, sand from the pumps was discharged
into it and in this way the walls were relieved; the height of sand
inside, between the caisson walls and the pier, being maintained at
the level of the sand on the outside. The design was to keep the
masonry constantly built up above water, as the pier descended.
The failure of the granite company to deliver in time the granite
which was intended to form the exterior of the pier above low water
mark, prevented me from following out this design. At the time the
* An inspection of the lithographic plates in the Appendix will give the reader a clearer
idea of the method and appliances used in sinking the east and west piers.




THE EAST PIER,                     7
granite was wanted, the river was from ten to twelve feet above
low water mark, and- there was left only the alternative of
stopping the descent of the caisson that much above the bed-rock
and awaiting the receipt of the granite, or else allowing the
pier to descend and trust to the walls of the caisson to exclude
the water above the top of the pier. Three courses of granite,
each two feet thick, were received in time, but when the pier
reached the bed-rock the top of the masonry was about six feet
below the surface of the water.
To prevent danger to the workmen in the air-chamber, in the
event of the walls of the caisson giving way under a pressure of
water and condition of things deemed possible, but not expected,
the shafts or wells through the masonry, by which access was
obtained to the air-locks which were at the bottom of the pier,
were kept securely built up above the level of the water.  On
one or two occasions leaks occurred in the caisson walls, causing
the top of the masonry to be flooded.   But this precaution
prevented a suspension of the work of filling the air-chamber
under the pier with concrete, in consequence of such accidents, as
the shafts remained free of water. During the month of April
however, the river rose to within nine feet six inches of the City
Directrix, and was nineteen feet six inches above the top of the
masonry. From the moment the caisson touched the rock up to
this time, the filling of the air-chamber with concrete had been
progressing almost without interruption. When the water reached
this point, the walls of the caisson suddenly sprung a leak, and the
pier was again flooded in a few moments. As the flooding of the
pier usually increased the leakage in the shafts very much, it was
not found practicable to continue work in the air-chamber when it
happened, if the depth over the pier exceeded a few feet. When
this accident occurred, the men were immediately signalled to come
up from the air chamber, then one hundred and ten and a half
feet below the surface of the river. This they did with entire safety.
A suspension of work on the pier was then ordered until the *ater
should subside.
Examinations of the caisson walls were made by our diver, Capt.
Quigley, and careful soundings were made to ascertain the depth
of the sand on the inside and outside of the caisson. It was then
discovered that fifty-five feet in depth of the sand on the east side
of the caisson (on the outer side) had been scoured away by the




8                    THE EAST PIER.
action of the current, the soundings showing but thirty-five feet of
sand remaining above the bed-rock. This had reversed the strains
on the caisson walls; the corresponding fifty-five feet of sand on
the inside between the pier and the caisson, had by its gravity
burst out the eastern wall of the caisson, now no longer supported
by the sand on the outside. This wall was of plate iron i of an inch
in thickness. This rupture rendered the further use of the caisson
as a coffer-dam, forever after totally impracticable. The upper
portion of it was consequently removed, and in its stead there
was secured around the pier a wooden dam, joined together in
sections corresponding to the sides and ends of the pier, and
having a large cushion near its lower edge, on the inner side, to fit
against the fourth joint of the masonry below the top course of the
pier. This work was admirably executed under the direction of
the Superintendent of Construction, Mr. W. K. McComas; the submarine work being very skillfully performed by Capt. Quigley. On
the 13th of August the water was again pumped off from the top
of the masonry.   Since this time the pier has been carried
up to the height of one hundred and seventeen feet above the bedrock on which it rests, and is now (October 1st) nineteen feet nine
inches above the present level of the river surface. The coffer-dam
has been removed from around it, and no apprehension of further
trouble from water during its completion need be entertained. It
measures at its base, in the direction of the current, eighty-two feet
in length, and transversely sixty feet, and on the top at its present
height, seventy-five feet seven inches by thirty-five feet one inch.
It now contains thirteen thousand two hundred and forty cubic
yards of masonry, concrete and brick work. The vertical wells or
shafts through it, were lined with brick work from thirteen to
twenty-two inches in thickness. These openings were carefully
filled up with concrete and the entire pier is now one solid mass of
masonry.
In addition to the almost constant presence of the Superintendent
of Construction, in charge of all the work<, it was made the duty
of three civil engineers, familiar with the plans, machinery, etc., to
give their personal superintendence to the sinking of the pier'; one
being constantly on duty, watching the progress of the work, and
keeping a record of everything of interest occurring. The necessary number of steam engineers, machinists and firemen, were
employed for keeping the engines at work night and day, and the




THE EAST PIER.                      9
machinery in perfect order, so that the work could go on without
any interruption whatever. To accomplish this desirable object,
sufficient duplicate pumps, engines, boilers etc., were provided, so
that the failure of any one piece of the machinery would cause no
stoppage in the progress of the work. The pier was placed on the
rock in one hundred and twenty-six days after the laying of the
first stone, which period included the most inhospitable season of
the year. The mason work was suspended during twenty days of
this time on account of bad weather. During fifteen days it was
impossible to tow a barge of stone to the pier on account of the
running ice.
TELEGRAPH.
When this pier had reached the depth of sixty-six feet, a telegraphic instrument was placed in the air-chamber, and a wire was
led to the office of the Superintendent of Construction on one of
the derrick boats at the pier, and also to the office of the Chief
Engineer in the city. By this means messages were transmitted
to and from the air-chamber, and between the offices of the Superintendent of Construction and that of the Chief Engineer, by which
I was, when not present, regularly advised of the progress and
condition of the work during the sinking of the pier.  The
knowledge that a means of communication with the upper world
was constantly at hand in the air-chamber, and one which was not
likely to be interrupted by any accident endangering the lives of
the workmen in it, was productive of a very salutary moral effect
upon them.
These telegraphic arrangements were courteously put up complete under the gratuitous superintendence of Col. Chas. H. Haskins,
General Superintendent of the Pacific and Atlantic Telegraph
Company. The instruments were of the kind known as' alphabetical," and were easily understood and operated by those placed
in charge of them in the service of the Bridge Company.
FILLING THE AIR-CHAMBER.
The filling of the air-chamber was executed in the most careful
and substantial manner under the immediate direction of the
Superintendent of Construction.  The preparation and disposition of the concrete were made in the air-chamber, (from one




10                   THE EAST PIER.
hundred and three to one hundred and ten and a half feet beneath
the surface of the river,) under the immediate supervision of Mr.
Rud. Wieser, C. E., Chief Inspector of Masonry, and his assistants,
Mr. Rich. Richardson and Mr. Fritz Eberley, master masons, one
or the other being constantly on duty. From frequent personal
inspection, I am warranted in saying that this part of the work is
unsurpassed in excellence by that of any part of the masonry
above water.
The filling of the air-chamber with concrete commenced on the
second day of last March, and was finished on the 27th of last May.
The working time being fifty-three days. The space filled may be
fairly stated at nearly thirty-six thousand cubic feet. The area of
the base of the pier is four thousand and twenty square feet and
the height of the chamber nine feet. The caisson was stopped as
soon as it touched the bed-rock. This was at its southwest corner.
At the northwest corner its edge was eight inches from the rock.
The northeast corner was sixteen inches and the southeast corner
eight inches. It will be seen from this that the rock was fortunately,
very nearly level. The sand beneath the edge of the caisson was
removed, the rock laid bare and the space filled carefully with concrete, the air pressure being sufficient to prevent a more rapid
infiltration of the water under the edge of the caisson than could
be managed by the pumping arrangements within it. The sand
seemed packed so firmly that no trouble was taken to barricade it
out of this space between the rock and the edge of the caisson.
When the entire edge of the caisson and the space under its
two great girders were thus concreted, the rock in its interior was
gradually cleared of sand and the concrete placed directly upon it
in layers of nine or ten inches in thickness, the closing courses
under the roof of the chamber being stoutly rammed in place.
The air-locks were then filled with the same material, and
finally the shafts. The concrete was made of broken limestone
thoroughly washed, the interstices being filled with mortar made
of equal parts of Akron cement and pure sand.
During the sinking of the pier the sand pumps designed by
me for this special purpose, gave great satisfaction and proved entirely successful. One pump of three and one-half inches bore was
found quite capable of raising twenty cubic yards of sand one hundred and twenty feet high per hour; the water pressure required to
supply the jet being about one hundred and fifty pounds per square
inch.




THE EAST PIER.                     11
PRESSURE IN THE AIR-CHAMBER.
The pressure of air in the air-chamber was very accurately determined by the depth of the caisson below the surface. Any greater
pressure than that due to the depth, caused the air to escape
beneath it, but when the caisson had penetrated into the sand to a
considerable distance, it was discovered that the water level formed
by the air under or across the bottom of the chamber was nearly
a foot lower than the bottom edge of the caisson. When the
caisson was but a few feet in the sand, the air forced its way up
by its sides in one or two currents of large volume, but as it
penetrated more deeply, the passage of the air through the sand
evidently became more difficult, and it appeared in small
bubbles sixty or seventy feet distant from the caisson.  This
retardation of the escapement of the air from beneath the caisson,
caused an increase of air pressure by which the water was held at
a greater or lesser depth below the line of the bottom of the air-chamber. The sand inclosed in the air-chamber and forming its floor was
usually one or two feet more elevated than the lower edge of the
chamber, and was entirely devoid of water, the air pressure expelling the water from it down below the edge of the caisson as
just stated. The distance to which it was thus expelled and maintained, was at no time discovered to exceed ten inches, and generally it was not over eight inches. This would give an air pressure
equal to nearly one foot more than the depth of the caisson, or
about one-third of a pound per square inch more. This difficulty
of escapement of the air through the sand was increased somewhat
by concreting under the edge of the caisson on the rock, and the
actual air pressure could then be no longer accurately determined
by the height of the water above. The pressure gauges usually
indicated a pressure of one or two pounds more than the depth of
water would give by calculation. This was caused in a great
measure however, by the friction of the air in the pipes, the gauges
being at the pumps and not in the air-chamber. A column of water
one hundred and ten feet six inches in height would be equal to a
pressure of 47.96 pounds per square inch, assuming the weight of
the water to be 62.5 pounds per cubic foot. The greatest pressure
marked by the gauges was fifty-two pounds, and it is not probable
that the pressure in the air-chamber ever exceeded fifty or fifty-one
pounds.




12                   THE EAST PIER.
EFFECTS OF COMPRESSED AIR ON THE MEN.
The first symptom manifesting itself, caused by the pressure
of the air, is painfulness in one or both ears. The Eustachian tubes
extending from the back of the mouth to the bony cavities over
which the drums of the ears are distended, are so minute as not to
allow the compressed air to pass rapidly through them to these cavities, and when the pressure is increased rapidly, the external pressure
on the drums causes pain. These tubes constitute a provision of
nature to relieve the ears of such barometric changes as occur in
the atmosphere in which we live. The act of swallowing facilitates
the passage of the air through them   and thus equalizes the
pressure on both sides of the drums, and prevents the pain.
The pressure may be admitted into the air-lock so rapidly that
this natural remedy will not in all cases relieve it. By closing
the nostrils between the thumb and fingers, shutting the lips
tightly and inflating the cheeks, the Eustachian tubes are opened
and the pressure on the inner and outer surfaces of the tympanum
is equalized, and the pain prevented.  This method must be
used and repeated from time to time as the pressure is let on,
if it be increased rapidly. No inconvenience is felt by the reaction
when the pressure is let off, as the compressed air within the
drums has a tendency to open the tubes, and thus facilitates its
escape through them; whereas increasing the pressure has the effect
of collapsing them and therefore makes it more difficult to admit the
compressed air within the cavities of the ears. It frequently occurs
however, from some abnormal condition of these tubes, as when
inflamed by a cold in the head, that neither of these remedies
will relieve the pain.  To continue the admission of compressed
air into the lock, under such circumstances, would intensify
the suffering, and possibly rupture the tympanum, therefore
the lock  tenders were   particularly instructed to  shut off
the  compressed air at the moment any one in the lock
experienced pain about the ears; and then, if it could not
be relieved by the above means, the lock was opened and the
person was not permitted to go through into the air-chamber.
Sometimes fifteen minutes were occupied in passing persons
through the first time, after which they usually had no further
trouble from this cause.
The fact that the depth penetrated by the air-chamber was con



THE EAST PIER.                    13
siderably greater than that hitherto reached in any similar work,
left me without any benefit from the experience of others, in either
guarding against any injurious effects of this great pressure upon
the workmen and engineers subjected to it, or of availing
myself of any known specific for relieving those affected by it.
When the depth of sixty feet had been attained, some few
of the workmen were affected by a muscular paralysis of the
lower limbs.  This was rarely accompanied with pain, and
usually passed off in the course of a day or two. As the penetration of the pier progressed, the paralysis became more difficult to
subdue.  In some cases the arms were involved, and in a
few cases the sphyncter muscles and bowels. The patients also
suffered much pain in the joints when the symptoms were severe.
An average of at least nine out of ten of those affected, suffered no
pain whatever, but soon recovered, and generally returned to the
work.
The duration of the watches in the air-chamber was gradually
shortened from four hours to three, and then to two, and
finally to one hoar.
The use of galvanic bands or armor seemed, in the opinion of the
Superintendent of Construction, the foremen of the chamber, and
the men, to give remarkable immunity from these attacks. They
were all ultimately provided with them. These bands were made
of alternate scales of zinc and silver, and were worn around the
wrists, arms, ankles and waist, and also under the soles of the feet.
Sufficient moisture and acidity were supplied by the perspiration to
establish galvanic action in the armor, and as the opinion among
those most accustomed to the chamber was almost unanimous in
favor of this remedy, I am very much inclined to believe it
valuable.
Immediately on the manifestation of greater severity in the
symptoms, a hospital boat was fitted up at the pier, and one of the
ablest physicians in the city (Dr. A. Jaminet) was engaged to
attend those affected, and also to institute such sanitary measures
as his judgment should dictate. A careful examination of the
health and bodily condition of every workman was daily made, and
none were permitted to engage in the work without the approval
of Dr. J aminet. Those most severely affected were sent to the City
Hospital and had the benefit of the advice and treatment of its




14                   THE EAST PIER.
resident physician, Prof. E. A. Clark. The total number of men
employed in the air-chamber of this pier was three hundred
and fifty-two.  Of this number about thirty were seriously
affected. Notwithstanding the care and skill with which those
most severely attacked were treated, twelve of the cases proved
fatal.  Each one of these, without exception I believe, was
made the subject of careful inquest by the coroner, aided by an
autopsy conducted usually by some of our most skillful surgeons
and physicians. Whilst the exciting cause in all of these cases
was doubtless the exposure of the system to the pressure of the
condensed air of the chamber, the habits and condition of several
of those who died were at the time they went to work, such as
would have excluded them from it if subjected to the examination
of Dr. Jaminet, and the verdict in about one-half of the cases
gave a totally different cause for the death of the patient;
Nearly or quite all of these deaths happened to men unaccustomed
to the work; several of them to men who had worked but one watch
of two hours.  In contrast to this, is the fact that quite a large
number of the men (certainly one-half of those constantly employed,) commenced with the work at its inception and remained
throughout its continuance entirely without injury or inconvenience.
The gentlemen composing the engineer corps of the bridge all
visited the air chamber, some of them quite often, either in the discharge of their professional duties, or from motives of curiosity,
and none of them suffered any injury whatever.
Much diversity of opinion was expressed by the medical gentlemen
who investigated the symptoms and held autopsies of the deceased.
Some of these gentlemen maintained that a slower transition from
the abnormal to the natural pressure would have been less injurious;
others claimed on the contrary, that it was from the too rapid
application of pressure in passing from the natural into the compressed air. The fact that the air-lock tenders were in no case
affected, although subjected many times during a watch of two
hours in the air-lock to rapidly alternating conditions of the atmosphere, at one moment in its normal state in the lock, and
five minutes later exerting a pressure of fifty pounds per square inch
upon every part of the body, would seem to prove both of these
theories unsound, and lead us to believe that in the length of




THE EAST PIER.                    15
time to which the human system is subjected to this extraordinary
pressure, exists the real source of danger, and not from any rapid
alternations of pressure to which it is exposed. After the caisson
reached the rock, I have frequently, when passing through the airlock, admitted the compressed air into it so quickly that none but
those well accustomed to it could relieve the pressure upon
their ears, and yet I felt no ill effects whatever from this rapidly
increased pressure; and in going out I have let the pressure
off so fast that the temperature in the lock has fallen thirty-two
degrees (Fahrenheit) in consequence. These transitions occupied
but three or four minutes.
The fact that the air-chamber was briefly visited by thousands of
persons, including many delicate ladies,even after it had reached the
bed-rock, some remaining as long as an hour in it without any
of them experiencing the slightest ill effects from the pressure,
and the fact that no cases of any importance whatever occurred
among the workmen after the watches were reduced to one
hour, satisfies me that this is the true cause of the paralysis,
and that by lessening still more the duration of the watches, a
depth considerably greater can be reached without injury to the
workmen. Too long a continuance in the air-chamber was almost
invariably followed by symptoms of exhaustion and paralysis.
Dr. Jaminet, on one occasion, remained in two and three quarter
hours when the depth was over ninety feet, and was dangerously attacked soon after reaching home.
Symptoms of paralysis rarely occurred in the shaft, but generally
after the stairs were ascended, and never in the air-lock or airchamber.
A large amount of money has been gratuitously expended by
the company in assisting and providing for those who were seriously affected, and in relieving the wants of their families during their
illness.
As a merited recognition of the courage, fidelity and services of
the men employed in the air-chamber of the east pier; and as an
evidence of my personal appreciation of their faithfulness and
good conduct, I have placed in the appendix to this report, the
names of those engaged in the work when the pier reached the
rock.




16                   THE WEST PIER.
THE WEST PIER.
The sinking of the west pier began on the fifteenth day of
January of last year, that being the day on which the first
stone was laid on the caisson. As it was commenced twelve weeks
later than the east pier, advantage could be taken of the experience gained in sinking the latter. On the east pier the interruption
and annoyance caused by the necessity of riveting on the iron
plates outside of the masonry to exclude the water, as the pier
descended, led me to devise some means whereby this inconvenience might be prevented and the cost of the iron saved.
The design of the west caisson was so modified that the use of
these plates could be abandoned after the first twenty-nine feet;
nine feet of this height being the air-chamber, and the remaining
twenty feet enveloping the masonry above the chamber.  This
height of plate iron was deemed requisite to give such rigidity to
the caisson as would insure it against any twisting or straining that
would endanger the bond of the masonry. After a depth of
forty or fifty feet was reached by the east pier, it was found
evident that brick linings in the shafts, although surrounded by
many feet of masonry carefully laid in hydraulic cement, were not
sufficient to exclude the water, which at this depth filtered through
quite rapidly. To preventthis and enable the iron around the outside of the pier to be dispensed with on the west pier, its shafts
were lined with white pine staves, three inches thick in the centre
shaft, which was ten feet in diameter, and two and a half inches
thick in the smaller wells, which were four feet nine inches in
diameter. This device answered admirably, and the estimated
saving in plate iron over the original design was about $10,000.
The lower staves were however, found to be too weak to sustain
the high water of the Spring freshet without expansion bands of 1 x3
inch bar iron, which were placed against them in the shafts in
time to avoid any disaster. This novel feature of wooden linings
and no exterior envelope for the masonry, will be introduced in the
east abutment. The caisson for it is however, made so strongly,
that the iron will only be brought up to a height of twelve
feet around the base of the masonry.  The lining staves in its
wells will be so much stronger that they will require no internal
supporting rings of iron.
Nothing could have exceeded the perfect working of these
economic improvements in sinking the west pier, had not the




THE WEST PIER.                    17
contractors failed to deliver the granite for it in time.  When
the point was reached at which the granite became necessary, the surface of the masonry was six or eight feet above
water and the base of the pier eighteen or twenty feet from the
rock. A judicious view of the case, having reference solely to
the cost of constructing this pier, would have dictated a suspension of the work until the stone should have arrived, rather
than continue the sinking of the pier and suffer the top of the
masonry to descend below the surface of the water. Other questions however, were involved of a more serious nature. It was
universally conceded that any effort to negotiate the securities
of the company would be fruitless unless preceded by an absolute
demonstration of the practicability of putting down the channel
piers of your Bridge upon the bed rock of the Mississippi, through
the unusual depths of water and sand that were to be encountered.
The east pier was in a similar condition at this time, and there
remained no alternative but to continue sinking both, to do which,
it was necessary to build up and brace out the iron plates which had
been carried up from the air-chamber around the masonry of the
last pier, and thus dam out the water from covering the top of the
masonry as it descended below the surface of the Mississippi, so as
to be able to resume the laying of stone when the granite should
arrive; and as the west pier was without any such iron envelope, it
became necessary to attach the wooden walls of a coffer-dam to
its sides in such a manner as to exclude the water from the top of
the pier. This was done by padding the lower edge of the dam and
attaching its sides securely to the masonry, several courses below
the top. This answered very well until the last nine inches of this
pier was sunk to the rock. Some of the bolts holding down the
dam at the south end of the pier gave way, and the friction of the
sand, then fourteen feet high on the outside of the dam and nearly
to the surface of the water, prevented the settlement of the dam
with the pier at this particular point.  Two outside courses of
the limestone to which it was attached was held up by the dam
nearly across the entire end of the, pier. This mishap made it
necessary to pump away the sand outside this end of the dam, and
put down a large pad or wooden covering reaching below the two
disturbed courses, then pump out the dam and relay them,
This was done at a depth of about seventeen feet below the surface,
The dam was not pumped out till the granite arrived, as the water over the pier did not prevent the work in the chamber, access to




18              THE EAST ABUTMENT PIER.
it being obtained through the wooden shafts or wells which passed
down through the water that was over the masonry. This unexpected trouble, and that at the east pier, were caused solely by the
failure of the Richmond Granite Co. to deliver the granite in time.
Its delivery was due several months before it was needed, and
yet it was not delivered until several months after it was wanted.
The loss to your company, resulting from this failure. i estimate to
amount to at least $50,000.
THE EAST ABUTMENT PIER.
The complete success which attended the sinking of the east
pier convinced me of the practicability of sinking the east abutment
pier to the rock in the same manner. The original plans of the
bridge did not contemplate resting this abutment on the rock. It
was believed quite practicable to protect its foundation with riprap stone, and to secure stability and safety by resting it on piles,
which were to have been driven to a distance of fifty feet below
low water mark. When the Directory of the company were assured of the practicability of resting this abutment upon the bedrock itself, and of thus terminating forever all doubts as to the absolute stability of each one of the four great piers of your Bridge,
the desire that this, the largest of them all, should be placed on the
rock also, was unanimous; although the excess in cost involved over
that of the original design was understood to be about $175,000.
No less than ten thousand additional cubic yards of masonry
below the line of the tops of the piling on which it was originally
intended to start the masonry, are required to sink the pier to the
rock. Consequently below this line, it will require nearly as much
masonry as will be contained in the west abutment when the
latter is completed.
This abutment, when completed, will contain twenty-two
thousand four hundred and fifty-three cubic yards of masonry,
including concrete and brick work, and will measure in height,
from the rock to the top of its cornice, one hundred and ninety
six feet nine inches.
The depth of the rock at the site of the abutment was ascertained
by careful borings to be eight feet lower than that at the east pier,
or one hundred and thirty-six feet below high water mark.
It is not probable that we shall have to contend with much deeper




THE EAST ABUTMENT PIER.                  19
water or much greater air pressure than that encountered in sinking the east pier. The Led-rock at this abutment is ninety-four feet
below extreme low water mark, and the river is not likely to be
more than eighteen feet above that during the seasons occupied in
sinking the pier. Extreme low water mark is only reached when
the river is gorged with ice above the city, and the volume of
water below the gorge becomes in consequence greatly lessened.
The ordinary low water rarely reaches a point within five feet of
low water mark.
In accordance with the wish of the Directory, preparations for
sinking this abutment to the rock were commenced, and are now
nearly completed.  The caisson is nearly ready for launching,
after which it will be immediately towed into position, and be
made ready for sinking. It will cover when on the rock, five thousand square feet of surface, and is therefore about one quarter
larger than the base of the east pier. A large trestlework has been
erected immediately east of the site of this abutment extending a
few feet above high water mark, on which is being placed a portion of the necessary machinery for sinking the caisson. This
trestle-work was rendered necessary because of the shallowness of
the water on that side of the abutment. On the west side of the
caisson one of the derrick boats used at the east pier will be
located, and on the trestle work will be placed much of the
machinery of the other derrick boat, which was used on the other
side of the east pier.
THE EAST ABUTMENT CAISSON.
This caisson will have several novel features in its construction,
which I think will make it superior to those used for the east and
west piers. The main shaft will have at the bottom two airlocks, each eight feet in diameter, instead of one of but six feet.
The main shaft will be carried down into the air chamber ten feet
in diameter, instead of but five feet.
The east pier caisson had six other shafts of four feet nine inches
in diameter, with air-locks of the same diameter at the bottom of
each.  This caisson will have but two other shafts of four feet in
diameter, with an air-lock at the bottom of each, eight feet in
diameter. These last two shafts are enlarged below the roof of the
air-chamber to eight feet in diameter each. The increased di



20              THE EAST ABUTMENT PIER.
ameter of the locks will contribute to the health of the men, as it
is sometimes necessary for twelve or fourteen of them to be in
one of them at the same time, for several minutes, until the
pressures are equalized.  As all four of these locks are within
the air chamber, and also the lower ends of the three shafts, and as
about one thousand eight hundred cubic feet of space is occupied
by them, there will be that much less space to fill in the compressed air when the pier is down.
The centre shaft alone will be used for the workmen, unless some
unforeseen accident should render it necessary for them to use one
or the other of the side shafts, which are provided almost solely
for safety. The extra size of the locks makes either one of them
capable of holding in an emergency all the men that will be in the
air chamber at one time, and hence their security will be increased.
To avoid the labor of walking up a circular stairway about one
hundred and twenty feet high after leaving the air chamber, the main
shaft, in addition to the stairway,will have an elevator or lift to bring
the men up. This, it is believed, will contribute greatly to their health.
When they are at work in a pressure of forty-five or fifty pounds
above that of the natural atmosphere, there ensues a rapid
exhaustion of the physical energies. When relieved from duty, a
considerable degree of prostration is frequently manifested, and the
foremen of the different gangs were in consequence instructed by
the physician to cause the men to ascend the stairs leisurely, to
avoid increasing it.  I confidently believe therefore, that by
bringing the men to the surface in the elevator, there will be much
less danger of injury occurring from their employment in the airchamber.
LIGHTING THE CAISSON.
A different method of lighting the air chamber will likewise be
adopted. In the other caissons much inconvenience was experienced on account of the particles of unburned carbon thrown off
from the flames of the candles used. The consumption of the candles
under the action of the compressed air was much more rapid than in
the normal atmosphere. At the depth of one hundred feet, they
were found to be consumed in about three-fifths of the time required in the open air. Large quantities of smoke were emitted
from the flames, and the air was filled with particles of floating




THE EAST ABUTMENT PIER.                 21
carbon which could only be removed thoroughly, by placing a rose
jet on the nozzle of a water hose in the chamber, and discharging
the spray in every direction. Some amelioration of the evil was
obtained by burning the candles under an inverted funnel or
chimney, communicating with one of the shafts by a small outlet
pipe, through which the escape of the compressed air was regulated by a cock, thus creating a draft above the flame by which
the smoke was carried off.
The calcium light would probably prove the most satisfactory one which could be employed in the chamber, were it not
for the excessive cost of it in this city. For the one hundred
and fifty days which will be required in sinking this pier and
completing its foundation on the rock, the cost of lighting the
three compartments of the chamber with calcium lights would
be at least $5,000.  By the means devised for the purpose the
cost cannot exceed one-fifth of that sum.
The difficulty of extinguishing a flame in an atmosphere of such
density, caused me to forbid the use of oil lamps in the chamber
before a depth of eighty feet had been reached. The clothing of
two of the men having taken fire from contact with some of the
hand lamps or candles used in the caisson, it was found exceedingly difficult to extinguish the flames.  One of them was
severely burned, although his garments were almost entirely woolen.
It was deemed unsafe to risk the danger of having the clothing
of the men saturated with oil from the accidental breaking of a
lamp which might, by the same casualty, ignite their garments and
thus endanger their lives. The use of oil was therefore forbidden.
At the depth of eighty feet it was found that the flame of
a candle would immediately return to the wick after being blown
out with the breath. At the depth of one hundred and eight and
one-half feet below the surface of the river, I blew out the flame of
one of them thirteen consecutive times in the course of half a minute,
and each time excepting the last, it returned to its wick. Almost
as long as a small portion of the wick remained incandescent,
the flame would return, and when the glowing particle of two
separate candles failed to possess sufficient heat to restore the
flame to either, it would reappear at once by placing the luminous portions of the two wicks in contact.




22              THE EAST ABUTMENT PIER.
The chamber of this caisson will be lighted by candles contained
within glass globes of similar construction to those used in lighting railway carriages.  The glass will be of strength sufficient
to sustain the external pressure of the condensed air. The chimney will consist of an outlet pipe of one inch diameter, communicating with one of the shafts, and the compressed air will only
be admitted within the globe in which the candle is placed,
through a small regulating valve. The candle will therefore be
burning under the normal pressure of air. A stop-cock in the
chimney will prevent the escape of air from the chamber through
the globe, when it is desirable to put in another candle, or to
clean the glass.
TIMBER WORK.
Beneath the masonry piers of suspension and truss bridges,
it is quite common to employ a considerable amount of timber.
Where the pressure upon the pier is a vertical one, this economical substitute for stone is admissable, but in the piers of an
arched bridge, where some one span is at times loaded while the
others are unloaded, the thrust of the loaded arch has a tendency to
oscillate the piers, and with a few feet in thickness of a material so elastic as wood under their bases, this oscillation would
prove a dangerous feature.  In the abutment piers where the
thrust is only from one side, and oscillation is prevented by the
works on shore, timber may be safely used to a considerable extent.
To give the desired stiffness to the caisson for this abutment and
avoid the more costly use of iron, the roof of the air chamber is
made of timber four feet and ten inches in thickness. A large
amount of timber is also used in constructing and stiffening the
sides of the air chamber, which are ten feet high, and in forming
two horizontal trusses or girders through the air chamber. These
two girders are each ten feet thick at the top, three and onehalf feet at the base and nine feet in height.  They are about
seventy-three feet long, and are interlocked at each end with the
sides of the air-cham ber. They divide this chamber into three nearly
equal compartments in the direction of the length of the Bridge.
Communication is made between these compartments by means of
two openings through each girder.
The sides of the chamber are eight and one-half feet thick
at the top and eighteen inches at the bottom, and are




THEZ EAST ABUTMENT PIER.                 23
composed of timbers, some placed vertically, others horizontally, and  some inclined  at an  angle of about fortyfive degrees, and the whole, including roof and girders, thoroughly interlocked together, and bolted with large iron bolts.
All of the timber is of' the very best white oak, and was squared
up with two steam planers belonging to the company. In addition
to the iron bolting used, these timbers are thoroughly secured
together with large white oak tree-nails.
The wood work of the caisson has been most admirably executed
under the superintendence of Mr. John Dunlap, master of shipcarpenters.
PLATE IRON WORK.
Enveloping this entire wooden structure is an iron covering
riveted together to prevent the escape of the air which is
to supply the workmen.    This is of three-eighth inch plate
iron, and its sides are increased in thickness at the bottom
edge to three inches, by riveting four three-quarter inch plates
together.  These extend several feet up the sides. This iron
edge extends ten inches down below the wooden sides and
forms the cutting edge of the caisson. Every two feet the
iron sides are strengthened by vertical angle irons three by
seven inches in size, riveted on flat wise on the outside of the
caisson. Through these angle bars, bolts one and a half inches in
diameter are inserted, and by them the iron and wooden sides are
strongly held together.  This iron covering extends over the
wooden top of the air chamber and forms a floor on which the
masonry will be laid.  The three shafts passing through this
floor, by which access to the chamber is obtained, are tightly
riveted to it.
The iron sides are carried up twelve feet above this floor,
where they will terminate.  The masonry above this point will
therefore have no exterior envelope such as the east pier had.
Nearly all of the iron used in this envelope was obtained from
the hull of the iron gun-boat Milwaukie, the wreck of which was
purchased about eighteen months ago. This iron work has been
executed by Capt. Wm. S. Nelson, the skillful and energetic contractor who built the caissons of' the two channel piers.




24              THE EAST ABUTMENT PIER.
WATER TIGHT LININGS.
The water penetrating the masonry will be excluded from the
three shafts by white pine linings, arranged like the staves of a cask.
The staves composing this lining in the main well or shaft,
which is ten feet in diameter, will be ten inches thick in the
lower part, and will be gradually diminished to three inches at the
top. In sinking the pier, the top of the masonry and shafts, will
be kept constantly built up above the surface of the river.
FILLING THE AIR CHAMBER.
The most valuable improvement in the design of the caisson
will I think, be found in the method devised for filling the chamber
when it has reached the rock. It is a well established fact that
sand constitutes one of the most reliable and durable materials for
foundations known, if availed of in positions where it can be securely retained under the structure erected upon it. It is an equally
well established fact that timber, when entirely submerged in fresh
water foundations, is indestructible. These two facts will be relied
upon in filling the air-chamber and fixing the foundation of this pier
upon the rock. Instead of concrete, sand will be chiefly used for
filling the chamber. The sides of the caisson are of great thickness, and are thoroughly interlocked at the corners of the air-chamber, and at the ends of the girders. The possibility of the sand
surrounding the pier ever being scoured out to the rock, at the site
of this abutment, is a very remote one. It is certainly much more
improbable than that it may be scoured thus deeply at the sites of
the two channel piers. To avoid all danger from this very remote
possibility throughout all time, whatever space there may be existing between the timber walls of the caisson and the bed-rock, after
the caisson shall have reached it, will be thoroughly concreted, so
that these walls will have a substantial bearing upon a solid material which cannot be affected by any current that may possibly
wash the base of the pier. The walls of the air-chamber are so
framed as to be sufficiently strong to resist the bursting pressure
of the sand within the chamber, caused by the weight of, the masonry of the pier and half the side span upon it, even after all the
iron used in it shall have been corroded away.  The base of the
pier is 5,000 square feet in area, and the weight of the entire pier
including one-half of the span, will be about 46,500 tons. The




THE EAST ABUTMENT PIER.                 25
pressure per square foot on the rock would, therefore, be 18,600
pounds.  The area of the wooden edge of the caisson, including
that of the bottom of the girders, air-locks and shafts, is about
1,250 square feet. This area alone would be capable of sustaining
the pier, without any additional support from the sand contained
within the air-chamber.  Without this sand filling, the pressure
upon the wooden base of the caisson (including the locks and shafts)
would be about 74,000 pounds per square foot, or 514 pounds per
square inch. This pressure is not beyond the power of good white
oak to resist, nor would it be sufficient to crush the concrete that
will be used in filling the small space between the oak and the
rock.
Tests made with our testing machine upon a number of blocks of
concrete only six weeks old, gave an average resistance to crushing,
equal to one thousand two hundred pounds per square inch. Of
course, with the integrity of the exterior of the caisson unimpaired,
the escape of sand from the interior would be impossible. With
the interior compactly filled, the pressure of the superincumbent
mass must necessarily be very nearly equally distributed over every
part of the caisson, and hence it cannot exceed about 16;(00(1. o
pounds per square foot.
The tedious process used in concreting the air-chamber of the
channel piers, together with the objections to working men at
such great depths, induced me to devise some method by
which a smaller amount of manual labor could be made to accomplish equally good results. By the plan determined on in this case,
I confidently hope to accomplish the necessary work in the airchamber with a fifth or sixth of the manual labor which was required under the east pier.
This method is so simple' as to be readily explained. So soon as
the rock shall have been struck by the iron edge of the caisson
the space then remaining between the wooden walls of the caisson
and the rock, will be thoroughly concreted. The sand under
the two girders will be left intact. The borings indicate that
the rock is quite level, and it is not probable that inequalities of more than eighteen inches will be found in it. It is estimated that one hundred cubic yards of concrete will be sufficient to support these walls, forming a bed of an average width of
three feet six inches, by two feet six inches in height. This
concreting being done, all of the pipes passing vertically




26                 THE WEST APPROACH.
through various parts of the pier, and used for air, water
and sand-pumps, will be closed at the top, and the pumps
valves and pipes connected with them in the air-chamber
will then be taken off.  There will be nineteen of these vertical
pipes, each either four or six inches in diameter, the lower
ends of which will be enlarged conically through five feet of their
lengths. These pipes being opened at their lower extremities, and
one of the inner doors of an air-lock being secured from being
clogged by sand, the air from the chamber will be permitted to escape, and the chamber will be filled with water. This being done,
sand will be introduced through the various vertical pipes mentioned. By means of plummets in these pipes, we shall be able to
determine the height of the sand discharged in them, and when it
is near the roof of the chamber the air will be again pumped in,
and workmen will be sent in to level it off. By repeating
this process two or three times, the chamber can be filled nearly
to the roof with sand compacted in the water, which will insure
its solidity. The remaining space can then be filled with concrete rammed in under the roof of the chamber.   The great
thickness of the walls and of the girders where they join the roof
reduces the area of the upper part of the chamber very greatly.
The upper three feet of it measures only two thousand and twenty-five cubic feet in area exclusive of the air-locks and shafts.
To fill the air-chamber of the east pier required one thousand
three hundred and forty cubic yards of concrete which was placed
in position by manual labor, under an air pressure of nearly fifty
pounds per square inch. This pier is twenty-five per cent. larger,
and will require only about two hundred cubic yards of concrete
to be placed in it under similar conditions; hence the work required to be done in this chamber will be greatly less than in
that of the east pier.
WEST APPROACCH.
In this approach, which will be entirely of stone, there will be
five arches of twenty-six feet eleven inches span, and forty-two
feet ten inches in height above the level of the street. The foundations for the piers to support these arches, all rest upon the
rock underlying the wharf.  These foundations are nine feet
by forty-six feet six inches.  The one next to the west abut



THE EAST APPROACH.                    27
ment is the deepest of the five, the rock being forty-one feet
eight inches below the City Directrix. This foundation is already
completed and will soon receive the Missouri rock-faced red granite,
which will form the base for the fine cut sand stone, of which the
approaches, on each side of the river, will be built. The foundation
for the second pier of this approach has just been commenced. The
foundation for the third and fourth piers are completed and are
ready for the sand stone. The masonry already laid in these three
approach piers measures nine hundred cubic yards.
The foundation for the fifth pier, will be in the line of the houses
fronting the wharf, and has not yet been commenced. The cellars
of the houses, where it is to stand, have been blasted out of the solid
rock, and this pier will be rapidly and easily constructed when
commenced.
TIHE EAST APPROACH.
The piers for this approach will be built upon pile foundations
none of which have yet been commenced.
There will be no difficulty in completing this approach within the
desired time, and it will be more convenient and economical to
begin it after the work on the west approach is more advanced.
In design it is almost exactly like the west one.
SAND STONE.
In the selection of sand stone, the greatest possible care has been
exercised. This selection was more especially supervised by Col.
W. Milnor Roberts, Associate Chief Engineer, who, personally
visited and carefully inspected every sand stone quarry of any note,
within available distance of the work. The one selected lies close
to the Mississippi river, near St. Genevieve, and is distant about
sixty-five miles from St. Louis. The stone is a pure sand stone, of
a warm yellowish tint. It is of uniform color, free from blemishes, and from the tests made of it promises great durability.
A force has been at work for some time getting out and cutting
the stone, and no fear of delay is entertained on account of nondelivery of it in time.




28                       GRANITE.
GRANITE.
Col. Roberts has also devoted much of his time to the selection
of the granite used, and to be used, in the construction of your
Bridge. Several examinations were made by him of the Eastern
granites, and almost every quarry from Richmond to Buck's Harbor
in Maine was visited by him, with a view to obtaining the best and
cheapest that could be had. A contract for the first seven hundred
cubic yards was made by him     with the Richmond Granite
Company, for gray granite. Subsequently, another was made with
Messrs. Thomas Westcot & Son, of Maine, for all the gray granite that
will be required for the two abutments and the two channel piers.
All four of these piers will be faced with granite ashlar above
extreme low water mark, except those parts of their sides which
are above the springing of the arches and beneath the roadway,
and inclosed between the spandrels of the arches. This portion,
which is much less exposed to the weather and to view, will be of
cut sand stone.
Missouri red granite will also be used as ashlar, but only
in the bases of the approaches, and on the T walls of the abutments, and will appear only to the height of the curb stone on
the St. Louis wharf, which is nearly the level of the City Directrix. For this purpose about fourteen hundred cubic yards will be
required. This granite promises to be equally as durable as the
gray, and that which is already laid up in the work is greatly admired, on account of the richness and beauty of its color.
It is to be regretted that proof could not have been given,
at an earlier date, of the capacity of a Missouri quarry to
supply a material so excellent and desirable.  One of the
good results of your enterprise, is the discovery and developement of this extensive quarry, owned by HLon. B. Gratz Brown,
with whom   a contract was made last April for 1,400 cubic
yards of the stone, a great part of which has been already
delivered.  This quarry is distant ninety miles from St. Louis
and three miles from the Iron Mountain Railroad.
Several hundred dollars were expended by the Bridge Co., in
fruitless endeavors to obtain the proper quality of gray granite
from a quarry in this state, prior to this contract and this dis



TESTS OF GRANITE.                   29
couragement, together with the unfavorable results of examinations made of other unproved quarries in this state, created a
reasonable assurance that no suitable granite would be discovered
in Missouri, in time for our wants. Hence, no alternative remained, but to seek for it elsewhere.
TESTS OF GRANITE.
In the Appendix will be found two interesting reports, one from
Prof. Felix McArdle and the other from Dr. Eno Sander, chemists
of high standing in St. Louis, giving the chemical tests applied by
them, and the results produced by these tests upon the samples of
red granite now being used in the construction of your Bridge.
The very careful experiments and report of Prof. McArdle were
made gratuitously.
A resolution thanking him for this generous manifestation of his
interest in your enterprise was passed by the Board of Directors of
ye-u- Bridge Company.
MAGNESIAN LIMESTONE.
The interior of all the masonry will be of magnesian limestone from the Grafton quarries. None of this stone will be
exposed to the weather. It is remarkably strong. Many tests
of its compressive strength have been made in the company's
testing machine, where its resistance has, in several instances,
exceeded 17,000 pounds per square inch, which is equal to that of
granite.
A curious fact has been developed by these tests, which is that,
the modulus of elasticity of this stone is about the same as that of
wrought iron. That is, a given weight placed upon a wrought iron
column and on a column of the Grafton stone of the same size, will
produce an equal shortening in both; while the elastic limit (or
breaking point) of the stone is not far below the limit at which the
wrought iron would be permanently shortened. A column of the
stone two inches in diameter and eight inches long, was shortened
under compression in the testing machine nearly one quarter of an
inch without fracturing it. When the strain was removed the
piece recovered its original length.




30                   TESTING MACHINE.
TESTING MACHINE.
The testing machine, the design of which was made by Col.
Henry Flad, Chief Assistant Engineer, has been in operation for
several months, and has given the greatest satisfaction. By means
of a very simple little instrument, suggested by Chancellor Chauvenet, and matured by Col. Flad, the most delicate changes in the
length of the specimen can be accurately recorded, with a degree
of minuteness never before obtained or even approximated, in any
testing machine, so far as my information extends. By this instrument it is perfectly easy to detect a change in the length of the piece
equal to the two hundred thousandth part of an inch.
A brass collar is slipped over each end of the specimen, and these
are secured by three pointed set screws in each collar. Any shortening or lengthening of the piece, will of course, alter the distance
between the two collars. One collar has on the side of it a small
flat surface or vertical table. Against this table is placed a little
vertical steel cylinder, which is held against the table by the end
of a little flat horizontal bar that is secured at its other end to the
other collar. This bar is held against the steel cylinder by a spring,
having sufficient strength to keep the cylinder from falling. It is
evident now that if one collar be brought nearer, or is moved
farther away from the other, the steel cylinder will be rotated, as
one side of the cylinder is pressed against the table, which is attached
to one collar, while the other side is pressed by the little bar that is
fastened to the other collar. If the specimen be subjected to pressure it will be shortened and the collars will approach each other.
If tension be applied to the specimen, the piece will be stretched
according to its intensity, and in either case, the rotation of the
little steel cylinder will indicate the measure of the disturbance that
has occurred between the two collars, and it will give it absolutely
without any element of error entering into it from any change of
the dimensions of parts of the machine under strain. By placing
on the top of this little cylinder a small vertical mirror, the extent
to which the cylinder has been rotated may be determined in
the following manner: Twenty-five feet from the mirror, an arc of
a circle is struck, the little steel cylinder being the centre of the are.
On this arc is erected a scale of inches with decimal subdivisions.
This scale being illuminated by gas-light can be easily read in the
mirror by means of a small telescope placed immediately above the




TESTING MACHINE.                    31
scale. The angles of incidence and reflection at the surface of the
mirror being equal, it follows, that one-fourth of a complete rotation
of the mirror, would be equal to a half circuit of the circle of which
the arc is a part; or, in other words, a movement of the mirror of
but one degree would be shown on the scale, by the reading of a
space equal to two degrees, or the one hundredth part of an inch
on the scale, would really be only half so much, or the two hundredth part of an inch, when seen in the mirror. The diameter of
the little cylinder is so proportioned to the radius of the arc as to
make the smallest subdivision of the scale equal to the twenty
thousandth part of an inch, but the observer, after a little practice,
can subdivide these divisions, which are magnified by the telescope, so as to observe the two hundred thousandth part of an inch.
The power is applied to the specimen under trial by means of a
hydraulic press, the ram of which moves horizontally. The ram
has a steel rod extension passing through the rear end of the
cylinder. Specimens for testing by tension have one of their ends
secured to this steel rod, and the other to the end of a scale beam.
Specimens for crushing are placed at the other end of the cylinder
and are compressed between the end of the ram and a crosshead.
This crosshead is attached to the end of the scale beam before mentioned, by four powerful rods of steel surrounding the cylinder and
leading back to a crosshead attached to the beam. This latter
crosshead is detached from the beam when tensile experiments are
being made.
It will be obvious, on reflection, that when a piece is being crushed
by the thrust of the ram, the four bolts sustaining the crosshead
against this thrust must stretch in proportion to the power applied,
and hence the specimen will be moved bodily in the same direction,
and that this will affect the accuracy of the readings of the mirror,
as it too will be moved horizontally with the specimen to which it
is attached. To correct this minute error in the readings, a second
mirror and scale are used to ascertain the extent of this horizontal
movement. The table holding this second mirror, against which
the little cylinder rotates, is secured to the frame of the testing
machine, which has no strain on it, and the little bar for rotating
the cylinder is attached to the crosshead; of course, any movement
of this head causes a rotation of this second mirror by which the
extent of the movement can be at once ascertained.




32                  SUPERSTRUCTURE.
It is equally important to know the exact weight applied to the
specimen as well as the change of form assumed by it when subjected
to the weight. Having no faithin the accuracy and durability of
the ordinary mercury and spring gauges for such high pressures
as are required in a hydrostatic testing machine, I determined
that the absolute strain on the piece must be weighed on the
balance. This, Col. Flad has very ingeniously accomplished by a
system of levers, balanced on hardened chrome steel knife edges
and boxings, sufficiently powerful to stand a strain of one hundred
tons, and yet so delicate as to be turned by the weight of one-half
of an ordinary cedar covered drawing-pencil when placed in the
balance. One pound weight placed in the balance, equals a ton of
two thousand pounds weight on the specimen.
I feel safe in asserting that the company have a testing machine
which can scarcely be excelled in the accuracy, delicacy and min teness of its results.
It has been placed in charge of Mr. Paul Dalhgren, C. E., by
whom a carefully tabulated record is kept of all tests made with it.
A great variety of these have already been made upon specimens
of steel, iron, woods of various kinds, granite, brick, limestone,
concrete, cement, models of tubes, trusses, &c., &c. Much valuable
information having direct reference to the work in hand, has been
already obtained by these experiments.
SUPERSTRUCTURE.
On twenty-sixth day of February last, a contract was made with
the Keystone Bridge Company of Pittsburg, for the construction
and erection of the superstructure of your Bridge, including that
of the approaches. By this contract the Keystone Bridge Company
undertakes to furnish all materials at the same prices per pound
and per foot, at which they were estimated in my published report
of May, 1S68, excepting cast steel work, which is to be furnished
at $20.00 per ton less than the cost set forth in that report. There
will be about two thousand five hundred tons of steel used, therefore
the saving on this item will amount to about $50,000. The contracting party will, however, receive $40,000 more for erecting
the three spans than the estimate in the report. Every other item
of cost as set forth in the report referred to, is the price per pound
or foot to be paid the Keystone Bridge Company. The amounts




SUPERSTRUCTURE.                    33
set forth under the head of engineering and contingencies, in that
report, and aggregating $149,512 14, for superstructure of Bridge
and approaches, are reserved by your company, and will be ample
to cover any excess of materials required over the amounts estimated, and for engineering expenses, &c.
By the terms of the contract with the Keystone Bridge Co., it
agrees under a severe forfeiture in case of failure, to complete the
structure ready for use in all its parts in seventeen months from the
time working drawings, were furnished to it: provided it is not
delayed by masonry work after the first of March next. In case
of such delay, the time of completion is to be extended no longer
than the time it is so delayed. Completed working drawings were
not furnished until the first of July as the completion of certain parts
of them was dependent upon data that were obtained from the
testing machine, and which could not be ascertained at an earlier
period. This delivery of drawings fixes the time for completion
of the Bridge on the first of December of next year. I have no
apprehensions that the masonry will not be completed in season to
prevent any claim for an extension of time on the part of the Keystone Bridge Company.
I have been informed that the Keystone Bridge Company has
contracted with the Wm. Butcher Steel Works Company, of Philadelphia, to furnish the cast steel that will be required in the work.
Specifications for the cast-steel work will be found in the Appendix to this report.
I have tested so many samples of steel made by this company
which surpassed in strength the requirements set forth in these
specifications, that I have no fear of its not being able to supply
the quality required.  Several pieces of this steel have shown
limits of elastic reaction ranging from seventy thousand to ninetythree thousand pounds per square inch.
Since my report, 1st May, 1868, in which the plan of superstructure was described, I have made several modifications in the
general arrangement of the arches and in the details of their
construction, which will considerably improve the architectural
appearance of the Bridge and simplify its fabrication.




34                   SUPERSTRUCTURE.
These changes consist mainly in using but one cast-steel tube of
eighteen inches diameter, instead of two of nine inches, in forming
the upper and lower members of each one of the four ribbed
arches composing each span; and in increasing the depth of each
one of the arches from eight feet to twelve feet from centre to
centre of these tubes.
The railways (which are below the roadway) are raised four feet so
that in no place will they appear below the arches, as they did in the
original design. In that design the railways were eight feet lower
than the centre of the middle span. By deepening the arch four
feet and raising the tracks four feet, they are brought level with
the centre of this span, or above the soffit of the arch. The lower
ribs or tubes of the arches spring from the piers at their original
level, consequently the arch has four feet less versed sine or rise
than before. To lessen the grade of the railways it was necessary that
the tracks should descend each way from the centre of the middle
span. This would cause them to fall below the centres of the side
spans, to avoid which the level of the springing of these two spans
has been lowered eighteen inches at each abutment. That is the
ends of the arches of the side spans resting against the abutment
piers, will be eighteen inches lower than the other ends which rest
against the channel piers. These arches, like the central ones,
have four feet less rise than as originally designed, and by lowering
their shore ends as stated an additional gain of nine inches depression is obtained at their centres, by which the gradients of the
tracks are proportionately lessened towards the ends of the Bridge.
Raising the tracks to the height of the centres of the arches will
unquestionably improve the appearance of the structure, and it is
generally conceded that the alteration in the level of the springing
of the shore ends of the side spans is likewise an architectural
improvement. The effect upon the eye caused by it, will be somewhat similar to that produced by the camber of the Bridge.
Of course these changes involved the necessity of revising the
former investigations and results, so as to ascertain the difference
in the strains, and to determine the alterations requ ired in the sec
tional areas of the various members of the structure, when thus
modified. An entirely new set of detail and general drawings were
likewise required in consequence of these changes.




WIDENING THE AVENUES TO THE BRIDGE.            35
The lithographic view of the Bridge in the appendix is a very correct representation of the structure as it has been definitely determined upon, and is now being constructed. This view also shows
the depth of the bed rock at the site of the different piers, and the
depth of sand overlying it during ordinary stages of water.
CONDEMNATION OF LAND FOR APPROACHES.
Since my last printed report, the land required in Illinois for the
eastern approach to the Bridge has been obtained by condemnation,
and paid for by the company.
Judicial proceedings have been commenced in this State for the
condemnation of the requisite ground for the approach on this side
of the river. About one-fifth part of that which will be required
has already been obtained by purchase. A commission has been
appointed by the Court to fix the values upon the remaining pieces
wanted. No delay in obtaining possession of all the land required
is anticipated. These matters are entirely under the control and
in the charge of the Executive Committee.
WIDENING THE AVENUES TO THE BRIDGE.
During the last session of the General Assembly of the State of
Missouri, a law was passed requiring an election to be held by the
citizens of St. Louis to decide upon the question of taxing the city
with a sum not exceeding $500,000.00, to defray the cost of widening the streets leading directly to the Bridge. This election was
decided affirmatively by a very handsome majority. Steps have
already been taken by the Mayor of St. Louis, Hon. Nathan Cole,
to carry the will of the people, thus expressed, into effect.
Washington avenue is the most centrally located avenue in St.
Louis, and is also one of the most beautiful. It runs nearly in the
direction of the Bridge, which is located at its eastern terminus.
By the Bridge this avenue is virtually extended across the Mississippi river into the State of Illinois. The law referred to requires
this avenue, which is eighty feet wide, to be widened at Third
street, where the roadway of the Bridge begins, to 140 feet, and at
Fourth street to 117 feet.
Third street which is intersected by the roadway of the Bridge
is at this point only sixty feet wide, and immediately south of the




36         CHANGES IN THE BED OF THE RIVER.
Bridge it is only thirty-eight feet wide. The law contemplates the
condemnation of the fronts of seven blocks on this street, three on
one side and four on the other side, so that it will be 116 feet wide
at the Bridge. This width will be maintained throughout two
blocks north, and one block south of the Bridge. From this latter
point it will be gradually narrowed from one hundred and sixteen
feet to seventy-six feet, in the length of the second block south.
Thence south, Third street is but forty or fifty feet wide. The
widening of Washington avenue will however, afford easy access to
Fourth street, which extends southwardly from  the Bridge a
mile or more in one uninterrupted width of eighty feet, by which
the southern travel will be conveniently accommodated. North of
the Bridge, Third street, or Broadway, as it is called, will afford
one grand highway, one hundred feet wide, to the northern limits
of the city.
These improvements will no doubt be completed by the city
authorities as soon as the Bridge is finished. They will contribute
greatly to the appearance and beauty of it, and will vastly
promote the convenience of the public. The wisdom and liberality of those who voted in favor of providing these magnificent
highways to accommodate the vast tides of travel that will hereafter
flow to and from the Bridge, will be more fully appreciated when
the structure is completed.
CHANGES IN THE BED OF THE RIVER.
I think the propriety of placing the channel piers of the Bridge
upon the bed-rock can be no longer questioned, if we consider the
facts developed in sinking them. The remarkable scour of fifty-one
feet below low water line made in the bed of the stream at the
east pier, by the freshet of last April, is sufficient to prove that
the scour extends much deeper than was supposed to be possible
by many distinguished engineers.   The depth of scour was
assumed by them as never exceediln thirty feet below low water
mark. At more than twice this distance below low water mark,
(sixty-six feet,) pieces of bituminous coal, as large as a cocoanut were found imbedded in the sand at the site of the east pier.
This coal had evidently been mined by man and had not been
carried any great distance by the current, as its surfaces were brilliant and the angles which had been formed by fracture were sharp




CHANGES IN THE BED OF THE RIVER.              37
and perfect. From these facts it would seem evident that the coal
must have been carried by the current to where it was found, after
the era of steam navigation, as we have no knowledge of stone
coal having been used on the Mississippi before that period. These
pieces of coal had doubtless been lost from some steamer navigating
the river above the city and lodged where they were found during
a deep scour, resulting from some unusual under current acting
upon the bed of the stream.  These currents, I am convinced,
extend to a greater depth in the winter season, than in time of
floods, which occur in the spring and early in the summer.
The channel opposite this city is very narrow and during severe
winters it usually freezes over very firmly before many wider
places above are closed. From these open parts floes and fields
of ice float down and are driven under the fixed and frozen crust
at this point.  The floating ice, being lighter than the water,
occupies the part of the channel immediately beneath the frozen
crust and there stops, and as this engorgement in the narrow
channel is increased by constant accessions from  above, the
current must be gradually forced deeper and deeper.  In this
way it is not at all improbable that where these gorges occur
in the river, its sand deposit may be totally removed in midchannel, and the bed-rock exposed to the action of the current.
When this occurs a continuance of the supply of floating ice soon
chokes the passage of the water between the rock and the gorged
ice, and thus a natural dam is created across the stream. Sudden
rises of the river above these gorges, attaining in a few hours several
feet in height, are not at all unusual on the Missouri and Mississippi
(dring severe winters. When they occur, the immense pressure of
the water tfiially sweeps away the obstruction, and fills the open
spaces in the river below, for miles distant, with ice so discolored
with river sediment as to be scarcely capable of flotation, and giving
ample evidlece of its imprisonment beneath the surface.
(ool. Roberts found a bone in the sand within a foot or two of
the Iedl-rock, under the east pier. It is a part of the femur or
thigh-bone of anl animal larger than mann, and is not petrified; from
which fact I assume that it could not, probably, have been in the
place where it was found during any long period of time.
While on this subject I will state, as an interesting geological
fact that a piece of the bed-rock was broken off in which is found a




38                  THE ICE-BREAKERS.
considerable amount of white coral. It appears on the surface of
the piece, which is about three inches thick, and extends through
it, appearing on the lower or fractured side. The walls of the cells
are incrusted with quartz, the crystals of which are so minute that
they can only be seen through a lens.
Beneath the west pier logs partly charred were met with at the
depth of fifty feet below low water mark. During the last pumping
of sand from the east air-chamber, eighty-four feet below low water
mark, particles of charcoal were constantly discharged from the
pumps with the sand.
The bed-rock was found to be of dark   colored limestone or
marble of such close texture as to admit of a moderate degree
of polish.   Its surface was worn smooth and covered with
corrugations of from three to six inches il size, evidently proving
that it had been exposed to the direct and constant action of the
current, probably at some very remote period.
TIlE ICE-BREAKERS.
The lateness of the season when the sinking of the east pier commenced, made it absolutely necessary to provide some adequate
protection for the requisite boats, machinery, &c., at the site of the
pier, against the heavy floating ice which invariably makes its
appearance here during the winter.  This floating ice frequently
attains a thickness of ten or twelve inches, and often covers the
entire surface of the river, moving along at the rate of about three
or three and a half miles per hour. In proportion as the weather
becomes more intensely cold, the volume of the ice increases, and
the rate of its movement decreases, until it finally comes to a fill
stop and then quickly freezes over, affording, even within a few
hours afterwards, a safe highway across for pedestrians. In a day
or two later the frozen mass becomes so strong as to support the
largest and most heavily loaded wagons.
The freezing over of the river at St. Lotuis is niot however.an
invariable rule, as it does not occur, perhaps, oftener tha ththree in
every four years on an average. Last winter was fortunately an
exception to the rule. For several days however the floating ice
was so heavy and compact that itwas with the utmost difficulty that
the most powerful steam ferry boats, built expressly to meet such




THE ICE-BRE AKERS.                   39
contingencies, could force a passage through it. One or two trips
across during an entire day, being all that they could accomplish,
frequent attempts in the meantime proving abortive.
To establish in mid-channel any temporary works to withstand
an element so apparently resistless, and of such exhaustless volume,
was an untried experiment on the Mississippi that presented several
very discouraging features. The two chief difficulties were first, to
place any construction above the pier that would not be quickly
scoured out by the current, and second, to make such construction
so strong as to resist the power of the ice to sweep it away. The
method devised by me to accomplish the desired result will be fully
understood by the following description.
About two hundred feet above the pier at a point from whence
the current would flow to the centre of the pier, a pile was driven
which formed the apex of a triangular system of piles shaped like
the letter A. From this pile two lines of other piles were driven at
distances of eight feet. These two lines extended down stream to
the distance of two hundred feet and represented the two sides of
the letter A. At their lower extremities these two sides were about
one hundred and eighty feet distant from each other. The triangle
thus formed was filled in with other piles driven in transverse lines
from side to side at distances of about fourteen feet, and the tops of
the entire system were then thoroughly braced together with hewn
oak timbers ten by ten inches square, well bolted to the piles, which
were of cypress.
The water was from forty to forty-seven feet deep, when this part
of the work was executed, and many of the piles were washed up as
the work progressed. It was difficult to drive them into the sand
more than twenty feet deep, even with a steam pile driver of 3,500
pounds weight.
About fifty feet above this triangle was placed a clump of nine piles
driven close together, and this was encased in sheet iron throughout about twelve feet of its length to prevent the ice cutting the piles,
About one hundred and fifty feet above this clump of piles, a large
iron pile made of the shell of an old cylindrical steam boiler
five-sixteenths of an inch thick, twenty-eight feet long, and fortytwo inches in diameter, was sunk nearly to its full length into the
sand vertically.  From the middle of this iron pile twelve feet




40                  THE ICE-BREAKERS.
below the river bed, was attached before sinking, a wire cable of one
and seven-eighths inches in diameter. This cable was led over the
clump of piles and firmly secured to it, and from thence it was carried
down to the apex of the triangular system below, where it was hauled
taught and securely fastened. The object of the rope was to aid in
holding the piles steady until the entire protective system was completed; and also to form a cutting edge on which the large floes of ice
could be raised and broken asunder before striking the works
below. To the triangular system of piles the caisson was secured,
and was held by it against the current until it entered the sand.
The iron pile was open through six feet of its lower part to form
a sand chamber into which one of the sand pumps was introduced
to withdraw the sand and permit it to sink. Above this chamber, the
pile was filled with ore from the Iron Mountain of Missouri to insure
its sinking in the sand. A central tube fourteen inches in diameter,
made of an old boiler flue, enabled the sand pump to be passed
through the pile to the sand chamber at the bottom; the ore being
contained in the annular space surrounding this tube. The water
was about thirty-five feet deep at the site of this pile when it was
sunk. After its lower end had penetrated to a point about sixty-two
feet below the surface of the water, and the cable had been tightly
stretched, fifty or sixty cubic yards of large rubble stone were
thrown in around the pile to protect it from scour.
After this work had progressed thus far, a subsidence of about ten
feet in the river, enabled us to bolt on to each side of the triangle of
piles, about ten feet below their tops, a longitudinal timber about
ten inches square, running the entire length of the system. These
two longitudinal timbers placed near the surface of the water, and
well secured to the sides of the triangle, constituted hinges by
which two enormous ice aprons were attached, one on each side, to
the triangle.
The object of these aprons, which will be presently described, was
to present an inclined surface on each side of the triangle of piles on
which the impact of the ice should be received. Any obstruction
opposing a vertical surface to the action of the ice would be soon
crushed to pieces or ground away, whereas by presenting an inclined
one, the ice would slide up on it and be broken to pieces, and be
thus made to pass off harmless from it just as the soil does from the
plowshare and mould board.




THE ICE BREAKERS.                     41
To protect the piles from the scouring action of the current it was
necessary to provide some means of keeping the current from them.
To do this with broken stone would be very expensive as well as
unreliable, and would besides create an obstruction much larger
than the pier, which would be difficult and costly to remove after
the masonry was completed. By planking the ice-aprons down
their inclined sides to the very bottom of the river, the current could
be deflected by them from the piles below, and the ice from them
above, and thus both objects be attained. This was done.
The ice-aprons were two hundred feet long and sixty feet wide.
It was necessary to place them beneath the water at an angle of
forty-five degrees, and with the lower edge or side of each resting
on the sand, and to make them of such strength as not only to
resist a powerful current, but also to withstand the great pressure
of the ice which might by the fluctuations of the stream be made to
impinge as low down on their sides as to the middle of their surfaces, as well as at twelve or fifteen feet above that point.
The frames of the aprons were made of strong squared oak timber
placed transversely at intervals of eight feet, so that the upper end
of each one of them would rest by the side of a pile, and on the
longitudinal timbers before mentioned.  The transverse timbers
were each sixty feet long, and were held in place by three equidistant string pieces, each two hundred feet long, bolted beneath them.
Two of these skeleton frames were thus constructed on shore, above
the works, and were launched with sufficient pine timber beneath to
float them. They were then towed, one to each side of the pile
structure, and the end of each transverse timber on the side next the
piles was placed on the longitudinal timber or hinge before named
and secured temporarily to them by chains. The outer edges of these
frames were then secured to barges placed alongside of them, and
the pine floats under the frames were then taken out. In this
position as the two frames lay on the water, they were planked
with three inch oak plank. On that part where the ice was expected
to impinge, No. 16 sheet iron was placed over the planking. A
space on each apron about twelve feet wide and extending their
entire length was thus covered with iron. Below this iron covering
some openings were left through the aprons for the current to pass,
to prevent the formation of a bar of sand below the structure in
the eddy that would be created by the ice aprons, after they should
be in place.




42                  THE ICE-BREAKERS.
When the aprons were both completed the lines holding up their
outside edges at the barges,were simultaneously cut away; these edges
then quickly disappeared beneath the current and were swept by it
to the bottom. Both aprons assumed the desired angle. The upper
extremities of the transverse timbers forming them, then rested
upon. the longitudinal timbers forming the hinge by which their
lower ends were rotated down to the bed of the stream. The upper
ends of these transverse timbers were then each bolted to its respective pile, and that portion of the sides of the pile system extending
vertically above the aprons, was planked with two or three strakes
of ten by ten oak timber, at the part nearest the aprons, and above
that point with lighter oak plank. At the apex of the breakwater
thus formed, about one hundred and fifty cubic yards of rubble
stone were thrown in, to thoroughly close any space left between
the upper ends of the two aprons.
This structure sufficed to completely turn the ice during the
winter and made a thorough protection to the works and barges
about the pier. A deposit of sand rapidly formed behind the ice
aprons which gave great support to them, whilst they in turn, protected this deposit once formed, from the action of the current.
Before our magnetic telegraph was erected, the ice was so heavy
for several days as to completely suspend intercourse between the
workmen at the pier and the shore. This contingency had been
provided for by provisioning the men with two weeks' rations and
providing them with bedding.
During the greatest severity of the ice, Mr. McComas, who remained at the pier, continued to operate the sand pumps, and every
morning and evening reported the progress of the work in a conspicuous place and in characters so large as to be read by telescope
from the shore. The closing sentence of the report was constantly
"Ice-breaker all right."
This structure was duplicated at the west pier with equally
successful results. Both ice-breakers are still standing having successfully withstood the April flood which attained a height of
twenty-six and a half feet above low water mark, and although the
current is much increased by them and the river scoured out in
proportion, the original angles assumed by the aprons seems to be
almost entirely unaltered.




CONCLUSION.                      43
CONCLUSION.
I avail myself of this opportunity to express my thanks to the
several gentlemen assisting me in the various departments of the
Engineer and Construction Corps of the Bridge, and to commend
them to the kind consideration of the Company, for the faithful
and efficient discharge of the important duties assigned them.
Respectfully submitted,
JAMES B. EADS,
Chief Engineer.








APPENDIX.
SPECIFICATIONS FOR CAST-STEEL WORK.
The steel shall be of the kind known in commerce as Crucible
Cast Steel. It will be required to stand the following tests; and
failure to stand anyone of such tests will be sufficient cause for the
rejection of the piece.
The staves composing the tubes will be required to stand a
compressive strain of sixty thousand (60,000) pounds and a tensile
strain of forty thousand (40,000) pounds per square inch of section,
without permanent set.
They must stand a tensile strain of one hundred thousand
(100,000) pounds per square inch without fracture.
The modulus of Elasticity shall not be less than twenty-six
million(26,000,000) nor more than thirty million (30,000,000) pounds.
This variation should be avoided if possible; in which case the lower
amount will be preferable. If a variation occurs in the modulus,
bars having the same modulus must be selected in making up the
tubes, so that one side of a tube shall not have a greater power of
resistance than the opposite one. Those having the same modulus
shall be placed in the same tube. Each bar will be tested by the
contractor, and the modulus stamped on it by the Illinois and St.
Louis Bridge Company's Inspector.
The steel pins will be required to stand without permanent set,
a tensile strain of forty thousand (40,000) pounds per square inch
and an ultimate tensile strain, without fracture, of one hundred
thousand (100,000) pounds. As it will be inconvenient to test these
pieces, the Engineer will require to have two or more of them made
in one piece, and of sufficient length to cut from the middle or ends
of the piece a sample for testing. In such case, failure of the sample will cause the rejection of the entire piece.




46                       APPENDIX.
Rods, bolts, eye washers, rivets, &c., will be required to bear an
ultimate tensile strain of one hundred thousand (100,000) pounds
per square inch without fracture, and forty thousand (40,000) pounds
per square inch without permanent set; such parts of the work will
not be tested in tension beyond forty thousand (40,000) pounds,
sample pieces only being subjected to ultimate tests. Such tests as
in the judgment of the Engineer or Inspector may be necessary to
detect flaws or other imperfections, when the pieces cannot be conveniently subjected to trial in the testing machine, may be used,
and any flaw or other imperfection existing in any piece, will be
sufficient cause for its rejection, if in the opinion of the Engineer or
Inspector it is injured thereby.
The one-quarter (1-4) inch plate steel for enveloping the
staves will be required to have a resistance to compression and tension, without set, equal to forty thousand (40,000) pounds per square
inch, and an ultimate tensile strength of one hundred thousand
(100,000) pounds.
The staves must be so accurately formed that when six short
sections of the same bar two (2) inches in length are bound
together by an elastic hoop, they will fit accurately at the joints,
and form a true circle seventeen and half (17 1-2) inches outside
diameter. They must be as straight as it is possible to make them,
without planing them.
Steel templates will be provided by the contractor, under the direction of the Engineer, and will be verified and stamped by the
Engineer before being used. The various parts of the work must
be made to fit these templates with the greatest attainable accuracy, and it will be the duty of the Inspector to reject any piece
which in size and direction of its parts, shows the least imperfection.
All holes through the steel work must be drilled, and all bolts
turned, unless otherwise directed by the Inspector, in writing.
Wrought-iron bands on tubes must be turned on the inside, and
faced- on each edge, as shown in drawings, and must be heated and
shrunk on.
The steel pins will be accurately finished according to the drawings. The central part, where it is reduced in size, will not require
turning off. The conical portions must be large enough to fill




APPENDIX                         47
tightly the holes in the tube couplings. The couplings for tubes
may be of rolled steel. The portions next to the main braces of the
arch, must be true and parallel, to insure accurate contact and adjustment of the braces to them; but these surfaces and the outside
of the couplings need not be finished work.
The surfaces of these pieces coming in contact with each other, or
in contact with all other parts of the work, except the main braces,
must be accurately finished.
Pieces of the material of these couplings will be subjected to the
same test as the pins.
Only the six (6) bolts nearest the centre of tubes, extending
through and through, are to be put into them before erection. The
caulking of the iron bands on the tubes and the remainder of the
through bolts, must be put in after the superstructure is erected. Four
(4) one and three-eighths (1 3-8) inch steel bolts will pass through the
ends of each set of main braces, and will be tapped into the tube
couplings. This work must also be done after erection of superstructure; the object of these bolts being to prevent any movement
of the braces around the pins and of the braces on each other.
The tube envelopes must have their edges planed and brought
closely together to insure accuracy of diameter of the tubes before
riveting on the butt straps. These latter will be caulked.
Each piece, after being examined and accepted by the Inspector
is to be covered with a coating of paint or other material, as may
be directed by the Inspector, to prevent rusting.




CHEMICAL TESTS OF GRANITE.
CAPT. J. B. EADS, Chief Engineer, Illinois and St. Louis Bridge
Company:
SIR:-I have submitted the samples of granite to a series of experiments, with a view of determining their capacity for the absorption of moisture; the effects of the action of alternate freezing
and thawing; and their resistance to decomposing influences, with
the following results.
The absorptive per cent., or the amount of water taken up by
one hundred parts by weight, was determined in each instance by
immersing the sample in boiling water for fifteen minutes, and
removing the superficial moisture before weighing.
RED GRANITE.                 GRAY GRANITE.
P. r cent. of                 Per cent. of
No. of                abso;rtion in                absorption in
Expe'mt               100 parts by                  100 parts by
weight.                       weight.
1.194700.341673
2.189670.341942
3.233641.345082
4.253420.351791
6.275892.364073
6.233641.351912
7.195640.342510
8.194520.342089
9.194591.341783
10.194702.311599
11.194711.341690
12.194794.341687
After the twelfth trial the absorptive percentage was found to
be constant and therefore the absorptive power of the red granite is
determined to be.1947 of one per cent. in every one hundred parts
by weight,,and the absorptive power of the gray granite is.3416 of
one per cent. in every one hundred parts by weight; a difference
of.1469 of one per cent. in favor of the red,




APPENDIX.                         49
The per cent. of loss was calculated from the effect produced by
the disintegrating power of the sulphate of soda whilst crystallizing.
A boiling saturated solution of the salt being used in which each
sample was immersed for fifteen minutes, and when taken out the
sulphate of soda was allowed to crystallize thoroughly, then the
salt was washed out with boiling water and the loss in weight
noted.
RED GRANITE.               GRAY GRANITE
Per cent. of                Per cent. of
No. of                loss in 100                  loss in 100
Expe'mt                 parts by                     parts by
weight.                      weight.
1.566511.335821
2.456282.343314
3.524111.369341
4.614235.341107
5.767891*.380140
6.513218.344316
7.524106.339149
8.531620.338555
9.541510.336032
10.559102.335619
11.565901.335739
12.566010.335793
After the twelfth trial the percentage of loss appeared constant
being for the red granite..566 of one per cent. and for the gray.3357
of one per cent. or in the proportion of 88 (gray) to 147 (red) as
calculated from the specific gravities. The specific gravity of the
red granite being 2.62038 and the specific gravity of the gray
granite 2.64224.
The disintegrating effect of nitre was tried in the same manner; the average loss of the red granite in five experiments was.5597 of one per cent. and the average loss of the gray granite was.38994 of one per cent.
The great disparity between the absorptive and loss percentages
of the red granite is explained when the large size of the crystals
of quartz and feldspar composing it is taken into consideration;
the loss being due to the bodily separation of the crystals from the
surface of the specimen. In a specimen, the surface of which was
ground, the loss by the disintegrating force of the above solutions
amounted to.3089 of one per cent.; and a hammered specimen
treated in like manner lost.384702 of one per cent.
*Loss partly mechanical, by handling.




50                      APPENDIX.
The gray granite further suffers a loss from the solvent action of
water alone; a sample of it immersed in distilled water for a number of days at the ordinary temperature was found to give to the
distilled water traces of lime and the alkalies, and the bottom of
the vessel containing the water was covered with a delicate layer
of silica. This was not the case with the red.
The disintegration of granitic material may be produced by the
agency of chemical change, which change is undergone by some
foreign substance accidentally present, such as iron pyrites, garnet,
etc., or the disintegration by chemical agency may proceed through
the decomposition of the normal constituents, mica and feldspar.
The disintegration of a granite may be safely assumed to commence always in the chemical decomposition of the feldspar when
foreign substances are absent. The decomposition of the feldspar
may be effected in different ways; first, by water containing free
carbonic acid; second, by the action of sulphurous acid; and
third, by water holding traces of alkaline or other substances in
solution.
The red varieties of fieldspar are less affected by these agencies
than the white varieties, since the red varieties contain a large
quantity of sesquioxide of iron, and in some cases the alumina is altogether replaced by sesquioxide of iron; whenever this occurs the
feldspar obstinately resists the decomposing agencies above recited.
The superior durability and resistance to chemical change of
granitic material containing the red varieties of feldspar over that
composed of the white varieties of feldspar is clearly shown in the
case of these two granites.
The red granite lost after three days immersion in water containing free carbonic acid frequently renewed,.0913 of one per cent. in
one hundred parts by weight; the gray granite treated in the same
manner lost.64087 of one per cent. in one hundred parts by weight.
Treated with a weak solution of sulphurous acid for three days,
the red granite lost.1456 of one per cent. of its weight; and the
gray granite.4253 of one per cent. by weight. Exposed to an
alkaline solution for four days the red granite lost.2287 of one per
cent. of its weight, and the gray granite lost.7349 of one per cent.




APPENDIX.                       51
of its weight. An atmosphere of carbonic acid or sulphurous acid
gas causes an appreciable change in the gray granite, whilst it has
apparently no influence on the red. In recording the results of the
examination the term "granite" has been retained on account of
its familiarity and common application.
With a view of determining the joint effect of the mechanical
and chemical disintegrating agencies, the samples, after treatment
with carbonic and sulphurous acids and with alkaline solutions,
were again severally subjected to the action of boiling water, (to
determine the increase or decrease of the absorptive per cent.,)
and to the action of the sulphate of soda solution (to determine the
loss percentage).
After treatment with solution of carbonic acid the absorptive per
cent. was found to average in the red granite.2(37 of one per cent.
and in the gray granite.6601 of one per cent. in one hundred parts
by weight; the loss percentage averaging in the red.53041 of one
per cent. and in the gray.9764 of one per cent. After treatment
with solution of sulphurous acid, the absorptive per cent. averaged
in the red.2399 of one per cent. and in the gray.7731 of one per
cent; thebloss per cent. in the red was.63021 of one per cent. and
in the grey.8999 of one per cent.
After treatment with alkaline solutions the absorptive per cent.
of the red granite averaged.21227 of one per cent. and of the gray.38419 of one per cent; the loss percentage being respectively.70121 and.97929 of one per cent.
The destructive forces, therefore, most to be feared are not those
of the rain and the frost, but the acid-laden air of the great city
and the alkaline waters of the great river. These are constant,
those only periodical.
I am, very respectfully, your obedient servant,
FELIX McARDLE, M. D.
ST. Louis, March 30, 1870.




52                      APPENDIX.
AUGUST 4th, 1870.
CAPT. JAMES B. EADS, Chief Engineer, of llinois and St. Louis Bridge
Company:
DEAR SIR:-It was a peculiar task with which you have honored
me by asking a chemical examination of several specimens of
granite with a view to determine as far as it could be done, not only their comparative value as a building material for the piers of
the bridge under your supervision, but especially whether or not
they might equally and successfully resist the decomposing influence of an atmosphere, impregnated, like ours, with the smoke and
gases produced by the combustion of the vast amount of fuel consumed in a city of such dimensions as St. Louis has already acquired,
and which will continue to increase with her manufacturing interests.
Would it be possible to arrive by experiments of but a few weeks'
duration in the laboratory, and by the aid of chemical reagents, at a
result in any degree similar to the slow but steady and continuous
action of the combined influences of moisture, change of temperature, the mechanical force of wind and rain, and the obnoxious and
deleterious gases, which are engaged incessantly in the immense
laboratory of nature in producing the final decomposition of all
matter? It was obvious, that if chemical researches would yield
any results at all, they could be only approximate, and would have
to be produced by substituting for the slow action of time, the
strong and immediate influence of the same, or similar forces as
act in the atmosphere but in a more concentrated form, which,
although they could not insure precisely the same condition of
things, would at least test the resisting power of the materials to
which they were applied. With such reflections I commenced my
investigations.
There was a considerable difference in the external appearance
of the three specimens; while the Richmond Granite, with its fine
grain and smooth appearance, seemed to be suitable, almost, for
fine works of architecture, it shared with the granite from Maine,
only the similarity of its gray color; for the structure of the latter
was much coarser, resembling in this respect, more the sample from
the Iron Mountain, which again differed from it widely in regard to
its pinkish color. The firmness and compactness of the rock seemed to be the same in the three samples, which were broken up into
small pieces of about two square inches each, and thus exposed to
the action of the reagents, the result of which is hereby submitted
to your consideration.




APPENDIX.                       53
Although the rock of the piers will never be exposed to the action of acids, I thought it would manifest at least its quality, if their
influence upon it should be ascertained, and consequently immersed
a piece of each separately in diluted muriatic, nitric, and sulphuric
acids. After ten days' immersion, they were examined and found
uneffected; then they were boiled in the same solutions for about
an hour, replenishing the evaporated water. After cooling, no visible change could be discovered on either piece, nor did its strength
seem to be the least impaired; the remaining liquor however,
from the Iron Mountain granite contained iron, which must have
been uniformly dissolved from its surface, for the rock was appalently as firm as ever, and even stood well the following accident.
While one of its specimens was being boiled with diluted nitric acid,
it became accidentally neglected, the liquid evaporated entirely and
left the granite dry and hot; but nevertheless, its surface remained
intact, and even its sharp corners had retained their firmness.
I had received but a small quantity of Maine granite, barely
sufficient for the foregoing experiments; it could not, therefore, be
included in the other series of tests.
In order to determine the probable action of the gases produced
by the combustion of fuel, which are mostly sulphurous and
carbonic compounds, several pieces of each specimen of granite were
treated with a slow current of a mixture of street gas, sulphurous
acid, carbonic oxide and carbonic acid, and as it is recognized in
nature, that moisture increases the force of the most powerful agencies for decomposition, it was believed that the destroying efficacy
of these gases would also be increased by keeping the rock moist
during the whole process. This experiment was continued for
nearly two weeks, and when the specimens were taken from the apparatus, their surfaces were closely examined, but no sign of decay
was visible, nothing could be discovered to indicate that the adhesion
of the component parts had been loosened and the validity of the
rock destroyed; its edges even remained firm and sharp.
In the foregoing experiments, the gases referred to were administered pure and without any admixture of air, smoke, etc., but as it
might be inferred that their action would be modified when applied
in the same state as they are discharged from the furnaces into the
atmosphere, I considered it my duty also to determine the concentrated action of such mixture. The best place for such a test
seemed to me the front of the flues of a boiler, through which
the gases pass directly from the fire, with all the smoke and




54                      APPENDIX.
undecomposed air, and where the temperature rises and falls from
sixty to two hundred degrees, and more according to the amount
of fuel and the rapidity with which it is consumed. The specimens intended for this test were put into a dish with water, and
then placed into the front of a boiler-flue, so that all the smoke
and the gases had to pass over and touch and envelope them. The
water was allowed to evaporate and dry up occasionally before being replenished, and the rock was thus exposed to all the imaginable
changes attendant on such a situation. On the expiration of twelve
days, the dish was removed from its position. The rocks showed
no signs of decomposition and were hard and firm yet. It must be
mentioned however, that I succeeded in loosening a few fragments
from the edges of both samples, which from all appearances seemed
to have been disintegrated by the sledge when the samples were
broken up into smaller pieces.
It remained yet to test the influence, upon the samples, of frost and
mild temperature. One piece of each granite was soaked in water
and then repeatedly exposed, alternately to the low temperature of
a freezing mixture, and the atmosphere of 55 deg. F., before they
were placed into the dish in the flue where they remained several
days. The treatment had made no apparent influence upon the
rock.
If you will now recapitulate and compare the results of the experiments fully and minutely described in the foregoing paragraphs,
you will find that there was no test made which has not terminated
in the favor of the resistive power of either specimen of granite,
against the decomposing influences of the reagents, and if any such
experiments and tests are of any value whatever, or give permission
to draw conclusions in regard to the technical value of the material
experimented upon, it is manfested beyond a doubt that both specimens of granite from Richmond, as well as from the Iron Mountain are capable of enduring and resisting the combined influences
of all the deteriorating gases, which may contaminate the atmosphere of our city.
Hoping that the manner in which I have treated this matter will
meet with your approval, I remain most respectfully,
Your obedient servant,
[Signed]                       ENNO SANDER,




APPENDIX.                      55
NAMES OF THE MEN WHO WERE EMPLOYED IN THE CAISSON OF THE
EAST PIER, FROM THE TIME IT ENTERED THE BED OF THE
RIVER UNTIL IT WAS FILLED WITH CONCRETE.
FOREMEN:
JAS. DONNELLY,                         C. W. BROWN,
B. O'KEEFE,                            FRANK JORDAN,
PETER WHITE,                           JAMES LOVE.
MACHINISTS IN CHARGE OF THE PUMPS:
PATRICK GLINN,                             C. MALBIN.
LOCK-TENDERS:
H. J. HARVEY,                          THOS. CAVANAUGH.
WORKMEN:
W. BURNES,                 JOHN WILEY
JAS. LYONS,                 WM. LOOMIS,
W. S. HAWKINS,              JOSEPH BOWERS,
P. McPIKE,                  ED. RAND,
P. PETERSON,                M. IRWIN,
JNO. FLINN,                 WM. LAFFERTY,
WM. McKENNA,                M. O'KEEFE,
CHAS. WAGNER,               FRANK SANFORD,
HENRY COOPER,               M. MEYERS,
AUG. DONKA,                 WM. MURPHY,
P. DUMAS.                   M. McDERMOTT,
P. GALLAGHER,               D. FORD,
WM. FOSTER,                 F. FORD,
J. McDONALD,                D. BRADLEY,
W. McCARTNEY,               LOUIS GIBELER,
DENNIS McCARTY,             A. DEFORNEY,
P. FORD,                    N. NICHOLS,
ANDREW COOPER,              CHAS. SHAFER.
JAS. SHEEHAN,               WM. CONNERS.








t~ PPlate ][
A I'deau*'Jarbr anr*nf'terils.",scalS< Y^fan I  totbe1root.. Screwrs I keep l-ce cfasson level 6efore reachuing sand
H  1/Mrema/zz JIs pporlbmolinfzctr laeses.W. 7                                                                                                                          ide piles for cais son.
if Skafls fIr esitnIrt  an ttdy.oppi( c, raveh~ic prcihases.                                                                                                      Z Trisses for oide piles.
I tac/v or Arazaieyja7-ckr raising g loering n aterials.  FOR SINKING CAUSSON AND LAYING M/LASONRY    OF THE EAST PIER OF TIE ILL.& ST.10O1IS BRIDGE.             AA.MB ing room.
P Z'enm eS dnrc aier puaamps.                                                             Jasne  3 B. E a mLs En ie esr.                                          (ccrorm ways for ce7needt a7med stoner cars.:     o   an  In':ut......ot








..........       I... 
I   .   I -.....   - I. -.......
I
. 
" 
.,. 
I. 
i. 
I I I i!.... 
.; 
"
-....:.. —_ —-.. —-.-:
I.: il.'..-.
-.....-.......-.
I'     I I I 
I............. -''C, -R -,', ",,.-',:,. -......"-. I-.,,..: -.
I;  i! 
: 
fi    I 
   4. 1!!:.. 
-             r-  It 
i 
iI,,
I 
-- -:1" —-,I... -.. —
--— 4i                  I -.'-  lc....-1 ] " -:!-:.'.-.'... -'..'
i:  
-i  c.,.......
(1)            - 
I'..:!'. -..': -:' -:
+             Cd 
it 
-...
I.I 
I                  -,   it 
t i  (L)  a).'.,....
I; 
—'''-:....;. 
I. I I.; —.4 
(Z.)  C4 —-i.I 
"i 
';'.:'.'-::-.'.. ...'.' ".
I.,...,-'I-..
I 
-....
I 
i 
i......
-.".1.... 1
&   P-4.': 
I 
I.iI 
IL-H.
  P -IIIII. 
:
C.D         I 
1-4  C  P -.. 
-f 
I 
I                     C.) 
.  - , 
I     -;.(:::i,-7.- -.         -t..
4 
-,
i.,
   -. 
 
iI    i! I 
i.! 
i.'
i 
P-4  C8  g  ' 
I: . 
-  I 
11  Cd     O'  Z4 
u'  :4 
(:.-'- "...'........,.;:.::,:.. -!'t,..
i 
C)  _';  1 —o;11  -'!: I.        7-)  C.)..  (a. 
I.- "... "....-.,..'....-.... 1.,.,....:.. "I':.,!-.,. r"; - - -..'...- -.:.,.:.,.,I...:: 
I'"
— 1..:...I.:...I -
71  11-I  iE'  C')  -d   ..:'.-.-"....I.,..I.. 7... -.' - -:.:: I. I -.:
C 
-,
-'...'-J1.    8 
lm   — i.'-, 
I 
I. 
"  P- 'c "
-1:'-'  D)  .,:.. I.,... -.,.....:....  I 
I i 
I..I;                          -'...,,-':.
I. - -'."-.. -. I-S...::"" — ".     -....:! 
 
I.....
-i             I  1  I 
I;      I I..I;i'
I 
i 
.; 
I  I....I. 
I.. 
I 
I  I 
-.
I:  I 
I.I 
I..:::.: -.I! 
i.. '''''.:.:. -,-.......:'.'..I.'...'....-.,....  I''"4".-.:.'" -, —'. -  - -.'.!.'-.-...  -'...'          ",'',  -... -
i I  1 1 
-...
/ —'...-......,,......
4".......'. ..
I 
-...
I I' 
I........ -.
I...111..I'''..
-..:':.,.....:'::'-':-.:.. 
I...'' —-.. I....-.I,.:...::...
I..
- II   i 
 
i....................
I..... - -..7. 
t 
I 
 
I!. 
I 
111111111111111.      I i I  I I I L ii   M    
-- --- : —- -—:
M 
I   .   - 
oil  - 
-- I" 1/' I 
I,
",,- I 1.    co 
. 
I.,,
-'
2 
'I"i Li 
< I 
--- 
I 
-T- 
. 
-1''-.  —-!
--—.. --:.;-.
I % 
- 
u 
I 
I, .,. 
I  -,.....;.-- i. -."':::
-,:' - it.-.;-i 
i 
M      r 
 —- 
III,,.q 
L I  Ili   -::'  T   lil  M.;   11'. . "'!: I;.:"I -...'.-— I       -   iI   11'1'1$j.- 
= — ",11 
-               ','.. - 
" 11   1111 -'
D 
.
-
I -- F:  -.  a), 

<      9 
- 
1-1 
-   I,         
-- -I1- M!
IBal tic.   I I. 
9). 
" 
-:!I 
ll 
- -'   1   ;i;llj..-   i' I. 
----          l 
OJ)  I 
R, cm -,
I          I-' I, 
-. 
I -            ii. " I 
t"Y   -.   I " 1'. 
 ilj i"" 111111   I   ill   11,)I,, 
i,   - -I   I   I - 1.,, 
L 
- Ii 
-.  I  '  - 
co 
-E-     - 
-. 
.' ",Fiji   " A,,,- 11111111,! 4 
go   C Z   CL.  44 
I.'     ,!  L lill.   ;")!I. 
I 
I 
11  I,  5:;T4  rz- 
--—. -: 
-i                    
I'        - 1j 
C)'.  I "   -—,  11  " 
----, —-     -. —--.
[::   9  -  =) ul 
-- 
4 
-IT   —'  I ---- I.1; 
LL-.0- p   - 7-) 
- 
- -.-;  i ii,,,,I .'   10 - -.-   -—: 
M 
<    - 
el. " 1; ,,,
I -;!;  --..1. 
-r           -1 
I Im- Ii!".
"
I
I








m no Wo >
~ -~l             3:~; |~ I::-    | |       | |        ||,1W11
|'gIII  I.lD  IIIIIIIII11r11 IiEDII j  IN  I  IjI I II ij f l!D lDaIII'lb,  I  I  I  III I,I *i llll X,  1'...........................1.
~I:               1.n' 1l................  l.....L..-..........  c...........:.
M... 
I~ —---
e I-1cr,~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~u'U~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~L?'....,;II' 0E~~~~~~~~~~~~~~
~........................1~/...........~-.....~..... -~~I....'....J...~...`...........~f~~ ii~~~~2~- ~~~...T~-.  ~..~.o~ ~o~r~ ~ r~~~r~~~~~.~*Tj~
iii ~ ~ ~ ~ ~ ~ ~  ~  ~  -           m@lll II~'   rlEI'iilllim~
~ Mi 
FS1                                      i!~'
~'~~~~!i~                                                  O








..................atea]IT...............................~..................l     1~l.~...:~.'~~.1..........................................................ir~.........................I..............
uj..................~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~:::;I
~i..r~~~r~~~..:;I.........
sr-   I                        ~~~~~r~~~~:,....,..~~~~~~~.........
Lj  S~.:t:::::::::::::::::::::::::::::r::1: — ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ — c~~~~~l::::~~~~~: ~ ~ ~ ~:=::::::~~~~~~~Ir:i  E   sl  Ii. ~!~     ~    ~   ~   ~   ~~K
Scale-.. ~ ~    ~         ~        ~        ~        i'l                            ":I
-r.  Mai Sh'aft
K. Side, Shafts.        ~:~:;~~:~;:F.,o~
-L. Apes for........Aird...Sai...d -A mps.'":~~~~~~~~tC S"F & P                                     V
r   i~- i:  ~~~~~~~~~~~~~~~~~~~~~~~~                                                I —:~~~~~~~~Tibr i
r-i:                               i:-~~~~~~ams.etsEngner








z
a am  194 f;; ~ ----— i-.__ —--—  —,z. — —- r-_
bZ
OhlN ~ a I —
r                                      8   I t!rr              lit,

1.r;
I;'~: 
"

Ii ~;r*!~  %
"
4                                                                                                                  --:-, —;T;;;,-'-- —-a~ ,,
CP):I a

o
i
/:::: 
"
o ;a
-f I
I:I-T~T-3::                                                                   r
0-.  I%                                         _
s " t
%I:6:::::::::::::: r:II i\i (.W 
m,i: 4
" jii i i
rs
I/jili E ~ i:
I::I!j;7\~: Y
jib 3 x r
i: — I.: i I-iiti-t~;r ---------- I; —.)-::--'
-.1.~~.: 41 —-----
Siil;'"??::::.'::f::::::'.'.",I,":
i;"v
n              Elr
I.Bgsf4tCs: 
I:b::d:.::::::...::::.::: t —3: tlII.j   :                                          ii                                                          I — —
=
st r
o
CO->
a
V, L1
W il
m %I/::K rn                                                                      = —-- —,,,,:L.,,L /,/ Fil~
d
Pi
CQ
I.I lil I.I inll -3:':          i:l-i —

"o I "II
II:.'.:'.'.:::::::::::::I
" lulJ
ii 311) ii:
ili
                                                                                            Q1:::::.'.'.:::':::::,:):
I::  i 
 ii
r'li -l.r~
jl): 
ii I
j'i! i
i "
m i1: i P
C).ui_.
 illl
 ~i-I::::::::,:::'.'.':::.I::I~~~~~~~~~~~~~~~~ I~
O ~s -1 —
3        z                                                                                         V)
lil iii iII I
:::::::::::-:::fl:
I
s
o, I t
CJ 1
C3  g
31: 31.;7-;;;,a;TT;;l-;~IS:I~~ ~~ ~~ ~~- ~~ ~~
dl
la
(O cJ~
icli.:
""'I
i i
13ii6f6S'll
:"\3:II1 rl~
-I";opa 5111'j
-~- j I__I_1
4
t  —i_ -— __ I%(gT_:i
i;y Ia:~~
k:
OJ ij,(; —-- — I
lli!
X ii::- —--— ~_
`i!r                           5~_-!1 — ---— = —-=I-;;
~41(hr:.~L i;-ei~a
tIi:
P)
ri~... z
 _..._. --------- ------------------- ~ —---------—  —- ~30' --------------- ------------ --------- --— 








..Plate V,                                                           L
I   fa7.ui S/a/f                                                               Sc ale                                                               P Airchazbel
K  9Sc  Zhafts                                  "  O                   lo o'                  20         s          30                              0D  J 7cinbe 7 ei6desl._.}:.
L _Ppes ferc r -and SandcP]mt..'fl-4 4-4 —--— L —4 —----                                      I I I -,I | | 4-J —                               R Thmiber Jec,~:
M _Tron crders                                                                                                                                      S S. Iroc nZne/ope
N N fron _Dec                                                                                                                                       T kicee Sides......_________________                  ~JAMES               B. EADS, Chief Emn imeei;....... ALAugu                        tt a  & Co., St. Lois.








ILLINOIS AND ST. LOUIS BRIDGE COMPANY.
REPORT
Of THIB
CHIEF ENGINEER
OCTOBER, 1871.
ST. LOUIS:
IlBOURI DEMOCRAT BOOK AND JOB PINTING HOUSE.
1871.








CHIEF ENGINEER'S REPORT.
ST. Louis, October 1, 1871.
To the President and Directors of the
Illinois and St. Louis Bridge Company:
GENTLEMEN-I have the honor to make the following report:
MASONRY.
The total amount of masonry of every kind, including brickwork, laid up to this date is sixty-eight thousand seven hundred
and thirteen cubic yards. There remain yet to be laid thirty-three
thousand nine hundred and two cubic yards.
EAST APPROACH.
The foundations of the five piers and towers of this approach
have all been built, and are now completed ready to receive the
sandstone, excepting pier No. 2, the red granite being laid up,
including the wash-course, on all of them with this exception.
These approach piers (excepting No. 5, to which the towers are
attached) were all sunk by the same process used in sinking the
large ones. This was believed to be more certain and economical
than by coffer-dams, the method used on the west approach. The
result has fully confirmed this opinion. With pier No. 2 the cofferdam would have proved very expensive, as this pier encountered
in its descent the hulk of an old and strongly-built ferryboat. Its
planks were three inches thick, and floor-timbers four by eight
inches square. This hulk had to be cut away to permit the descent
of the pier, which has gone directly through it, the kelson of the
wreck having been intersected by the caisson. This not altogether
unexpected difficulty has delayed the completion of this pier about
three weeks. It has now passed this obstruction, and needs to be
sunk only one foot more to reach its final destination.




4             REPORT OF THE CHIEF ENGINEER.
It has not been deemed necessary to sink the foundations of the
first four piers of this approach deeper than twelve feet below extreme low-water mark. The caissons used are made wholly of oak
timber, excepting only the air-locks and the spikes and bolts used
in fastening the timbers together. They are ten feet high, consequently they are entirely below low-water line, while the masonry
extends only down to that line. The surface of the wharf will be
about nineteen feet above the bottom of the foundation of No.
1, and about thirty-two feet above that of No. 4.
The towers and pier No. 5 are laid on a foundation of concrete
six feet deep, under which a system of piles was first driven to an
average depth of twenty-five feet. The concrete is based, fifteen
feet above low-water line.
The east abutment pier, resting on the rock at the depth of
136 feet below high-water mark, and standing between these
approach piers and the river; and the wharf pavement covering
the river shore, will, I think, be amply sufficient to prevent the
current of the stream from ever affecting the foundations of these
small piers.
From the towers eastward the upper roadway of the Bridge will
be carried by trestling to Fourth street in East St. Louis, with a
grade of four feet to the 100 feet. The railways will curve out to
the north from under the upper roadway, immediately east of the
towers, and will be of such grade and construction as shall be
determined upon by the railroad companies who have contracted
to use the Bridge, and who will pay for these railways. A sixtyeight foot grade will bring the railways down to the level of
the present tracks in East St. Louis, within a distance of about
3,000 feet.
EAST ABUTMENT.
This pier has had no work done upon it during the last four
months. It is completed on the river front to within two courses
of the wash-course. The wash-course is the one immediately below
the lower tier of skewbacks.
On the T wall the red granite is now being laid, four courses
being required to finish it ready for the sandstone. The sandstone
commences upon the wash-course, which course in the T wall is of
red granite, and corresponds in height with the wash-courses of
the approach piers on each shore.




REPORT OF THE CHIEF ENGINEER. 
EAST PIER.
The masonry of this pier lacks two courses of being high enough
to set the lower tier of skewback-plates. These will be laid while
the contractors are setting the plates on the west pier; there being
but one outfit of hoisting apparatus for carrying up the masonry
of these two piers, it is alternately used at each. At present no
work is being done on this pier. It will, however, be finished
considerably in advance of the wants of the contractors for superstructure.
WEST PIER.
The masonry of this pier has been carried up high enough to
receive the lower tiers of skewback-plates, where it must rest until
these are fixed against it. The stone-cutters are now engaged in
making the necessary recesses in the granite to receive the plates,
and so soon as this shall have been completed, these plates will
be set. This pier will be completed about the first of February
next. The necessary steel and iron work for it are in such a
state of forwardness that no further delay need be feared in its
completion.
WEST ABUTMENT.
The masonry has been carried upon the T wall or western face
of this abutment to a level, four courses below the springing of
the first stone arch over the St. Louis wharf. This portion of the
masonry (the T wall) is of sandstone.  On the river front of the
pier the masonry is finished to a level five courses lower than the
T wall, and is now ready to receive the lower tier of the skewbackplates. These four plates are now ready and will immediately be
set by the contractors for the superstructure. So soon as this shall
have been done the masonry of this side of the pier will be carried
up to receive the second tier of skewback-plates.
These skewback-plates are seven feet by three and two-thirds
feet six inches in thickness; and to insure an equal bearing against
the granite throughout their entire area, iron cement will be driven
in between them and the stone after they are in proper position.
This work must be done before the next courses of granite are laid
above the plates. These plates will sustain the entire thrust of the
arches. It is against them that the skewbacks of the arches wil
rest. The skewbacks are of wrought-iron, and will be firmly




6             REPORT OF THE CHIEF ENGINEER.
secured to the masonry by large anchor-bolts passing through them
and the skewback-plates into the abutments, where the bolts will
be fastened to cast-iron anchor-plates built in with the masonry.
Into the upper and lower tiers of skewbacks the cast-steel tubes
which form the upper and lower members of the arches will be
inserted. The anchor-bolts are five and three-fourths inches in
diameter. Some are made of steel and others of iron, according
to the different degrees of strain to which they will be subjected.
The lower skewbacks will each have four bolts and the upper ones
three. These bolts will sustain no portion of the weight of the
arches. Their object is simply to prevent any movement at the
ends of the arches which would otherwise occur from extremes of
temperature and excessive inequality in the distribution of load
upon the arches.
When the second tier of skewback-plates are fixed upon this
abutment, the remainder of the masonry of it will be rapidly completed. I see no reason now to fear any further delay in this part
of the work, and think the masonry of this pier will certainly be
completed by next February.
WEST APPROACH.
The piers of this approach over the St. Louis wharf are all
up to the springing of the arches which they are intended to
carry, except pier No. 5.
One complete sandstone arch is turned upon piers Nos. 2 and
3, and about one-fourth of an arch is turned upon the centering
on each side of this completed arch. The other two arches of this
approach can not be started until the T wall of the west abutment is four courses higher, and pier No. 5 is completed. This
pier stands in the house-line of the wharf, and is completed from
the rock to the wash-course inclusive, and is therefore ready to
receive the sandstone on the red granite base-course.
The foundation of the south tower is also completed to the same
extent, and sandstone is being laid now on this pier and tower.
The foundation of the north tower has just been commenced.
Between the wharf and Commercial street the masonry is well
advanced, piers Nos. 6, 7 and 8 being completed. On these piers
is now being placed the centering for the brick arches which will
surmount them, each arch having twenty-seven feet span.




REPORT OF THE CHIEF ENGINEER.              7
Piers Nos. 8 and 9 stand on opposite sides of Commercial
street, twenty-seven feet apart. Each is completed ready for the
centering of the arch, which will soon be placed in position on
them.
Piers Nos. 9, 10, 11 and 12, standing on the block between Commercial and Main streets, are all completed, and the three brick
arches (twenty-six and a-half feet diameter) upon them are likewise completed. The spandril spaces between these arches are
being filled with ballast to level up for the railway sleepers.
The vaults under the arches between the wharf and Main street
are inclosed by substantial masonry, and the rooms thus formed
are connected by arched doorways through the piers. A large
warehouse is thus created on each of these blocks, which will be
available for revenue. Arched entrances are provided to these
apartments through the piers on Main and Commercial streets.
The land over which the approach is carried further west, is similarly utilized for warehousing purposes as far as to the middle of
the block between Second and Third streets.
Pier No. 12, now finished, forms the eastern abutment on Main
street for the support of the truss which carries the roadways
over that street.
Pier No, 13, on the west side of Main, is built up from the rock
to the street level, and the sandstone work on it is progressing.
From this point west no large piers occur, except on each side of
the alley between Main and Second, and on each side of Second
street, at which point abutment piers are being placed to sustain
the iron trusses for carrying the roadways over these thoroughfares. The foundations for all of these abutments are built up to
within one or two feet of the street levels, and sandstone work is
progressing on one of them on Second street.  The others are
covered at present with hoisting machinery and wagon-ways for
transporting materials. From Main street to the middle of the
block, between Second and Third streets, the railways and roadways will be supported on strong side-walls of masonry, and two
rows of brick pillars placed equidistant between these -walls.
These side-walls are almost entirely completed, ready for receiving the railway timbers, from Main street to the end of the approach at Third street. Between Second and Third streets they
are still more advanced, being nearly completed for the upper
roadway.




8             REPORT OF THE CHIEF ENGINEER.
The brick pillars between these side-walls are well advanced,
east of Second street to the alley, one tier being nearly completed
to the level of the railway beams.
The entire masonry for the west approach, from the Levee to
Third street, will be completed ready for the superstructure, I
think, by the middle of January next.
SUPERSTRUCTURE.
CAST STEEL.
The Keystone Bridge Company, contractors for superstructure,
have made but little progress in their contract, owing chiefly to
delays in obtaining the proper quality of steel and iron for the
work. The contract for steel was let by them to the Wm. Butcher
Steel Works, near Philadelphia. Extensive additions, consisting
of buildings, furnaces, rollifg-mill, straightening machines, &c.,
were made to the works, and every indication seemed to encourage
the belief that the steel would be promptly supplied. Unforeseen
delays, however, occurred in getting the necessary machinery into
proper working order.
The first large forgings required by the Bridge were steel
anchor-bolts, five and three-quarter inches in diameter, and
from twenty-two to thirty-six feet long. The first bolts, when
tested, were found to be of inferior quality. Having been injured
in forging, they were broken by testing. Each bolt is required
to sustain, when tested, a tensile strain of 519 tons, without
being permanently elongated, being twice as much as the maximum strain to which it can be subjected in the Bridge. Before this test could be applied to the defective ones, sufficiently
to prove their weakness, the testing machine itself was broken
twice. This involved several weeks' delay. When the defects of
the machine were remedied, the inferiority in the bolts was fully
discovered. New mixtures of steel had then to be tried, and
greater care was used in its manipulation. Many bolts had been
made before the testing machine was repaired and could reveal
these defects.  These were, of course, rejected when tested, and
others had to be made in their stead.
In the novel operation of testing such large forgings, and in the
management of the machine itself when exerting such great strains
upon them, many unexpected accidents occurred, both to the
machine and to the instruments required for measuring the exten



REPORT OF THE CHIEF ENGINEER.             9
sions of the bolts and the amounts of strain imposed upon them.
For instance, a piece of one of the bolts, which weighed over 1,000
pounds and twenty feet in length, was shot out of the machine
like an arrow, when the bolt broke, and fell fifty or sixty feet distant; whilst the other end of the bolt reacted with such force upon
the machine as to break the piston rod or pulling bolt by reversing
the tensile strain upon it, thus driving it out of the ram at the
other end of the hydrostatic cylinder, and breaking by its reactive
force the fastenings by which it was secured.
The expense of testing was assumed by the Keystone Bridge
Company in its contract, and it in return sub-let this part of the
work, so far as it relates to the steel, to the Wm. Butcher Company. The design, construction and repairs of the necessary
testing machines and appliances are, therefore, a part of the
province of the contractors, they simply agreeing to subject the
various materials to the requisite strains under the supervison of
the Chief Engineer of the Bridge, or of his assistants, the instruments by which the strains are measured being all that is supplied
by the Bridge Company.
By these various delays several months were lost before any
bolts were supplied, capable of sustaining the strain above named.
These difficulties were, however, fully overcome, and twenty-four
bolts have been thus tested and received. Eighteen of these were
sent more than three weeks ago to the Keystone Bridge Works at
Pittsburg, to have screws cut upon them. Others are being made
from day to day at the steel works, and I believe they will now be
supplied as fast as they are needed. Twenty-six are yet to be
made.
Similar difficulties and delays were experienced in other parts of
the steel works. About four-fifths of the entire steel required, consists of tubes about thirteen feet long and eighteen inches in
diameter. These are each composed of six staves of the length
of the tube, and varying from one and one-eighth to two and
one-eighth inches in thickness, and each being about nine and
one-half inches wide. Several attempts were made to roll these
staves before the rolls were perfectly formed to accomplish it.
Each failure necessitated the removal of the rolls from the mill
for alteration to the machine shop, several miles distant. Each
of these rolls weighed several tons, and this usually involved a
loss of two or three weeks before they were in place again and
ready for trial.




10            REPORT OF THE CHIEF ENGINEER.
An option had been given the contractors to have these staves
rolled with a rib on each edge of the stave and projecting into the
tube, or to roll them without these ribs.  The steel company
elected to roll them with the ribs. After three or four alterations
of the rolls they determined to abandon the attempt to roll them
with the ribs. This involved the making of an entire new set of
rolls, ten or twelve in number, and when these latter were tried
they had to be twice or thrice returned to the machine shop before
perfect staves could be turned out with them,
In this way at least six months were consumed before a stave
could be offered for testing. When this was done the steel proved
inferior. Repeated changes had then to be made in mixing the
steel. When a satisfactory mixture was obtained it was only then
discovered that the same degree of strength was not present in all
the staves made from it.  This was believed to result from a
want of proper care in melting and in forging the ingots, and
in second heating for rolling.  No doubt the difference in the
the degrees of heat applied to the perfect and imperfect ones
altered the proportions of carbon and iron, or their relations to
each other, and thus caused a decided difference in the strength of
the staves thus made from the same formula. Of the anchor-bolts
about twenty or twenty-two which stood the test were carbon
steel, while nearly an equal number failed under it, although
many of the failing ones were made by the same formula as the
good ones.
These results proved the absolute necessity of using the greatest
skill and caution in the application of the requisite degrees of
heat in making carbon-steel.  The want of this extraordinary
care has caused great loss of time and money to the Bridge Company, and has doubtless occasioned serious loss to the Butcher
Steel Works. This unfortunate experience induced that company
to endeavor to find some other method which would insure with
less skill and caution a greater uniformity of product. With this
view, experiments were commenced at the works recently in
making chrome-steel under the patents of Mr. Bauer. These trials
were under the direction of Mr. C. P. Haughian, Superintendent
of the Chrome-Steel Company, and were attended with the most
satisfactory results. An arrangement has since been made with
the patentee for manufacturing this steel.
Chromium unites with iron and forms an alloy, similar in its
properties to steel. Chromium is quite different from carbon in




REPORT OF THE CHIEF ENGINEER.            11
some important particulars. It is a metal, while carbon is not.
It has little or no affinity for oxygen, and is not affected by
excessive heating, while carbon has a great affinity for it, and by
the application of heat it is liable to be burnt out of the steel.
One hundred trial staves were made last month of chromesteel, under the directions and from the formula of Mr. Haughian.
They were all beautifully and perfectly rolled, and there was no
failure in any one of them to stand the test required. This steel
comes from the rolls much more smoothly than the carbon-steel,
and it works quite as easily, being capable of sustaining a greater
degree of heat than the carbon-steel, it takes the form of the rolls
more readily.
Tests made of this steel by me, before the contract was made
with the Keystone Bridge Company, satisfied me that it possessed
qualities eminently suited for the Bridge superstructure.
In 1869 Mr. Haughian allowed me to be made acquainted with
the entire process of manufacturing chrome-steel. Commodore J.
W. King, of the Engineer Corps of the Navy, now Chief of the
Bureau of Steam Engineering, kindly volunteered to investigate
the subject for me, and on my personal pledge that Mr. Haughian's
trade secrets should not be revealed, Commodore King and my
chief assistant, Colonel Flad, were allowed to pass forty-eight
hours in the closest inspection of the works, during which time
they weighed out the proper mixtures, placed them in the crucibles, melted them, cast the ingots, and had the steel finished by the
hammer, all being under their immediate supervision. An elaborate confidential report was afterwards made by them to me of
their observations and experience.
As it it was however, a patented manufacture, and made by but
one establishment, to have required it to be furnished by my specifications would have been equivalent to compelling the Keystone
Bridge Company (whose contract with your company was made
before my specifications were complete) to forego all competition
in obtaining the steel; and as several other makers expressed the
fullest confidence in their ability to furnish an equally reliable
steel, it seemed but fair to state the necessary qualities which the
steel should possess, without prescribing any special formula that
would restrict competition.
I did not feel justified in assuming that crucible carbon-steel
of the qualities and forms required, could not be readily made,
when I was assured of the contrary by some of the most eminent




12            REPORT Of THE CHIES ENGINEER.
steel-makers in America. I was so fortunate as to be permitted
to make a careful personal examination of Mr. Krupp's great
works in Prussia, and also the mammoth works of Messrs. Petin
Godet & Co. in France, and was also assured by the managers of
both of them that our requirements were entirely practicable with
carbon-steel. I did not, however, hesitate at any time to express
my belief that the chrome-steel was most likely to meet the
requirements of the Bridge; nor am I justified now perhaps,
even with the experience developed at the Butcher Steel Works,
in asserting more than my opinion that carbon-steel can not be
made with as equal regularity and uniformity as the chrome-steel.
I think the sequel has proved that it was unfortunate for your
interests that a contract for chrome-steel was not made at first, for
the unsuccessful attempt to supply the carbon-steel for the Bridge
by the Wm. Butcher Works has seriously delayed its completion.
While this disappointment has resulted in great loss to you by
the delay it has involved, it must have inflicted serious damage
upon the Wm. Butcher Company.     The honorable disposition
shown by the latter company to discharge the obligations of their
contract with the Keystone Bridge Company, and furnish a quality of steel fully equal to their agreement, notwithstanding their
many unexpected losses and disappointments, certainly merits
your considerate notice. The President of the company, Mr.
Samuel Huston, has repeatedly assured me of the determination
of himself and associates to supply the Bridge steel as promptly
as possible; and that they asked no abatement in its quality, but
were resolved to make it fully equal to the requirements of the
contract, cost what it would.
Acting upon this commendable determination, that establishment has contracted to pay Mr. Haughian a royalty of $15,000
for the right to make chrome-steel for your Bridge; and I have
been assured by Mr. Huston that henceforth no other kind of
steel but this would be made for it.
From what I know of the manufacture of chrome-steel, and
from the tests of anchor-bolts, staves, and envelope plate-steel,
already made at the Wm. Butcher Works, from the formulas of
Mr. Haughian, I feel every assurance that the difficulties in the
way of supplying the steel for your Bridge are now surmounted.
The steel we are now testing is of a quality entirely satisfactory,
and the workmanship is unexceptionable. The tests made of its




REPORT OF THE CHIEF ENGINEER.            13
ultimate tensile strength are considerably in excess of the specifications. In compression almost any degree of resistance can be
obtained by the addition of chrome. To avoid unusual difficulty
however, in finishing the steel in the lathes, it is only made sufficiently hard to meet the requirements of the specifications.
IRON-WORK.
I regret to state that our iron-work is not being prosecuted by
the contractors with due diligence.
The Keystone Bridge Company contracted with other parties
(as I am informed by Mr. Linville, President) to make the main
braces which connect the upper and lower members of the arches
together. These members are scarcely second in importance to
the steel tubes which compose the arches. They are to be made
of iron, and I was informed by Mr. Linville that my specifications
form part of his contract with the sub-contractors. These specifications distinctly state how the iron shall be tested, and that it
shall bear an ultimate tensile strain of 60,000 pounds per square
inch. None of the iron yet offered by the sub-contractors has
proved capable of bearing over 51,000 pounds, and very few
samples have exceeded 50,000 pounds, and much of it only about
48,000 pounds. I have recently been assured, however, by Mr.
Linville that the sub-contractors will supply the iron without any
further delay, fully up to the standard required. As several other
parties have offered to furnish the requisite quality, and have
proved their ability to do so by the tests already made of their
iron, we have reason to believe that the Keystone Bridge Company
will not be longer delayed in obtaining it from the parties with
whom they have contracted, or from other makers.
FINANCIAL.
Statement No. I in the appendix gives the amount of money
yet needed to complete the Bridge. Statement No. 2 exhibits
the amount charged to construction by the Auditor up to the first
of September, 1871.
From these two tables it will be seen that the cost of the Bridge
will exceed the original estimate $1,479,582.72.
Justice to myself as well as to you, demands an explanation of
the causes which have led to this difference in cost. For this pur



14            REPORT OF THE CHIEF ENGINEER.
pose Statement No. 3 in the appendix has been prepared. It is
based upon the actual expenditures to September 1, 1871, and the
estimated cost of the work yet to be done. An examination of it
will show the various items of cost of the work over and above
that contained in the original estimate.
In my report of last October (pages 8 and 36) I referred to the
fact of a scour of fifty-five feet in vertical depth having occurred
in the river bottom alongside of the east pier, and of the proofs
that were met with in sinking that pier, that the scour at times
extended even to the rock itself. It was generally assumed by
others who had studied the peculiarities of the river that the scour
never extended deeper than thirty feet below-low water mark.
Although I believed it extended much deeper, and that there
would be no assurance of stability for the channel piers unless
founded on the rock, I believed, when designing the Bridge, that
the abutment on the Illinois shore could be made safe on a pile
foundation. The original estimates, therefore, only contemplated
a foundation of this kind, with the base of the masonry starting
from a level seventy feet above the bed-rock. The possibility of
founding this abutment on the rock having been demonstrated by
sinking the east channel pier, and the proofs of scour of the river
bed to such extraordinary depths being perfectly conclusive, the
Board expressed the unanimous desire that this abutment should
be founded on the rock. By an examination of the appendix it
will be seen that the cost of this extra work was $232,626.64, no
part of which was in the original estimate.
The original design of the Bridge contemplated an upper roadway wholly of wood. The sub-structure of this roadway was
designed to act as a truss to resist the action of hurricanes upon
this part of the Bridge. As now designed, this sub-structure or
wind-truss will be in the form of a flat or horizontal girder fiftyfour feet wide extending from pier to pier, and formed entirely of
plate-iron, thus giving much greater security against wind, and
furnishing a fire-proof defense for the upper roadway against the
sparks and heat from the engines when passing on the railways
below. The wonderful force exerted by the tornado last March,
which was so destructive in its effects upon the construction works of
the Bridge Company, leaves no question as to the propriety of this
change. Nothing equaling the power of this storm is believed to
be on record. A locomotive of over twenty-five tons weight was
actually hurled from the rails by it, and was thrown over on its




REPORT OF THE CHIEF ENGINEER.             15
side twenty-five or thirty feet from the track, on ground only four
or five feet lower than the rails. No evidence of the engine disturbing the rails, or of touching the earth within twelve or fifteen
feet of the track, could be discovered. These facts were reported
to me by my Chief Assistant, Col. Henry Flad, who personally
inspected the spot immediately after the storm.
From calculations based upon the data furnished by this tornado, I feel confident in asserting that the arches of the Bridge
will be capable of resisting an equally powerful tempest. The extra
cost involved in substituting this plate-iron wind-truss for the
wooden one will be $80,000.
The destruction of machinery, scaffolding, etc., caused by the
tornado, with incidental expenses caused by it, cost the Company
fully $50,000.
In my report last October, I stated the fact that the Company
had incurred serious losses, caused by the failure of the first contractor for granite.  I then explained how these losses ensued,
and stated their total amount to be fully $50,000.
From July, 1869, to May, 1870, owing to financial embarrassment, little or nothing was done in the construction of the Bridge.
The official and other expenses throughout this period of inaction,
not calculated for in my original estimate, were not less than
$35,000.
The original estimate contemplated granite ashlar on the piers
and abutments, from two feet below low-water line to high-water
mark. It was afterwards determined to encase the four main
piers with granite to their tops. This extra quantity of granite
cost $73,476.
The items of granite masonry were originally estimated at $32
per cubic yard. The price paid was from $14.94 to $20.76 in
excess of this. After the failure of our first granite contractor,
although the entire granite quarries of the East were canvassed,
we were compelled to pay these higher rates to the subsequent
contractors.  The excess of cost in this item over the original
estimate is $94,149.
The original estimate contemplated an extreme width of fifty
feet for the Bridge. This has been increased to fifty-four feet two
inches. Several reasons made this change desirable, if not absolutely necessary, By the increased width twelve inches of additional space between the arches are obtained for each railway




16             REPORT OF THE CHIEF ENGINEER.
track. This is believed to be an important improvement, and one
which was rendered desirable from the increased width given to
the Pullman palace cars since the original design of the Bridge
was made, in 1867.
On the upper roadway the local traffic will be more conveniently
accommodated by this change. Two lines of horse railway cars
are provided for on it, with room on either side of them for the
largest transfer wagons. The sidewalks will be nine feet wide
instead of eight. In addition to these desirable advantages, the
general appearance of the structure will be considerably improved.
The most important consideration, however, in determining upon
this modification in design, was the increased security afforded by
it against violent winds.
The extra cost of widening the Bridge four feet and two inches
is estimated at $127,652, and the additional land damages at
$30,000.
The cost of tramways for street or horse cars was not included
in the original estimate. This item amounts to $19,113.29.
The original estimate contemplated arches eight feet in depth,
with the railway tracks below the central part of the arches. The
depth of the arches has been increased to twelve feet, and the
railways raised up flush with the lower part of the arches at their
centers. The change in the grade of the approaches caused by
this alteration will cost $33,257.
The original design contemplated only two railway tracks on
the approach between the Levee and Third street in St. Louis.
The side-walls of this part of the approach have been widened, and
an extra tier of brick pillars will be built in it, by which three
tracks will be located there instead of two. The cost in masonry,
iron, etc., by this change will be $10,300. This alteration will
very materially increase the railway facilities of the Bridge.
The cost of widening the iron bridges over the streets and alley
to accommodate this third track will be $6,000.  The switches
required to connect these three railway tracks together in this
approach prevent the use of posts to support the upper roadway
where they occur, and involve the necessity of using iron girders
at an extra cost of $19,800.
The four towers of cut sandstone at the ends of the Bridge, in
which the stairways will be placed, will cost $30,000 over the
original estimate.




RBEPORT OF THE CHIEF ENGITTLER.            17
The grade of the upper roadway trestling in East St. Louis has
been altered from five to four feet in 100. This increases the
length of it 476 feet and the cost $28,000.
In consequence of unforeseen and unavoidable delay on the part
of the contractor in getting the three caissons for the east abutment and channel piers ready early in the season, the sinking of
each of these piers had to be done chiefly in the winter. This
involved the absolute necessity of protecting them from the ice by
putting ice-breakers in the channel above each one of these piers.
This was an unforeseen expense amountingAt  the east pier to........................................................$31,346  53
w   est..........................................................  33,088   33
east abutment.................................................... 11,090  61
Total..............................................................$75,525  47
The cost of sinking the east pier is found to exceed the original
estimates by $70,990.66.
The cost of sinking the west pier exceeds the estimated cost
$35,442.29. The cost of sinking the west abutment, owing to
several steamboat wrecks which were found embedded in the
wharf extension at the site of this pier, exceeded the original
estimates $47,770.80.
The original contract'for superstructure was made with the
Keystone Bridge Company before the detail drawings and specifications for the work were fully completed. The reasons for this
unusual proceeding were deemed of sufficient moment by the
Directory to justify it, but it caused several misunderstandings
between the contracting parties, and resulted in the making of a
supplementary contract in which, among other concessions made
to the contractors, was one in the item of forty.eight wroughtiron skewbacks. The Keystone Bridge Company asserted that
they made the agreement under the misapprehension that these
forgings were to be of cast-iron. Your Company consented to
exclude these members from the contract and to pay actual cost
for them. By this concession the cost of the superstructure is
estimated in Statement No. 2 to be increased fully $30,000.
It will be seen by aggregating these various items that $1,179097.15 of the $1,479,582.72 of excess over the original estimates
are accounted for by them, leaving $300,055.16, which has been
expended, according to the accounts of the Auditor, in extra
2




18             REPORT OP THE CHIEF ENGINEER.
cost of machinery and various other incidental and contingent
expenses, in excess of the original estimates.
I have no additional explanation to submit in accounting for
this difference between the original and the present estimated cost
of the work, except to say that in computing the amount of
contingencies liable to be met with in sinking foundations of such
unusual dimensions, to such unexplored depths, and under conditions so very novel, I evidently under-estimated them, and have
therefore, much to my regret, misled you. While I feel constrained, therefore, to ask your indulgence, I beg to call your
attention to the difficulty of foreseeing or preventing such incidental expenses as some of those which are charged to constrution account and which go to swell up this deficiency-such,
for instance, as "Hospital, medical attendance, and compromising suits, $9,569.21; law-suits in Illinois, $17,087.66; legislation
and similar expenses, $9,800."
Some compensation for the disappointment that must necessarily
result from the delay in completing the structure, and from its
increased cost over the original estimates, may be found in the
fact that the estimated amounts of traffic upon which the revenues
of the Bridge were predicated in 1868, have been ascertained after
a careful examination of actual facts and figures in 1870, by Mr.
Wm. Taussig, Chairman of the Executive Committee, to be greatly
under-estimated. It is confidently believed, from facts developed
by this examination, that the income of the Bridge will be very
largely in excess of the amount estimated in 1868.
With respect to the work remaining to be done, I do not think
there can be a possibility of the actual cost differing materially
from the estimates henceforth. It is all under contract, and the
weights and amounts have been very carefully verified. I feel
quite confident that the delay in the superstructure, which has
retarded the entire work, and added very considerably to the
expenses already incurred, is fast drawing to a close, and that
before many more weeks we will see some of it in the course of
erection.
The character of the work, as far as it is finished, is unsurpassed
in excellence and solidity by anything in this or any other country.
When finally completed, I feel confident that the magnitude of the
structure, the superiority of workmanship, and the quality of
materials used, will fully justify the total cost of its erection,
and that its revenues will amply compensate for the investment.




REPORT OF THE CHIEF ENGINEER.            19
When all of the many difficulties that have retarded this great
work shall have at last been surmounted, and the Bridge becomes
an accomplished fact, it will be found unequalled in the important qualities of strength, durability, capacity and magnitude, by
any similar structure in the world.  Its great usefulness, undoubted safety, and beautiful proportions will constitute it a
national pride, entitling those through whose individual wealth it
has been created to the respect of their fellow-men; while its
imperishable construction will convey to future ages a noble record
of the enterprise and intelligence which mark the present times.
JAMES B. EADS,
Chief Engineer.








APPENDIX.
STATEMENT No. 1.
Estimate of funds needed to complete the Bridge, September 1, 1871.
Trestling in East St. Louis.............................................  $71,880 00
Pier No. 5 and toll-houses in East St. Louis.................  24,060 00
East St. Louis approach over Levee...................................  84,020 46
Channel piers and main abutments...........................  143,909 22
Approach over St. Louis Levee.............................................  64,103 34
Towers on St. Louis Levee....................................................  25,897 00
Approach from    St. Louis Levee to Main street.......................  52,742 64
Bridges over Main and Second streets and alley.....................   18,197 84
Approach between Main and Second streets...........................   38,004 56
Approach between Second and Third streets......................  37,023 30
Ballast and ties in approaches..........................................  11,000 00
Rails on approaches...............................................................  6,500 00
Superstructure....................................................................... 1,528,153  21
Tram-rails on approaches......................................................  11,352 60
Engineering, etc.................................................................  60,000 00
$2,176,844 17
Less proceeds of sales of boats, barges, machinery,
materials, etc., at 40 per cent. off cost................$216,347 00
Railroad  iron.........................................................  11,000  00
— 227,347 00
$1,949,497 17
STATEMENT No. 2.
Expenditures for construction to September 1, 1871, according to the
books of the Auditor.
W est  abutment.......................................................................  $228,355  30
West approaches...................................................................  168,952 19
East abutment..................................................................  354,723 16
East abutment caisson...................................................  139,697 62
East  pier................................................................................  409,175  46
East  pier  caisson....................................................................  111,056  63
Amount carried forward...........................................$1,411,960 36




22                                      APPENDIX.
Amount brought forward..............................$1,411,960 36
West pier..................................................................  309.330 28
West pier caisson....................................................            62,437 22
Tools and machinery..............................................  387,119 81
Boats and barges...................................................................  118,012 68
Materials on hand...................................................................  283,428 65
Repairs.........,.......................................................................  76,467 06
Horses and expenses..............................................................  15,150 00
East approaches...................................................................  47,900 79
Superstructure.....................................................................  14,889 41
Cement account....................................................................  4,957 11
Blacksmithing......................................................................  6,912 53
Wooden buildings..................................................................  30,085 16
Expense account...................................................................  97,633 73
Contingent account................................................................  137,608 15
Engineering account.............................................................  176,080 74
$3,179,973 68
For land damages...................................................................  436,587 31
Total...................................................................$3,616,560 99
Add amount required to complete the Bridge......................... 1,949,497 17
$5,566,058 16
Original estimate, including land damages............................. 4,086,475 44
Excess of cost over original estimate...................................$1,479,582 72
STATEMENT No. 3.
Extraordinary expenses of construction incurred and to be incurred,
not contemplated        in original estimates of May, 1868.
Sinking east abutment pier to the rockMasonry, extra.................................................. $69,796 00
Caisson......................................................  139,697 62
Labor, sinking, concreting, &c...........................  71,127 02
$280,620 64
Less estimate for coffer-dam, pumping, &c., May,
1868........................................................48,000 00
Total extra cost of east abutment.....................                      $232,620 64
Estimated cost of upper roadway and iron windtruss over original estimate.................................                  80,000 00
Loss by the tornado last March..............................                      50,000 00
Loss ensuing from the failure of the first granite
contractors, as stated in October, 1870..............                           50,000 00
Official and incidental expenses incurred from July,
1869, to May, 1870.......................................            35,000 00
Amount carried forward.................................                  $447,620 64




APPENDIX.                                23
Amount brought forward.................................................. $447,620 64
Additional granite, over and above that specified in
the original estimate provided for the channel piers
and abutment, from the upper roadway to highwater mark.....................................................  73,476 00
Excess of cost of granite over the original estimate,            $ 94,149 00
Cost of increased width of Bridge, four feet two
inches.................................................................  127,652   00
Additional amount of land damages for four feet
two inches extra width of Bridge........................       30,000 00
Cost of tram-rails for street cars and vehicles.........          19,113 29
Change in the grade of the approaches, caused by
alteration in depth of the arches.........................       33,257 00
Extra masonry, iron, etc., required for third railway track in west approach..............................       10,300 00
The cost of widening the iron bridges over the
streets and alley to accommodate third track......               6,000 00
Iron girders required by use of the third railway
track  above  named.............................................  19,800  00
Extra cost of the four towers for stairways at the
end  of  the  Bridge..............................................  30,000  00
Alteration in the grade of the upper roadway trestling  in  East St. Louis..........................................  28,000  00
Ice-breaker at the east pier................................. 31,346 53
Ice-breaker at the west pier................................. 33,088 33
Ice-breaker at the east abutment......................... 11,090 61
---        7756,525 47
Cost of sinking the east pierCost of caisson.................$111,056 63
Labor in  sinking, &c..........................................  77,812  42
Guide  piles, &c..................................................  8,796  97
Piles  and  pile-drivers..........................................  10,481  78
Fuel...............................................  5,616  88
Coffer-dam............................................  7,329  34
Materials, supplies, &c...................................... 29,896 64
$250,990 66
Less estimated cost....................................   180,000 00
Excess over original estimate................................     70,990 66
Extra cost of sinking west pier, over original estimate...................................................................  35,442  29
Extra cost of putting in the west abutment foundation, over original estimate.................................  47,770 80
Extra cost of skewbacks.........................................................  30,000 00
$1,179,097 15




STATEMENT of Masonry, in cubic yards, completed up to October 1, 1871, and amount yet to be done.
GRANITE.           GRAFTON LIMESTONE.                 CONCRETE.
______(   a                                                                      C __  M  _m  _  a  ^  6 
NEr w CONTRCACT.
Between 3d and 2nd streets........................... 2,055.0......................... 2,055.0............
Between 2nd and Main streets               52.7.    1,021.1               55.3................ 2,129.1......
Bet. Main street and Corn-.....      1,909.3....3
mercial Alley.............................  11.7...... 1,148.0      589.....              I,..
Commercial Alley & Front st........      72.5...... 1,234.2     ~      22.................    235.2........
W est   Towers............................  44.2............  191.0.........................*  772.5...........
East Approach, Pier No. 5,
and  Tow ers...........................  111.6............  660.9......................................................
Total of New Contract.........  155.8  296.9. 7,310.2......      667.6................. 8,430.5  7,869.5 16,00.0 
OLD CONTRACT.
West Approach, Piers from
5 to 1..439. 1,573.8....................   2,453.8 *...................... 4,466.9  4,343.1  8,810.0
West Abutment.................  540.7  124.6  166.6 145.... 7,371.0......*-  *    1,120.0......f 8,347.9  3,112.1 11,460.0
West Pier......1...... 1026.7      30   130         7,922.5      262.0 219.01,340.0....... 10,710.2  3,360.8 14,071.0
East Pier...........................  985.5..            10,535.8  -   489.0 573     82.0...... 13,923.3  3,793.7 17,717.0
East Abutment...................  260  68.7....     16,395.9  411.6.. 422.0 141.0 669. 17,639.6  5,752.4 23,392.0
East Approach, Piers from 1
to 5.....................................  317  7.................  2,939.3............  2,683.0669.   4,067.0  5,670.0  9,737.0
Total of Old Contract... 2,812.9  950.3 1,770.4 275.. 47,618.3  411  751.0 1,214.0........ 59,154.9 26,032.1 85,187.0
EXTRA WORK BY
The Company........................................................  18.0............
Mr. Andrews.....................                      528.8.28............................. 1 0, 528.8...........
Messrs. Hennessey & Yowel.......................................... 581.2.............. 1.2.....
GRAND TOTAL.............. 2,812.9 1,106.1 2,067.3 2757,839.0 47,618.3 411.0 2,017.8 1,214.0 2,683.0 669. 68,713.4 33,901.6 102,615.0




3
AMERICAN SOCIETY OF CIVIL ENGINEERS.
INCORPORATED 1852.
TRANS SACTIONS.
NOTE.-This Society is not responsible, as a body, for the facts and opinions advanced in any
of its publications.
DISCUSSIONS OF SUBJECTS PRESENTED AT THE EIGHTH ANNUAL CONVENTION. *,
ON LEVEES.-t
fIR. CALEB G. FoRnHEY.-The paper under discussion has much
merit; especially in the history of this phase of hydrologic science.
The " Mississippi levees " have been treated with more or less detail by
all' writers on the history or physics of this great river; Pitman, Stodart, Martin, Flint, Monette, Forshey,. Thomassy, by Humphreys and
Abbot, and now by Bayley. It may be considered as well recorded in,
the pages of hydrographic literature.  Referring to the paper, attention
is first called toEFFECT OF LEVEES UPON THE FLOOD LINE.-The writer undoubtedlyplaces that question beyond cavil, if anything was before needed..
Prior to levees, every bend gave an outlet of 3 or 4 feet over the banks,
for 2 or 3 miles of current.  The sum total of this outlet capacity
was much greater than any crevasses that in recent times inundated thecountry.  "There are many banks on the river front where the natural
surface of the ground has never been overflowed in the memory of man."
This is termed a fallacy by the learned authors of "Pfhysics and Hydraulics of the Mississippi river," chiefly because " there have been crevasses
more or less, every flood." I respectfully dissent, and allege that if the
waters before levees existed were high enough to deposit these banks,
they were higher than at present under the influence of levees.
* Continued from page 298.  f Referring to-CXXI, Levees as a System for Reclaiming
low Lands, G. W. R. Bayley; page 115. $ Delta of the Mississippi, and Physics of the River,
Control of its Floods and Redemption of the Alluvion, a paper read before the American Association for the Advance of Science, 1872. The Levees of the Mississippi River, Transactions,
Vol. III, page 267.




300
The levees were built from below; and the gradual manner of their
application enabled the river's bed to adapt itself to the new servitudes.
We are forced to the conclusion that the river's bed has greatly enlarged.
In addition to the sections measured by me, as stated,* I have since assisted Lieut. Davis in a very careful remeasurement of Sections 5 and 6 of
the prime base of the Delta survey, as measured by me in 1851 for Gen.
Humphreys, chief of that survey.    The remeasurement of these sections
shows a total enlargement of 13 646 square feet of the sections sounded
by the Delta survey above the prime base, and within 2 miles above
Carrolton I have resounded 4 sections, and Mr. W. H. Williams has
re-computed them with care and precision.
They correspond to sections of Delta survey, thus:
SQUARE FEET.
SECTION.            1851.             1872.           DIFFERENCE.
56               226 267           233 193           + 6 926
58               227 458           226 877           -   581
69               209 211           229 325           + 20 084
78               189 128           194 945           - 5 817
These illustrate the allegation of an increase of channel capacity.
The facts, of the shifting banks of the river and the levees preventing the deposit of materials upon the ground formerly visited by the
river, are of some importance to record, while they are remembered;
the localities where this testimony is furnished proving the early floods
as great as more recent ones. I would, therefore, refer to a series of
points not changed, where the water has never been a foot above the
banks since the application of levees. This kind of testimony is now
rare, and the persons who remember the localities are much rarer.
Beginning below New Orleans and ascending, they are:
1. The upper extreme of Point La Hache, 40 miles below. L.t
2. Deer Rlange, above landing, 36 miles below. R.
3. Concord plantation, upper end, 25 miles below. L.
4. Fort St. Leon, 17 miles below. R.
5. Becka plantation, 13 miles below. R.
6. Belleville Foundry, opposite lower portion of the city. R.
7. Friendship, Labarre's place, 2 miles above Carlton. L.
8. Union, Dussieux plantation, at lower end, 13 miles above. R.
9. Red Church, 26 miles above. L.
10. Bonnet Carre Point, Glendale, Ferry landing, R.
* Page 137.  t L for left, and R for right bank.




301
11. College Point, 60 miles above. L.
12. Australia, lower line, 115 miles above. R.
13. Glennons, 160 miles above. R.
14. Bayou Sara, below opposite 168. R.
15. Hore Place, 235 miles above New Orleans, 30 above Red river. R.
16. Ellis Cliffs, be ow opposite, 252 miles above. R.
17. Goodman's plantation, 310 miles above New Orleans, 3 miles above Waterproof. R.
18. Perkins, above mouth of Vidal Bayou, 360 miles above. R.
19. Wilkinson's Point (T 15, R XIV lower side). R.
20. Hendersons, lower part, T 19, R XIV, 412 miles above. R.
21. Pilcher's Point, upper end, Bunch's Bend, 460 miles above. R.
At each of these points, there was in 1872, testimony of adequate kind,
that the highest flood mark was not higher now than when the land behind the levee was deposited.       The levee was small, less than 2 feet
high, in many cases not 1 foot, and the flood mark often less than 6
inches. The water depositing the land must have been 1 foot deep.
TENDENCY OF SOUTHWARD FLOWING WATERS TO IMPINGE AGAINST THE
WEST BANK. —Reclus is well sustained in his remark upon the Mississippi as not'confirming "the law of displacement of running waters."         I
however, dissent from     the doctrine as alleged by him     and affirmed by
Mr. Bayley, although it appears to be sustained by illustration.          The
laws of physics forbid it.   The earth in its rotation revolves as a whole,
water and all; and there is no appreciable tendency of water rather than
of solids, to incline to the West. The rate of velocity, 3 miles per hour
against 900, would be inappreciable in the revolution of the earth.*      The
weight of evidence is on the other side.
The Mississippi river hugs the bluffs of the eastern bank from Cape
Girardeau to Lat. 35~, below Memphis, a distance of 300 miles.          From
thence to Vicksburg, it bisects the alluvial area 380 miles, inclining to
neither side.   From Vicksburg to Baton Rouge, a distance of 250 miles,
it hugs the eastern bluffs again, and thence to the mouths it inclines to
the South East for 240 miles.       In its whole distance of 1200 miles it
touches but once the western bluffs at Helena, when as free as air to
choose its course. And when the mouth is reached, and it divides in
three, and ultimately into about seventeen mouths, it sends one-third to
the South Pass and two-thirds to the other directions, as moved above,
southeastward. And again; the deposit of sediment on the west side,
* I take exception to Reclus' doctrine of the westward tendency of rivers emptying southward; and chiefly because of the disproportion between the velocities. I quote the velocity
of the Mississippi river at 3 miles per hour; this is the rate of channel movement, and not the
movement in latitude, which is much less. The movement southward, of the Mississippi's
water is as one to two compared with the channel movement. The rate compared with the
movement of the earth upon its axis is then as 1.5 miles to 900 miles, and is, therefore, for a
stronger reason than the one assigned, inappreciable.




302
thus fending off the river, is in proof that all rivers with alluvion should
be driven to the eastward rather than toward the western shore.
The Mississippi river in its whole alluvial length is in singular defiance
of this hypothesis. An examination of the chief rivers of the American
continent, where they are larger than elsewhere, will be found to contravene the allegation of any appreciable influence. The Mississippi,
Pearl, Pascagoula, Mobile, Apalachicola, Savannah, Santee, Cape Fear,
Neuse, Roanoke, Pamlico, Chowan, James, Potomac, Delaware, Susquehanna, Hudson, Housatonic, Connecticut, Androscoggin, Penobscot, St. Croix, Sabine, Trinity, Brasos and Corao, with a singular
uniformity contradict the assertion of westward tendency, leaving only,
some small unimportant rivers on the whole North American continent
apparently sustaining the hypothesis.
In South America, the direction of flow of the Orinoco, the Amazon and
all smaller rivers to St. Roque, is coincident with the alleged influence,
and nothing can be inferred from them. Thence all the rivers down to
the La Plata run normal to the ocean beach, and do not sustain the hypothesis.
The rivers of Western Europe running south are all small, except the
Rhone, Guadiana, Guadalquiver and Ebro, which affords a striking contradiction. In Eastern Europe, the Dneiper and the Don, the Volga and
the Ural are all wanting in testimony in its favor.
The only rivers of any magnitude in Asia, that flow southwardly, are
the Euphrates, the Indus and the Ganges, and we look in vain to these
for testimony that their waters have a westward tendency in their flow.
The hypothesis would require that all rivers flowing southward, in
their approach to the mouth, should hug the uplands till they reach the
alluvion and then have a westward flexure in their discharge; whereas
the cumulative evidence is indefinitely great that no such influence is appreciable. I have been thus elaborate, because of the high authority of
Reclus, sustained by Mr. Bayley, and because of the necessity of being
right in the establishment the physics and projecting the works, of a
river, to which the attention of the engineering world is now drawn.
cur-OFFs.-The cut-offs of the Mississippi have been too numerous,
since its navigation by steamboats and occupation by man, to permit the
return to a previous regimen, from natural causes. The assumption that
it will maintain the same length, in any long period, is too violent, in recent times, whatever may have been the truth prior to the settlement of
its hydrographic basin. The number of cut-offs by artificial means has




303
outstripped the tendency to lengthen itself, and the plane of discharge
has been greatly increased in the present century. Every cut-off that
has occurred has been largely assisted by artificial means, No less than
seven have taken place below the mouth of Arkansas, within a half century. Probably not more that two or three of these would have occurred
without aid. They have abridged the length of the river by about 106
miles, and it has lengthened itself less than 20 miles, by increased caving
in the concave bends; all of which have travelled downward. The 618
miles as now measured, should be 724 miles or 17 per cent. more; the velocity of 3 miles per hour has increased to 3.4 miles; and the abrasions
have been in proportion.   The confinement of the water to the channel
by levees, has increased the servitude without sensibly increasing the velocity. They have assisted to increase the capacity of the channel by
widening and deepening, as shown by Mr. Bayley. The cut-offs, however,
have increased the steepness of the plane of discharge, and hence the ravages upon the banks. They are most disastrous in their effects, and should
be prevented by all the means that law and watchfulness can devise.*
OUTLETS OR WASTE WEIRS.-The effects of these are well elaborated
by Mr. Bayley, in his proofs that a bar will form almost immediately
across the stream, directly below the weir or outlet. This is amply shown
in my report upon Bonnet Carre in 1850.t      Soundings in the fall of
1850, showed section below to be contracted 75 613 square feet as compared with section at upper end of crevasse.    Soundings made during
high water of 1851, the crevasse having been closed, proved that the
great bar thrown across the river channel by weakening its transporting
forces, in the discharge of water through the crevasse, was entirely carried
away again, when the river was confined by the new levee. The difference was 23 000 feet in section. +
This paper of Mr. Bayley, is full of suggestion and a valuable contribution to the annals of the Society. 
* I herewith submit a paper on "Cut-Offs on the Mississippi River, their Effects on the
Channel above and below;" wherein the subject is discussed more at length. It will be found
mainly in harmony with the paper under discussion.
t Published in Public Documents, Louisiana Senate, 1851.
$ As I was first to announce this fact in the physics of the river, and it has given rise to
much discussion since, I deem it worth re-claiming, as a great practical principle in the treatment of the river.
~ Although the writer has given frequent references to authorities, one omission is to be
noted. The statistics in the remarks on Tones Bayou are furnished by a survey of Gen.
Jeff. Thompson and myself, one of the most rapid and exposed expeditions, and most fruitful
in the number and value of results, to be found in American engineering; see Report of
Commission of Levee Engineers, January 1st, 1873.




304
MR. GOUVERNEUR K. WARREN.*-My ideas on this subject, in relation to the Mississippi, are set forth in the report of the Commission on
the reclamation of the alluvial lands of the Mississippi,f of which Commission I was appointed President. This Commission was composed of
five members appointed by the President of the United States, under an
Act of Congress, which provided that two of them should be " civil
engineers eminent in their profession," and three of them, officers of
the Corps of Engineers. After a thorough consideration of the subject,
with the most recent data carefully collected and studied, that Commission made an unanimous report. One of the members, Gen. Abbot,
of the Engineer Corps, had shared with Gen. Humphreys, the laborious
investigations in regard to the subject of levees on the Mississippi river,
and also the honor and credit which their contribution, known as the
" Physics and Hydraulics of the Mississippi River," has received.
In transmitting the report of the Commission through the Chief of
Engineers, Gen. Humphreys, to the President, a common sense of justice caused me to say: " The foundation of the report of the Commission rests upon your invaluable surveys and investigations, which, begun
in 1850 and continued until 1861, are published in the great work'The
Physics and Hydraulics of the Mississippi River, and upon the protection of the alluvial region against overflow,' &c., and upon the further
contributions to these subjects contained in your published official
reports in 1866 and 1869." I continued; "the Commission has obtained
additional data upon subsequent floods and the results of the more recent
experience in building and re-building levees, as far as they are attainable, so that their report is in a great measure exhaustive of the subject,
and the conclusions reached may be considered entitled to confidence."
I wish it to be known that this letter of transmittal was written and
signed by myself only, just as the printed report shows, and this expression of confidence was but an individual opinion.   The report,
signed by all the members, gives the foundation for that confidence, and
is expressed in facts and reasons. To the report itself, I therefore invite
attention of those who wish to become informed of my views on the
subject of levees. t
* Presented June 15th, 1876.
t Report of the Commission of Engineers appointed to investigate and report a permanent Plan for the Reclamation of the alluvial Basin of the Mississippi River subject to Inunda-'tion. Washington. 1875.
+ I will endeavor to furnish a copy, to every one who will apply to me for it, giving his
address.




305
To show what these conclusions were, I will here quote them.
10. Cut-offs.-" So far from artificially aiding in their recurrence, it is
therefore the emphatic opinion of this Commission, that in every case
they should be prevented, or at least retarded, if this can be done at
any reasonable cost."
2~. Diversion of Tributaries.-" No such works are practicable except
at enormous expense, and the injury to navigation which would be sure
to result, would in any event forbid their execution."
40. Outlets. —" This Commission is forced unwillingly to the conclusion that no assistance in reclaiming the alluvial region from overflow
can judiciously be anticipated from artificial outlets. They are correct
in theory, but no advantageous sites for their construction exist."
50. Levees.-" There are certain theoretical views concerning the
effects of levee system which are raised again and again in discussing the
subject."  * *  " It is claimed, since the effect of embanking a river
is to confine its sedimentary matter to the channel, that the deposit formerly made on the banks must settle on the bottom, and thus ultimately
raise the bed and with it, the high water mark." * *  "No change of
the kind attributable to levees can be shown to have occurred in any
river, and the theory is therefore without any foundation in fact. Diametrically opposed to this is another theory, which, for the Mississippi,
is equally erroneous,  * * that the increased velocity resulting from
the confinement of its flood volume between levees will rapidly excavate
to a correspondingly greater depth, thus avoiding any permanent increase
in the high water mark. This reasoning, if true, would establish conditions singularly fortunate for the levee system; but unluckily the wish
has been father to the thought. Uncompromising facts show that the premises and conclusions are both erroneous for the lower Mississippi.
Very numerous soundings with leads adapted to bring up samples of the
bottom were made by the Mississippi Delta Survey* throughout the whole
region between Cairo and the Gulf. They show conclusively that the
real bed upon which rests the shifting sand bars and mud banks made by
the present river is always found in a stratum of hard blue clay, quite unlike the present deposits of the river. It is similar to that forming the
bed of the Atchafalaya at its efflux, and is well known to resist the action
of the strong current almost like marble. Clearly then the bed of the
Mississippi cannot yield, and if the velocity be increased sufficiently to
* Humphreys and Abbot.




306
compel an enlargement of the channel, it must be made by an increased
caving of the banks, an effect which is not quite so agreeable to contemplate."
I will here interrupt the quotation to allude to the reference to this
clay bed of the Mississippi, made by Mr. G. W. R. Bayley.*  He says,t
" It is claimed on high authority that the clay bed of the Mississippi
resists the action of the strong current like marble, also that the bed of
the Mississippi cannot yield." In a foot note, Mr. Bayley says, the "high
authority" is " Humphreys and Abbot."   This reference, I think, is a
mistake. I have looked the work of Humphreys and Abbot through
without finding it. It seems probable that Mr. Bayley's quotation is
derived from the report of the Commission, from that part I have just
quoted. But attention is asked to the fact, of his having omitted the
word " almost," so that he makes the quotation read-the clay bed
" resists like marble," whereas it should read-" resists almost like
marble." Even in this last form, it is perhaps too strong an expression
and might better have been omitted altogether. Nothing in the report
of the Commission depends upon it.
To say the clay bed resists the action of strong currents, as clay is
known to do, where the current can act upon the bars and banks, which
are largely composed of sand, is all that is claimed. I will return to this
subject of enlargement of the channel after I have finished quoting from
the report of the Commission, which I will now resume. " In truth, no
marked effect of the kind is to be anticipated, owing to the comparatively
short duration of the increased discharge; for evidently the levees can
produce no effect upon the regimen of the river where the water does not
stand over the natural banks." I will again interrupt the quotations from
the report of the Commission, to state that the data showing the number
of days the river water is against the levees, is given for several floods
from Carrollton to Columbus, in " Physics and Hydraulics of the Mississippi."$ At Carrollton, this period is, on an average, 100 days in a year;
at Donaldsonville, 50 days. The number of days rapidly decreases as
we ascend the river.
I will resume the quotations.  " Hence really the practical effect of
the levees will be limited to raising the high water mark, and to slightly
increase the caving. Since the absolute amount of the increased flood* One of the most eminent engineers of the delta region, in his interesting paper " Levees
as a System of Reclaiming Low Lands," page 115.  t Page 138.  $ Page 411.




307
height does not carry the cost beyond the limits of a remunerative investment, it is the part of wisdom to steadily continue work without indulging in groundless fears that the river bed will rise, or in the equally
groundless hopes that it will be sensibly depressed. * *
"The prolongation of the delta into the Gulf by the aggregation of
sedimentary matter is also assigned as a cause for the ultimate rise of the
bed, and hence a future necessary increase to the height of the levees.
A possible secular change of this nature is quite too remote in its effects
to merit attention from practical men of the present day, Simple calculation will show that hundreds of years will be required to raise the
flood-height at New Orleans, an inch from this cause. In fine, then, we
are to conclude that there is no mysterious agency, either favorable or
injurious, which may be expected to exert a controlling influence upon
the levee system.  X  *  It being certain that the alluvial regions of
the Mississippi can only be reclaimed by levees, it remains to consider
what experience has taught respecting them. The existing system was
begun a century and a half ago near New Orleans, and has gradually
extended upward until there are but few points on the river at which it
has not been tried. * * The faults" (of the system) "are only too
apparent. They are" First.-Vicious levee organization."
"Second.-Insufficient height, in adjusting which, only existing high
water marks have been considered, without remembering that there has
never yet been a great flood in the river in which the water has not been
greatly lowered by immense crevasses which occur with absolute
certainty."
" Third.-Injudicious cross seetions and constructions, which alone
would be sufficient to explain many of the frequent breaks, under the
combined influences of pressure, seepage, burrowing of crawfish, &c."
" Fourth.-Inadequate arrangements for inspecting and guarding."
" Fifth. —Faulty location of the embankments, which are often placed
so near caving banks as to insure an early destruction. Each of the
causes of this failure will be considered in turn."
I will not quote this,* but will invite every one interested in the
subject, to consult the report of the Commission and see how these matters
are there treated.
* As it occupies 11 octavo pages.




308
The report of the Commission ends with recommending a plan for
protecting the alluvial region of the Mississippi from overflow, premising it with the remark that in their judgment "no practical aid can be
derived from any diversion of the tributaries or making artificial reservoirs, that cut-offs are very pernicious, and that artificial outlets,
although correct in theory, find no useful application to the Mississippi.
The plan consists; first, in keeping open the Atchafalaya, and the La
Fourche, and if borings shall show it to be safe, in re-opening the Plaquemine; second, in a general levee system, extending from the head of
the alluvial region to the Gulf, including the valleys of the tributary
streams. The requisite laws to be enacted by the several riparian States
to give the right of way, to confer the authority to make borrow pits
and bench marks, to secure the levee from injury from cattle and hogs
running at large, and to order out in times of danger, under suitable
penalties for non-compliance, the population residing within a reasonable
distance from the levees. The main line of levee to be of sufficient
height (as already computed) to restrain the floods, and of the requisite
cross section to resist the action of the water. Where reasonable security against caving requires large areas of front lands to be thrown out,
protection against ordinary high waters is to be given by low front levees
closely following the bends, suitable sluices and gates in such cases to be
provided in the front and main levees for the rain-water drainage."
This quotation comprises the engineering features of the plan.
The report also gives an outline of the administration of the work, its
division into districts, and the manner of regulating each by itself and
also as a part of the whole.  The report expresses no opinion as to
whether corporations, States or general government could best carry on
the work, leaving that matter to the legislative body that instituted the
Commission.
It should be observed that the report of the Commission presents
a practical plan for the whole alluvial region, which is based on plain
facts, freed from hypothesis. It is applicable to the whole region at
once. Were there money enough, it might all be built in a very short
space of time. It accepts an increase of the flood heights under what
seems to be the most unfavorable conditions, the ascertainment of which
was one of the important contributions of Humphreys and Abbot.
The people in Missouri, Arkansas and Mississippi can go to work
under the proposed plan without waiting for a proper beginning in




309
Louisiana. No plan of protection from overflow, however theoretically
perfect, would be acceptable to the people of the Mississippi valley, which
had to begin at the mouth and be carried upward.
I will return to the question of the enlargement of the natural waterway in consequence of confining to it all the water which now escapes
over the banks. I will waive the question whether the enlargement is to
take place by scouring out the clay bed or by the increased width due to
caving banks, for the practical end would be the same, the enlarged bed
would carry off the increased volume of the water without raising the
flood height as well in one case as the other. But obviously, we cannot
get the increased scour until we build the levees and close the outlets, so
as to confine the escaping flood water. I will take as an example of what
must be done, the case of a levee at Natchez, a midway place of the alluvial region. There the river on March 6th, 1851, was level with the
natural banks, and on March 31st, was 4 feet above them.  This gives a
fair idea of the sudden nature of the rises even at high stages, with the
river imperfectly leveed. Had the river been perfectly leveed, so that the
floods which inundate the whole region 25 to 60 miles wide, were confined to the main channel, the rise would have been more rapid and
higher. This is the view held in the report of the Commission. But
the theory of an enlarging channel says no, the channel would have increased so as to prevent this increase of flood height.  Granting everything to this hypothesis of an enlarging channel that can be claimed for
it, does any one believe that this enlargement could take place throughout the length of the river between Natchez and the mouth, in the short
space of 25 days? Think of, or compute the amount of material that
would thus have to be carried away by the river in a few days and thrust
out into the Gulf, to keep down a rise of a few inches, if adequate levees
were built for the whole region.  It must not be forgotten nor kept out
of sight, that it is the whole alluvial region we are considering.  It does
not seem reasonable that the enlargement could, even under the most
favorable conditions, keep pace with the increasing volume.  Does it
not seem more probable that the levees at first would have to be as high
as if the bed was unchangeable, even though, when once the water was
actually confined to the channel, the flood heights should afterwards,
through many years of erosion, gradually diminish?
Common experience, and all accurate measurements, show that in a
rising stage of a stream, as the volume is increased the surface of the




310
water in the stream rises, and it cannot make any difference where the
increase of volume comes from. The increase of volume in the Mississippi below, by closing the Atchafalaya, is just the same as if the volume
of water before carried off by it came from a new affluent of the same
capacity; or more exactly to indicate what I mean, suppose at a flood
stage in the Mississippi, the Atchafalaya just carried off the water
brought in by Red River, then the closing of the Atchafalaya would
practically add the volume of the Red River to the Mississippi, and cause
it to rise. Increased levee heights would then necessarily be needed on
the Mississippi below it, unless we can suppose the bed of the Mississippi
to accommodate this increased volume.
A belief in such sudden enlargement is not generally held by engineers. Very valuable experience on this point has been gained in the attempts to confine the Red river to one channel, a smaller stream, and
therefore more in our control. Some of the results of this experience on
this river are given in the report of Mr. C. M. Fauntelroy,* where he
states the views he obtained from Mr. W. C. Melvin, C. E., Asst. State
Engineer.  Mr. Melvin says: "As it would hardly be possible to carry
on the work of leveeing both sides (of the Red River) at the same time,
it is obvious where the work should be first entered upon. Time and
assistance must be given to the river to scour its bed to a greater depth,
and for its banks to cave. For this purpose, all the growing timber of
every description that is standing within 60 feet of the crest of the banks
should be cut away; roots and stumps loosened; in short, everything
done to facilitate the caving and scouring process."
It is held by some, this view that more water makes the stream rise,
is an hypothesis; disregarding all the experience of mankind, they
point to the one fact that the closing of Bayou Plaquemine has not increased the flood height at New Orleans, entirely ignoring the well
known fact that there has never been a flood since it was closed that has
not caused large crevasses in the Mississippi between the Plaquemine and
New Orleans, which crevasses have prevented the effect of the closure ever
being felt at New Orleans as an increased flood-height, or as an increase
of scouring power.
There is no doubt that closing the Plaquemine was a benefit to
planters along that bayou, and it was probably at the expense of the
planters on the Mississippi. A similar result will probably attend the
closing of the La Fourche, if it is ever permitted.
* Secretary to the Commission of Engineers, &c.




311
There is a moral point of view to this subject as well as an engineering one, and that engineering is always to be received with caution
which benefits its advocates at the expense of, or injury to others or the
public. Of such is the engineering of the bridges needlessly obstructing
navigation, and the flimsy structures of many kinds by which human
life is sacrificed to avarice.
In closing, I would earnestly urge those who would wish to be well
informed on the subject of levees on the Mississippi to study thoroughly
the "Physics and Hydraulics" of that river by Humphreys and Abbot.
The question is one too vast to be developed in the limits of an ordinary
discussion.
The report of the Commission treats of several points briefly, because they are so fully elaborated and disposed of by Humphreys and
Abbot, whose conclusions have been generally accepted.*
MR. GEORGE W. R. BAYLEY. t-I fully recognize and freely admit the fact that "cut-offs" lower the flood-line of a sedimentary
river, although the temporary     effect is to elevate it below.  The
first effect of the recent Vicksburg cut-off was an increased rise
below, but it soon ceased, and a reduction will follow.    The floodline at Vicksburg and above and below that point, has been reduced
several feet by the Terrapin   Neck and Palmyra Bend cut-offs of
1866 and 1867, and this reduction extends as far down as Natchez.
The lengthening ofthe river channel by the caving in of its banks in the
bends tends to lowr the high water slope, and because the velocity
required for the discharge of the flood waters or maximum quantity requires a greater slope, the surface rises until the increased slope is obtained.
As the length of the river increases, therefore, the flood line rises. On the
other hand, a cut-off shortens the river's length, and the slope, for the
time being, becomes too great.   It can only be reduced to what the
quantity and velocity require by excavating its bed and depressing its
surface. This always happens. As the river lengthens itself, the floodrises; when a cut-off shortens it, the flood-line falls.
The quantity of water flowing, its mean velocity and slopes of
bed and surface, are reciprocally dependent and proportional.     The
resulting effect, however, of excavating the river bed after a cut-off is to
* The "Report of the Commission of Engineers appointed to investigate and report on a
permanent Plan for the Reclamation of the alluvial Basin of the Mississippi River subject to
Inundation," from which, frequent and full quotations have been made, was officially published as part of the Annual Report of the Chief of Engineers for 1875-6, pages 536-678 (Appendix, 0).  t Presented May 25th, 1876.




312
increase the rate of caving in of the river banks and of the higher
cultivable lands near its margins; the lengthening of the river goes on,
then, with renewed energy until another cut-off occurs, and so the floodline rises and falls. If it is financially practicable to protect the river
banks from being undermined and from caving in, great good can be
realized by making cut-offs and shortening the river's length. In the
case of the Rhine, which is comparatively a shallow river, this has been
done. Can it be as successfully applied to the Mississippi river? When
the river, by becoming too long, elevates its flood-line too much, a
cut-off becomes necessary to reduce it, but none are practicable for 200
miles above its mouth.
In the case of Red river, next above and below the city of Shreveport, and down to Loggy Bayou (the outlet of Lake Bisteneau), I am
decidedly of opinion that the plan of cut-offs can be applied to very
great advantage, and that it should be utilized there. By means of it,
the flood-line can be very much lowered, its bordering lands reclaimed,
levees in great part dispensed with and navigation very much improved
there. Three cut-offs can be advantageously made above Shreveport,
and eleven or more others below, above Loggy Bayou.* I recommend
the application of the plan of cut-offs to Red river, above Grand Ecore,
most earnestly, and advise our Society to advocate it.
I have said that the lengthening of a sedimentary river, by reducing
its slope or prolonging it, causes a rise of its flood-line in the effort to
maintain its slope of surface. The advance of the river's mouth into
the sea has the same effect, and I wish to add a few remarks on this
point.
In.Louisiana, there are many old outlets of the Mississippi, and
secondary outlets from these, which were filled up and cut-off from the
river in times long past. I will name a few of them only, for they are
numerous. Below New Orleans, left bank, the Bayou Terre Aux Bcuf
had a course 30 or more miles in length. It is now but a stagnant ditch
or coule, bordered by old sugar plantations for many miles below its
former head, 12 miles below this city. Thence to the Gulf its alluvial
banks become more and more narrow, the swamps and marsh lands
back approaching nearer and nearer to its margins, until only sea marsh
is found. The Metairie and Gentilly bayous, back of New Orleans, have
their delta ridges formed in like manner. The Terre-Bonne and Bayou
* See lithograph map, herewith (on file at Rooms of the Society).




313
Black, right banks, were old outlets of the Bayou Lafourche, or
secondary channels, and the Little and Big CailloV's and Chacahopla
Hiere outlets of the Terre-Bonne and Black. Each and all have their
deltas, which are high, dry, cultivated lands now, and all were formed
by overflow deposits; all were old outlets or "passes."
In each, as its slope was prolonged and so reduced by the advance of
its debouchure, the flood line was gradually raised and the banks
were elevated and extended laterally. When the prolongation of the
main channel became too great for the quantity flowing, and for the
maintenance of the velocity of current necessary for its discharge, then
the mouth widened, shoaled and divided; a division of channels
occurred, and new and steeper slopes, more nearly such as the quantity
in the secondary channels required. The beds and banks of the new
channels were elevated by deposits, and this process backed the water up
in the old main channel, and caused further deposits in it- cLiet,and flattened, by elevating, its high water slope.  Lateral overflows
and outlets then caused a rise of its banks. These still further reduced
the quantity flowing in the main channel, added to the shoaling and the
rise of the flood line. In this way and by this process, the elevation of
the bed and banks went on, further and further up stream, until it
reached the parent river or the primary outlet and nothing more than a
portion of the high flood waters, during inundations/escaped into these
silted-up outlets. After the settlement of Louisiana, even this supply
was cut off by means of levees built across the heads of the old outlets,
and they became stagnant pools or rainfall drains.
The Bayou Lafourche is now, and has for many years, been going
through this very same process, but it has been modified and prolonged
by the maintenance of levees above where the lands are too narrow and
too often overflowed to be worth reclamation. Sixty miles below its
head, where levees 2 feet high were sufficient fifty years ago, levees
of 12 feet high are insufficient now unless crevasses occur through them
in some places, annually, to relieve, for that year, the flood in the
gorged channel. But each relieving crevasse adds to the difficulty year
by year, and the flood line rises in the effort to obtain that slope which
the quantity admitted into the bayou at its head requires. The head
level cannot be raised, for that is the surface level of the Mississippi
river itself.  There is no remedy but the closure of the outlet
next the river by a dyke, and    the substitution of slack water




314
navigation by means of a lock in it; for this contest against nature's
laws cannot be prolonged indefinitely. The length of the outlet is too
great, and its slope too little, for the quantity of- water passing into it at
its head.
I now come to the case of the Mississippi river itself, at its
mouth.    Before the era of levees a large proportion of the flood
waters of the great river were lost through outlets and over its
banks below its last tributary, Red river, and, probably, considerably
less reached the sea through its mouth than now.      Even now less
passes New Orleans, at high flood, than Columbus, Kentucky; the
Atchafalaya carries off the difference, less the quantity lost by
evaporation.
Before the settlement of Louisiana, some centuries at least, the
prolongation of the river into the Gulf below the present forts (21 miles
above the Head of the Passes), had reduced its slope to less than what
the quantity flowing required, and the quantity passing to sea was
further checked by tidal action. The river widened its section from less
than half a mile (opposite the present Forts) to three times that width,
and correspondingly reduced its depth. It then divided, right and left,
into two main channels, with one lesser channel between.      Here,
gradually, the surface level at flood time was elevated so as to give
steeper slopes to the divided channels. Now, the flood line of the river
at the Head of the Passes is nearly 3 feet above the mean level of the
Gulf of Mexico. As the distance to sea, southwest and eastwards, is
nearly 18 miles, and by South Pass, in a south-southeast direction,
12 miles, the flood line slopes are about 2 and 3 inches per mile
respectively, east, southwest and south-southeast.  Between the Head of
the Passes and the Forts, 21 miles, the average slope of the main river
is but 1l- inches per mile (the lower portion still less, not more than 1
inch), and from the Forts to New Orleans about 1I inches per mile.
Now, it is certain that before the division of the waters of the
Misissssippi at the present Head of the Passes, the mouth of the river
was at tide level, and the river surface slope below New Orleans was
greater than it now is.  As it discharged less water then than now
(because depleted by outlets above), its slope needed to be greater. The
increase of quantity and of current velocity, since the leveeing up
of the outlet channels below  Red river and down to the Forts has
enabled the channel to discharge a greater quantity with a less slope than




315
before; therefore, the closure of outlets under the levee system has had a
very beneficial effect and every outlet should be kept closed.
About one-third of the river's volume ig discharged through the
Southwest Pass, but its slope was becoming too flat, by prolongation,
for this quantity; a tendency to widen and divide at its mouth is manifest. Its discharge need to be increased to compensate for its reduced
slope.
About 55 per cent. of the river's discharge hina4-ee passig, out to
the eastward through Pass a l'Outre, Southeast, Northeast and Balize
B:you Passes and their subdivisions, the discharge through which Lt
eic, checked and their division hastened by the prevalent easterly winds
and storms. The repeated division of the waters flowing to the eastward,
and the consequent tendency to increase the slopes of the divided
channels by elevations of bed and surface, ka retarded the flow into
them and favored and increasedflow down Southwest Pass, because of its
having but a single channel, and down the South Pass because of its
steep slope, directness and short length.  The Head of Pass dyke or
jetty, built out into the edge of Pass a l'Outre, up stream from the east
side of the head of South Pass, diminishes the flow down Pass a l'Outre,
and adds to it down South and Southwest Passes, probably more,
down the latter at present than down South Pass, thus increasing their
currents and channel-making power and compensating for extension.
By encroaching, gradually, still further upon Pass a l'Outre, and
diverting still more of its water westward-the navigable channels to the
eastward are useless now-both the Southwest and South Passes may be
improved. The effect of the east dyke is to prolong, further up stream,,
the 2-inch per mile slope of Southwest Pass, but the 3-inch per mile slope
of South Pass cannot be extended, obviously, so far up. Hence the South
Pass slope should and must be extended into or to a junction with
the Southwest Pass slope, below its head, for on that side the fall is
greatest, because the distance (and from deep water to deep water as
well) is least. Flowing water will always seek the shortest route, if it
can, where the fall is greatest. I, therefore, conclude that, at the head
of South Pass, the best channel into it is and will be found on the west
side of the island in the head of the pass, and that the east channel
should be entirely closed by a dyke across to the island. Better water
canble maintained in one channel than in two, and one of them should
be closed for that reason.  Because the shortest distance from deep




316
water to deep water, and therefore the most rapid slope exists on
the west side, that side is the best for developing a channel into the
South Pass. If we work in accordance with natural laws, we cannot go
astray.
Again, South Pass loses 23 per cent. of its water through Grand
Bayou, about half way down from its head, hence the surface slope of
South Pass must be more rapid below than above Grand Bayou. We
have, by perfecting the prolongation of South Pass about 2 miles
seaward by jetties,extended (it was partially extended by side shoals
and reefs previously), and thereby flattened its slope. To compensate
for this, we should increase the quantity of water flowing to sea between
the jetties by closing Grand Bayou, and so provide for the prolongation.
I also think that the width between the outer jetty ends should be less
than above, in order to concentrate the discharge there, and excavate
thereby a deeper channel in the outlet to sea. By reducing the width
we will gain in depth, and maintain the current velocity further out into
the Gulf.
Because the great outlets, the "Jump," 11 miles above the Head of
the Passes, and "Cubitt's Gap," 4 miles above, lessen the quantity of
water in the main river, above the passes, and thereby reduce the
velocity of current and channel-making power (as is evidenced by the
shoaling above the Head of the Passes, from 10 to 12 feet across the
whole river since 1838), these should be closed, and for the same
reason, the great Morganza and Bonnet Carre outlets, between the
mouth of Red river and New Orleans, should also be closed. Every
cubic foot of water which passes the mouth of Red river, or enters the
Mississippi there, should by all means be confined to the river channel
until it reaches the sea through the river mouths proper; none should
be diverted.
The greater the quantity flowing, the greater the velocity, the deeper
the section, the less the frictional resistance to motion, and the less the
surface slope required for the discharge of a given quantity in the same
period of time.




CXXVI.
CUT-OFFS ON THE MISSISSIPPI RIVER.
THEIR EFFECT ON THE CHANNEL ABOVE AND BELOW.
By CALEB G. FOESHEY, C.E., Member of the Society.
PRESENTED JUNE 17TH, 1876.
Of the several problems that have presented themselves for the solution of the practical engineer, in the treatment of the physics of the
Mississippi river, none have been more obscure than the effect of cutoffs. Their application to the great desideratum-the redemption of the
alluvial lands from overflows-has been, and is yet advocated, by those
who believe the straightening of the river would prove beneficial. Fortunately they are few and newly acquainted with these grave questions.
The first cut-off was that known as Shreve's, at the mouth of Red
river. It seems to have been made in the interest of navigation, and
very little importance attached to the work, in respect to its influence
upon the regimen of the river. But from that time forth-1832-there
has been a discussion upon this method of redeeming the lands from
inundation. As one of its advocates, I made a series of measurements and
observations upon the results, to influence the Legislature to effect the
cut-off at Ra-ourci.  The facts are of record and I re-state them here.
Just below Lat. 31~ North, the river made a bend, northward and
westward; at about 8 miles it received Red river, at the fenders of
the bend, and sweeping around southwardly 3l miles, sent off the Atchafalaya; thence it returned nearly to the place of departure, after a
circuit of about 18 miles. The neck was very narrow, not more than 800
feet across. Capt. Shreve, under authority of the United States, by a




318
little trench guided the water through, and in a short time the cut-off
was accomplished. The river established its new channel by the shorter
route.
As soon as the river had time to accommodate itself to the new channel and assume a new and permanent regimen, in 1844, I measured the
water marks, and ascertained the levels of a flood approaching that of
1828, which, it will be remembered, was higher than any other recorded
in river annals. Then the next in height was that of 1844. In the 16
years of interval the river had time to assume the supposed new regimen.
SHREVE'S OR RED RIVER CUT-OFF.-At the mouth of Red river the fall
was 3 feet, by the marks, both of which were very plain upon the trees.
At Vidalia, 65 miles above, the fall was 8 inches, on a locust tree upon the
batture. It was 3 inches at Waterproof, 105 miles above the cut-off.
There was no mark above this point, but I assume that the two floods
reached the same elevation at St. Joseph, 120 miles above Red river.
Then below the cut-off, at Morgan's Point, 28 miles below, the depression was 18 inches, by the testimony of Col. Morgan himself; and at
Bayou Sara, 40 miles, by my own observation, the mark of 1844 was 3
inches below that of 1828. Beyond that point I had not any benchmark or testimony, but presume that it expired at Profit's Island, 60
miles below the cut-off.
RAcoURCI CUT-OFF was made by the State Engineer in 1848-9.  It
abridged the channel distance by 24 miles.  Its location was but 3
miles below Shreve's cut-off.  A canal, some 20 feet in width and 1 mile
long, was made in 1848; but the drift wood floated in it, and the river
failed to go through. In 1849, the woods were cleared away for 100 feet
on each side, the canal was dug deeper to the sand, below the blue clay;
some gunpowder was used in blasting, and ultimately, when the water ran
through the canal freely, a flat-boat was fitted with a wheel that worked
rapidly. This was very effective.  The blasting had loosened the earth,
and it commenced caving very suddenly.  The scene is described as one
of terrific grandeur as the widening reached the lofty forest. The falling
of trees and the whirl and boil of eddies were truly sublime. In two
hours' time, it was a river. The Natchez, Capt. Tom Leathers, with some
degree of recklessness, put her head into the tide, steamed through the
rushing current and t  the   terrible tempest of falling cypresses. After this,
the cut-off was established and the boats generally passed through the
shorter channel, though it was the end of the season, before the great
Mississippi lay snugly in the new bed thus prepared by the hand of man.




319
The observations of the Delta Survey, made in 1851, were commenced
but two years after this event, and before it had effected much in its new
state. Its first effect was to transform the region about the mouth of Red
river into a habitable country.  Its effect in lowering the water was
added to that of Shreve's, making the abridgment of distance about 24
miles,* and added to the shortening effect of the previous cut-off, amounted to 42 miles. The joint effect upon depression of water was equal to
6 + feet.
The effects were felt on the force and direction of the currents upon
the banks of the river. The ravages commenced immediately above and
below the cut-off, and especially in the direction of the points of attack,
where caving had never been experienced before.f The whole community, with one voice, cried out against such tampering with the river.
The policy opposed to cut-offs was established, and since they have not
been seriously advocated, except by some newly arrived engineer or
planter, whose experience was not cognizant of these disasters. The
policy is likely to remain, unless the councils of men newly introduced
to the history and habits of the river, shall prevail to disturb it.
TERiRAPIN NECK AND PALMYRA. —-In 1866, nearly 17 years later, Terrapin Neck, which had threatened for many years, was, by the aid of artificial means, cut-off, and abridged the river's distance by 14 miles. It
occurred at Lat. 320 30' North.  Palmyra cut-off occurred in the following year, at Lat. 32~ 10' North; only 20' latitude intervened between
them. Palmyra abridged the channel by 20 + miles.
It will be necessary to consider these two cut-offs together and as one,
because their effects were merged in each other so that they cannot be
separated. They will, moreover, have to be considered with reference to
a new water mark. The flood of 1828 was no longer legible upon the
trees, in 1851, and no bench marks had been established. The Delta Survey, under Gen. Humphreys, had to establish an arbitrary bench, which
was referred to the mark of 1828, where practicable.
The mark of 1851, which was a high water year, was established by
this survey, for future reference. It was preserved and revived by the
Commissioners of the Levee Company of Louisiana, in 1871, and in 1872
by the U. S. Engineers, for systematic observations.
* I do not lightly differ from Humphreys and Abbott, but this distance was uniformly
called 27 miles. They call it 21 miles. I have revised, and by way of compromise, put the
abridgment of steamboat distance, descending, at 24 miles.
- No analysis of the limits of influence was ever made. It is to be regretted; for the 17
years gave it ample time to distribute its effects over a long distance. Its relief was felt at
Natchez above and at Baton Rouge below, but how much will never be known.




320
In 1874, a great flood came, and made fresh marks all over the entire AMississippi alluvion, from Memphis down; and as these two cut-offs occurred
midway between the two periods, 1851 and 1867, we shall assume them as
reliable, and compare the effects by the two registers. The one is carefully recorded in "Physics and Hydraulics of the Mississippi River,"
and the other, in the report of the U. S. Commission for reporting a plan
for the reclamation of the Delta, as also by the State Engineer. We will
state the several differences as shown in 1874, from the mark in 1851.
The city of Vicksburg is about midway between the cut-offs. The
register at that city will be taken as the standard. The difference made
here by the cut-offs of 34 miles is 72 inches; that is, the water was depressed from 51.3 to 45.3 inches. The difference at Lake Providence,
68 miles above Vicksburg, is 63 inches; that is, the water-mark is
reduced from 45.6 feet on the gauge, to 40.37 feet. At Ashton, 88 miles
above Vicksburg, it is reduced to 2 feet by the marks upon the trees, as
compared with a previous mark, whether of 1851 or 1862 is not certain.
This is the limit of observation from below, and probably comes within the influence of the American Bend cut-off. It would have expired by
proportion, at 132 miles distance above Vicksburg.
At Natchez, 84 miles below Palmyra, the greater of the two cut-offs,
the depression was 1.5 feet.  The slope from 6 feet was gradually observed, though not measured above Natchez.    The depression was from
51.3 feet at Vicksburg to 49.8 feet at Natchez, or 1.5 feet. At Red river
the effect is still greater. It was reduced from 46.4 to 44 feet, or 2.4
feet. At Baton Rouge, again, it is reversed, and instead of being if,
it is dA -OiM  nearly 2 feet. We would place the lower end of the effect
at Bayou Sara, 191 miles below Palmyra.*
Thus the cut-offs have depressed the level of discharge for more than
130 miles above and 191 miles below the point where their influence commenced.   These cut-offs effected, respectively 30 and 42 inches at their
source, making a whole depression of 6 feet; that is, the abridgment
of the river by 34 miles, disturbs the regimen for a distance of 423 miles.
But this is not all that we observe. We shall see the ravages upon the
banks. At the Terrapin neck, the first effect was to cut away the Hawes
Harris plantation, and completely to occupy by the channel of the river,
this fine estate.  Then attacking the point below, it was carried entirely
* The effect at Ashton was only inferred from the overflow being largely reduced, which
effect we put at 2 feet. The crevasse here is now 3 miles wide, and was formerly very deep.
It is now so much overgrown with cottonwood that it vents comparatively little water, much
less than formerly. It has been open since the war, and is filling itself up rapidly.




321
away on the left bank. Then crossing over to the right bank, in the
Milliken's Bend, it caved in the upper portion of that plantation. It did
not stop there, but continued to ravage the two next estates of lower Milliken and Mrs. Maher, and thence continuing, it made havoc of Paw-Paw
island. In fact, it has disturbed every portion of the river down to the
bend above Vicksburg.  In a very short period the effect must be to cut
off that very narrow peninsula, and leave Vicksburg entirely an inland
city.*
The Palmyra cut-off has in like manner made havoc of Point Pleasant
plantation, one of the noblest estates in Louisiana. Thence it crossed
over to the Conger place, carrying away the best fields, then reacting
upon the bend opposite, the caving commenced upon Ships Bayou and
the Alligator Levee, that cost the State so largely, and, far as it was placed
back from the front for safety, it is now at the very bank of the river
and must be renewed. Then its effects upon Hard-Times must be to carry
the river through that bend. The crevasse there this season is the legitimate effect of this cut-off, and all the ruinous consequences upon the
plantations of Lake Bruin and lower Lake St. Joseph. It indirectly
influences the cavings above Rodney, and the disasters of Waterproof are
attributable to it; and the Kempe estate, whose levee cost the State
near $400 000, is legitimately attributable to the same cause.
In general, the effect of every cut-off is to change the points of attack
of the river, and with the full force of the fall, due to the shortening of
the channel, to precipitate it upon the land and levees that have been
reclaimed for many years before. The points of highest improvement are
doomed to new servitudes because in the normal condition, these reaches
of the river have been least exposed.
Cut-offs as a means of reclaiming the lands are, for all these considerations, so injurious, so disastrous, that they should be guarded against
and prevented by legislation and unceasing vigilance.
FLOODS FROM CUT-OFFS.-We should not dismiss this subject without
considering the effects beyond and below the depressing of the floods in
the neighborhood of the cause.
Below and beyond that influence comes the raising of the water, and
the elevation of the flood-line. Though its first effects occur at the cutoff itself, there must be some limit to the depression. At that place,
wherever it be, commences the congestion which produces an elevation
* Since this was written, the cut-off at Vicksburg has taken place-about April 20, 1876.




322
of the flood. This distance cannot be predicted, but every cut-off illustrates it and travels from it, towards the mouth of the river.   In this
instance it was 191 miles from the cut-off. The fact, that at Baton Rouge
the river was higher than before, is not isolated. It continued the congestion so far as observed.  At intermediate places down to Carrolton, in
building and repairing levees, they had uniformly to be raised.
At Carroiton, where it was carefully measured, the rise was 0.7 feet
where the range was 15.4 feet. At New Orleans, it was 0.7 feet, and at English Turn, 15 miles below New Orleans, it measured 6 inches. At PovertyPoint, it was raised about 6 inches.  Judging from the necessity of raising
the levee, and below that point, though I had much experience in the repair and renewal of levees and increasing their height, I could not state
precisely what the elevation was.  But down as low as Tagliaferos, 50
miles below New Orleans, I felt it necessary to raise the levees by a small
fraction.
Thus, for 215 miles, the volume of water was congested in the river
channel, as I verily believe, by the effect of the various cut-offs. This
was doubtless the cumulated influence of all the cut-offs, probably from
the Arkansas down to the Racourci; for if it had been the effect of the
Racourci and the Shreve's cut offs it would have shown itself earlier.
It may be stated, then, that the abridgment in channel of 106 miles
in the past 50 years, by seven cut-offs, has had the effect to congest the
waters by a small but measurable fraction in the channel for the last 260
miles of the leveed river.  Its maximum effect may be stated at 2 feet,
and this effect is at the upper end of the congestion.
It is believed that this will gradually disappear, as the result of accelerated velocity and enlargement of channel capacity from the confinement of levees and the lashing of steamboat waves. In what period this
will occur cannot be predicted.*
* I would take this occasion to retract all the arguments used by me in 1847, to induce the
Legislature to authorize tihe Racourci cut-off. This was done in 1850, but it had not the publicity that I desired, being buried in a public document. For the past 27 years, I have on all
occasions used my voice and pen against the resort to cut-offs to prevent the overflows o1' the
Mississippi river.




Nov. 15, 1873  380  $4 00  $17                                                      250.350  $13 76     2     5
"   29, "     248   "                                                        45    171.490   12 72
Dec. 13,       389         1                                                  54'   263.650   12 72
27, "       78," 90                                                           52.730   13 41
c     Mar. 21, 1874  519   "     2       p   H M;03    394.650   12 52     2~     6
April 4,    1 020   "     44                                               5 I  5  )80  805.830  11 54
18, "           594   495    2                                                L. j  >  40  450.740  1286        5,0             rn
May    2,       574   4 75   2                                                L75    409.480   13 30
16,       529          2 "                                               50    389.380   13 79     "      6
30,      541    "     21                        =                       35    402.980   13 24
574I                                  75V                                                       2
0   June 13,                           D 595 4 86  2  Q C         3               35    435.610  1323             5
Z     ";'!27,"     158                                L             e'^ "         20    119.230   13 34        1  6'    July 25,       167    "                                 r                     oo 00  120.980  13 91     "      5 2
Aug. 8,      481    "     2                                                 3 40   354.220  13 67
22,       274    "     1                     M                           10    204.830   1361
Sept. 5,      567         21. 90    423.880  13 42
y19, "            386         1       I             Q                          L 10    290.710  13 04
Oct. 3,         3531                                                          )70    262.560   13 33'     28, "  |389      <                                        t             2 05   298.730   11 64
e.,"404"1........ 
131  3 75                                                    1 45   405.350   1248      "
6  9720........ $4i                                          430   7210.570  $12985........




104
CVIII.
THE DELTA OF THE MISSISSIPPI,
CONSIDERED IN RELATION TO AN "OPEN RIVER MOUTH."
A Paper by JOHN G. IBARNARD, U. S. Corps Eng., Honorary
Member of the Society.
PRESENTED MARCH 15TH, 1875.
One of the most interesting problems of engineering, as well as one
of the most important works of internal improvement our government
has been called upon to engage in, is that of providing an adequate navigable outlet to the gulf from the Mississippi river-an outlet for the
mightiest system of internal navigation and for the commercial products
of the most extensive and fruitful valley of the earth.
It is well known that the mouths of the several delta arms, or "passes,"
of the Mississippi are obstructed by bars over which 15 feet of water is
rather more than an average maximum, even for the greatest of thesethe "Southwest."
A brief retrospect of the history of the question of augmenting this
depth will be quoted.*
"The necessity and feasibility of deepening one or more of the
passes of the Mississippi is not a new subject. It is one almost coeval
with the settlement of the country itself. Yet it is only of late years that
projects to accomplish this object have been seriously entertained."
" By reference to ancient charts, it would appear that the North East
Pass for a period of at least 70 years, maintained a depth of 12 feet on
its bar. This depth was found more and more inadequate as commerce
rapidly increased after the transfer of the country to the United States,
and about the year 1835 public attention was strongly attracted to the
necessity of increasing the depth in this and other passes."
" With a view to this object a preliminary survey was made under the
direction of Maj. W. H. Chase, Corps of Engineers, who based thereon
certain projects and estimates for increasing the depth of water by closing several of the Passes and dredging the channel through the North
East and South West bars."
* From a document (dated 1852) to be named hereafter.




105
"Congress appropriated for these objects the sum of $250 000; but the
subject was deemed so important by the War Department that it was
referred to a special board of Engineers, who, simultaneously with the
experiment of dredging, ordered a new and very thorough survey of the
delta, with the view of obtaining more extensive and accurate data on
which to base a project, and also to furnish a standard to which engineers
could in future refer in investigating the changes which are continually
taking place in the channels and at the mouths of the passes."
"The survey" (since known as Capt. Talcott's) "was executed with
all desirable precision and scientific skill; a powerful dredging boat and
tenders were procured and the experiment of dredging commenced.
JUnfortunately, the survey and building of dredging machinery nearly
exhausted the sum appropriated, and Congress having failed to make further appropriations, the experiment was necessarily abandoned before it
had progressed sufficiently to test its efficacy. Thus the operations of
this period failed to cast any light upon the important question now before the Board, viz.:' what is the proper method of securing a depth of
water over the bars adequate to the wants of commerce.' "
" As the North East Pass became more and more innavigable, it was
found that the South West, which had heretofore been little used, answered sufficiently well the existing wants, and it has continued to answer
them, without material inconvenience, till a recent date."
"During the past year, however, public attention has again been attracted to this subject. Vessels carrying large and valuable cargoes have
been detained for weeks and even months on the bar, and it has been
stated that the South West Pass has, in its turn, commenced shoaling
while the Pass d Loutre has commenced deepening."
" The Board does not find evidence to confirm the opinion that any
deterioration has taken place at the South West bar. * * *
" But the Board, though having little fear of a sensible or rapid deterioration of this pass, is nevertheless of opinion that it is not adequate to the existing and prospective wants of commerce, and it is this
inadequacy and not its deterioration which has now made it necessary to
look for some efficient means of deepening this or other passes."
The facts recorded in the above quotation had caused an appropriation to be made by Congress in the year 1852, in words as follows: "for
opening a ship channel of sufficient capacity to accommodate the wants
of commerce, through the most convenient pass leading from the Mississippi river into the Gulf of Mexico, seventy-five thousand dollars.
And it shall be the duty of the Secretary of War to apply said moneys to
the opening of said channel by contract, and at an early day in the next
session of Congress to report the progress of the work, the amount necessary to complete it and an estimate of the annual cost of keeping said
channel open, and any contract made shall be limited to the amount
hereby appropriated."




106
To carry into effect the above, a board was organized consisting of
one naval officer, the late Corn. W. K. Latimer, and three engineer officers, viz., Maj. W. H. Chase, (since deceased,) Maj. J. G. Barnard, and
Maj. G. T. Beauregard.   Corn. Latimer, (at that time in command of
the Navy Yard, Pensacola,) had been long on duty in the gulf waters,
and was familiar with the Mississippi. Maj. Chase had served since 1820
in this region, commencing at that date the very first work, (Fort Pike,)
of the modern, or third system of sea-coast defenses, which was located
in the marshes of Louisiana.
It is scarcely necessary to say, that to engineers of Louisiana, the
Mississippi river-its floods, its mouths, its levees, its crevasses, its encroachments, &c., are an ever-present theme, and Maj. Chase, as we have
seen, not only directed the first government surveys, but was the author
of " projects and estimates" for increasing the depth of water on the bars.
He was, more notably, the author of the first actual plan* for a " ship
canal," though not the first to propose one.
Maj. Beauregard and the writer (the former a native of Louisiana)
had both been stationed in New Orleans, in charge of fortifications of the
river and other navigable approaches. Both had had extensive experience
in construction on the peculiar soil of Louisiana and of excavation therein; indeed their experience had in part been, (at Forts Jackson and St.
Philip,) in the very locality where the proposed ship canal was designed
to be.
The officers of 1852 were selected, therefore, for their supposed familiarity with the peculiar problems of hydraulic engineering which characterize the Mississippi. In entering upon the duty assigned them, they
did not find themselves confronted with a novel-even though it were a
difficult-,problem. They spent several weeks on the spot in execution
of their task, eight days of which were devoted to personal examinations
of the bars, occupying themselves mainly with "studying the changes
which had taken place between the survey of Talcott and the more recent
one of Capt. Sands; the nature of the formations exhibited; the character of the bottom, particularly of the South West Pass and Pass a Loutre,
and the general slope of the bottom of the gulf seaward for several miles
from the Passes."
It would be foreign to the scope of this paper to follow further the
report of the Board of 1852;t it enumerates as the eligible methods of
* This plan which is elsewhere alluded to by the writer as " boldly, ably yet so imperfectly sketched out," is certainly superior to any subsequently made, till the Board of 1874
presented what may be called a thorough solution of the canal problem.
t Drawn up by the present writer, and from which a quotation has already been made.




107
proceeding: " 1st. Stirring up the mud at the bottom by suitable machinery, throwing it into the current whereby it is to be swept off. 2d.
Dredging. 3d. Jetties projected from the shores to contract the current
over the bar. 4th. Closing the useless passes."
But the members of the Board were limited by their instructions to
devise, if possible, a work which could be executed-without reference
to other appropriations-for the small sum of $75 000.
No other plan suggested itself to th'e Board by which this sum could
be applied with hope of obtaining any important result than that of " stirring up " the bottom, and upon its recommendation a contract was entered
into with the Tow-boat Association, by which a channel through the bar
at South West Pass, 18 feet deep and 300 feet wide, was to be made. The
execution of this contract was the very first successful application of any
artificial means to deepening the channels over the bars, and it demonstrated the efficiency of dredging by that method.
But the Board could not of course be certain of success by this means.
It further recommended: " in case the Tow-boat Association fails to
perform the contract, and it should not be thought expedient to resort
to dredging, the above sum, together with the available existing appropriation, should be applied to the construction of jetties; as the Secretary
of War may direct."
" The project of jetties is based upon the simple fact that by confining
the water which now escapes uselessly in lateral directions, to a narrow
channel over the bar, the depth of this narrow channel must be increased;
in other words, the existing bar must be cut away." * * *
"' The rate of annual expenditure for extension of these jetties the Board
is unable to estimate. The Board recommends the trial of the project,
in case all efforts of stirring up the bottom, or dredging, fail."
The Board of 1852 is, therefore, not only the inaugurator of the
first successful application of artificial means to deepening the channels
over the bars; but it was the first to give official sanction and recommendation to the project of jetties for the deepening of the outlets of one cr
more of the passes. But though this may be the first formal expression,
it was even then by no means a new or unfamiliar project. One of the
members, (Maj. Beauregard,) had some years before, somewhat warmly
espoused their trial.
After the appropriation of 1852 (by which a channel 300 feet wide and
18 feet deep was opened by the Tow-boat Association) no further appropriation was made till 1856, at which time, of course, that channel had become obliterated.  In that year an appropriation of $330 000 was made
" for opening and keeping open ship channels of sufficient capacity to




108
accommodate the wants of commerce through the South West Pass, Pass
a Loutre," &c., &c. It being further provided that the work should be
done by contract, proposals were advertised for, which were submitted
to a board consisting of the late Maj's Chase and Bowman and the (now)
Gen's Humphreys and Wright.
This Board recommended the acceptance of the bid of the New
Orleans Tow-boat Association for deepening a channel 300 feet wide
through the South West Pass to 18 feet depth for $125 000, and for
keeping the same open at the rate of $75 000 per annum; that the bid of
Messrs. Craig and Rightor be accepted for opening the Pass a Loutre channel for $125 000, and for keeping the same openfor five years for $40 000
(i. e. $8 000 per annum). With respect to this bid the Board remarks:
" The Board have great doubts of the practicability and efficiency of
this mode-of the practicability of construction and of efficiency, should
the work be accomplished. But an important point will have been ascertained either by its failure or success. If failure is exhibited, future projects on this score would at once be rejected.  If success is exhibited
two reliable modes for future improvement will have been obtained; the
one by dredging, scraping and raking, and the other by closing passes
and increasing current, or by jetties; and one or the other, or both,
maintained as the value of their several results may be determined."
The Secretary of War (Mr. Davis) decided that the bids of Craig and
Rightor should be accepted for both of the passes, and a contract was accordingly made, (November 13th, 1856,) in which "the said Craig and
Rightor contract to open,through the entire length of the bar or shoal at
or near that outlet of the mouth of the Mississippi river into the Gulf of
Mexico, known as the Pass a Loutre, a straight ship channel, having a
well defined width throughout its whole extent of not less than 300
feet, and a depth of not less than 20 feet below the level of ordinary low
water." * * *
" It is understood that the means by which the said Craig and Rightor
propose to accomplish the work are by stopping the minor passes or outlets of the river, and, if necessary, by constructing jetties on or near
said bar, adopting one or either or both of these alternatives, as may
be found expedient, substantially after the manner explained and set
forth in their proposals submitted to the Engineer Department."
Inasmuch as nothing was actually done by them in the way of " stopping the minor passes or outlets," and as the work of constructing a
jetty by a patent sheet-pile construction (an invention of Craig and
Rightor) was, at first, exclusively resorted to, it is to be inferred that the
jetty system was the essential part of Messrs. Craig and Rightor's undertaking. The late Col. S. H. Long,.U. S. Engineers, thus reports thereon.
April 6th, 1857:




109
"The operations of Messrs. Craig and Rightor have hitherto been
confined to the Southwest Pass. I visited their work on the 27th ult.,
and had the satisfaction to find that they had formed a line of piles, in a
direction obliquely downward towards the deepest channel across the
tidal bar, 1 064 feet long, the depth of water along the line varying from
4 to 8 feet. The line consists of a series of posts or piles, about one foot
square, driven vertically into the bottom to the depth of about 25 feet
below its surface, and at the distance of 15 feet apart. Longitudinal
strings or streamers, 4 by 8 inches and 30 feet long, are spiked to the
upper sides of these posts a little above the surface of the water. The
strings serve as guides to the plank piling, which is composed of planks
5 inches thick at top and 2~ inches thick at bottom, rebated in the manner specified in the patent."
" The plank piles vary in width from 12 to 20 inches, are driven 10 to
15 feet into the ground, and fastened by iron spikes to the upper sides of
the streamers. The spikes employed for the fastenings are about half
an inch square and 9 inches long, and the timber of the structure consists of yellow pine and cypress. The line thus formed presents a sub:
stantial and well built structure, is able to withstand a current of about two
miles per hour, and is quite impervious to water. On the upper side of
the line, the depth of the water remains as it was before the piles were
driven, while on the lower side, sedimentary deposits have been made to
the depth of 3 or 4 inches, the water on this side having become nearly
or quite stagnant."
" Since the date of the inspection, as above, I have been informed
that the line has been prolonged more than 500 feet,' making the entire
length of the line about 1 600 feet. The daily progress in extending the
line, the requisite materials being at hand, and the weather favorable, is
about 150 feet."
On December 31st, 1857, Col. Long reported that the "pile-dam"
had a length of 5 733 feet.
" At the upper or northerly extremity of the dam the water does not
exceed 3 feet in depth, and as we descend along the line of the dam
the depth gradually increases to about 12 feet, which is the depth at the
lower end of the dam."
"Through a distance of about 1 500 feet, on the upper portion of the
dam, indurated beds of sand and clay occurred at the depth of 6 to
8 feet below the water surface. On being penetrated by the piles, these
beds thus disturbed resolve themselves into a sort of quicksand, too
yielding to afford the requisite stability to the piles. In consequence,
and on the occurrence of protracted boisterous weather, soon after this
portion of the dam was built, the line of piles was ruptured in many
places, and in some instances thorough breaches of greater or less extent
were found in it."
"The residue of the dam being strongly fortified by substantial
square piles driven on both sides of the plank piling, and at intervals of
only A or 5 feet asunder, successfully withstood the buffetings of the




110
storms and the surf produced thereby, and still remains erect and
stable." * * *
" The materials used in the formation of the dam, exclusive of a large
quantity driven away and lost in storms, amount to about 2 000 000
superficial feet of yellow pine and cypress timber, and 5 tons of iron in
bolts and spikes." (The accompanying sketch shows the extent and position of this pile-dam.)
A report, dated December 10th, 1857, to Col. Long of Mr. Wm.
Johnson, supervisor (under the former) of the work, gives some interesting particulars.
" The contractors have been much troubled in driving any permanent work at the north or upper end of the dam. For about 1 500 feet
from the marsh, the bottom is composed of yellow sand indurated with
clay, and in some spots, of sand alone. The driving of the piles in these
places is a matter of great difficulty, and when driven they have a tendency to work up and out of place, and upon the smallest aperture or
opening being developed, the current immediately commences to wash
away the surrounding sand, and in an incredibly short space of time it
will excavate below the bottom of the piles in the immediate vicinity of
the old cranny, and the work thus affected will, during the first gale
thereafter, wash up and be drifted out to sea."
" In this section of the work I have known the current at one of these
apertures to excavate from 2 to 16 feet in 48 hours."
"Several breaches of this character have occurred during the last six
months-have been as often repaired, and in spite of every precaution still
continue to break out anew. With the exception of this division of the
line, the balance of the dam, over 4 000 feet (the whole length of the dam
being 5 733 feet) is driven in a blue clay bottom, with a depth of from 7
to 15 feet, below which an indurated bed of white sand is found, with
an average thickness of some 10 feet, below which again lies a stratum of
yellow clay to an indefinite depth. The work driven in this bottom is now
standing in perfect order; has remained so since the date of its construction; is unshaken, and, in my opinion, is unshakable, except by the
power of some irresistible tornado, or the more insidious effect of the
Gulf worm. I am happy to inform you, however, that I have in vain
sought for some trace of this animal in wood planted in the water last
February, and which has been submerged ever since up to this date."
A year later (December 22d, 1858), Mr. C. A. Fuller, who had succeeded Mr. Johnson as supervisor, reports:
" Since that date (May 28th). a portion of the pile-dam near its upper
end has been swept away, leaving a clear and continuous breach of about
900 feet in length. The remaining portion of the dam, with the exception of several breaches which are not of recent formation, appeared to
stan I well. In examining this dam in a light skiff I was obliged to keep




Ice' Iii,,. g                              t....
I~~~~~~~~~~~~~~~~I
~ r           I ~:~,   ~!14~..
0.~~~~~~~~~~~~~~~~~~~~~~~~~~~~o
t          0..-**,.      l......,'~~ ~ ~''l  a''lq~i~tT~ l~1~111 t1..'III1.! i,                                0
~'`1            I j         I    
i~~~~~~~~~~~* J,;-............ a~
~~ 9...          ~,,
~~~~~~~~ U. i *'  il''           ~  "....  ~'  -
Ii
~~ ~ ~                 I.       0~  ~........
t   a'~~~""-'w'a. "  (I''',   ". 4.....
~*'i 1,e          -.'a ~x~ 
~~~~~~~~~~~~~~~~~~~~- ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ -...,.'  ~  -. U:".
P;  5    Ii               j;      ~.....~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~t,3~...
g~~~~~~~~~~~~~~~~~~~~'"
c o.
03 C~0*
ao   3                         ~~J...~ —.~.II
-S~~~~~~~~~~~~~~~~~~~~~~~~~~~~+








11l
on its channel side, there being but a few inches of water on its easterly
side, the sand being compact and hard against the dam."
It should be remembered that a year previous to this last date, i. e.,
in December, 1857, when the pile-dam had attained the length of 5 733
feet, the contractors found their resources exhausted (they had made,
according to their own statement, an actual expenditure of $66 079.27)
and were constrained to apply to the Government for relief. It was
finally granted, though not exactly in the form they asked. They obtained an extension of time for completion of their contract to July 1st,
1858, and, what was more important, a relaxation in regard to the depth
prescribed, by which 18 feet depth was substituted instead of the 20 feet
originally contracted for.
The contractors then abandoned their jetty construction and resorted
to "harrowing," "scraping" and "'blasting with gunpowder;" the
latter agent applied principally to the " mud-lumps" at Pass A Loutre.
On May 29th, 1858, Col. Long reported, after personal inspection and
soundings, that a channel 300 feet wide with depth from 18 to 22 feet,
existed at the South West Pass; and on September 23d he reported, from
his own personal examinations, a channel opened at Pass a Loutre with
an average depth considerably more than 18 feet, and, at its narrowest
part, 325 feet wide;* and hence, in accordance with the stipulations of
the modified contract in so far as they relate to the " opening of these
channels," the compensation to be made for the opening thereof ($125000 for each) " is now payable to the contractors." It was accordingly
paid.
The foregoing particularst are given in relation to these contracts
with Craig and Rightor, because, involving nominally the use of jetties,
it hasbeen inferred on the one hand that the jetty system failed; on the
other that the attempted jetties were of such frail construction as to be
* The engineer officer now in charge of the dredging of the delta mouths having in his
recent annual report (1874) stated (see Ex. Doc. 1, 43d Cong., 2d Session) that he "has satisfied
himself that the depths and widths of channels reported in 1853, 1857, 1858, 1859, 1860, have
been erroneously reported," it is proper to remark that the only cases in which specific widths as.
well as depths were reported were those of 1853 and 1858, when, under contracts for specified
widths and depths, large sums of money were to be paid and, on the certificate of the U. S. engineer officer, were paid. These cases were as follows: in 1853, $75 000; in 1858, $125 000 for
(severally) the S. W. Pass and Pass a Loutre channels, or $250 000 in all. The official records
of Col. Long's inspections and reports have been published in Ex. Doc., H. R. No. 5, 36th
Cong., 2d Session. Gen. Beauregard, who was the engineer officer in charge of the execution of the first contract (that of 1853) has seen proper to publish an answer to so much of the
so-called "expose" as concerned himself.
t At the Pass a Loutre, about 1 000 feet of "pile-dams" were made by Craig and Rightor,
before they resorted to other methods.




112
quite incapable of endurance. Slight as the construction was, it has
been shown by what precedes, that most of it stood firmly during a whole
year; but what is more noteworthy, and what till recently was not generally known is, that though now very much worm-eaten it has endured
to the present time.
To have carried out the jetty system in its application to this pass (the
South West) required at lowest computation that this commenced " piledam " should have been extended beyond the bar crest (i. e., about 3
miles further), and that another, starting from the opposite shore, should
converge towards this one.
It may very well be doubted whether this " pile-dam " construction
would have stood in depth very much greater (12 feet) than that where the
incomplete work terminated. It has demonstrated, nevertheless, that
the shoals extending from the extreme points of unsubmerged land towards the bar channel have adequate compactness and tenacity to support
well adapted engineering works.
A few words must finish this retrospect. If the contractors (Craig
and Rightor) abandoned the jetty system, the cause is obvious. Their
own funds had given out and the amount of compensation ($125 000)
they were ultimately to receive was utterly inadequate for a proper application of that system. But the terms of their contracts required only
that, at a certain date, they should exhibit a certain depth (and width) of
channel. To exhibit such a result they then resorted to the temporary
expedients of scraping &c. the bar surface, and by so doing arrived at
the result desiderated.
But their contracts further provided that they should maintain the
prescribed depths for a period of five years for $40 000, i. e., $8,000 per
annum. Had they effected the increased depth by jetties, as they originally intended, they doubtless believed that the maintenance for five years
would not be costly. But it was clear enough that a dredged channel
could only be maintained by incessant dredging, for which the Tow-boat
Association, which had already had experience in this matter, demanded
in their competing bids, $75 000 per annum. Craig and Rightor, after
making a futile petition that the compensation be increased to the modest
sum of $17 000 per annum, abandoned the work, and the channels which
they had made gradually filled up.
There still remained an unexpended balance of the appropriation of
1856, of over $60 000. An abortive contract for reopening the South
West Pass channel was, in 1859, entered into and canceled, and then




113
Col. Long, hiring by contract a suitable steamboat, with scraping apparatus, undertook the operation himself; Mr. C. A. Fuller, (a graduate of
the Military Academy, for many years an assistant engineer to Col.
Long) being in immediate charge. A channel 18 feet deep was, through
parts of the years 1859 and 1860, maintained; the width is not stated,
but according to Mr. Fuller, it never was so great as 300 feet.  The civil
war interrupted operations for the time.
Subsequently, the government resumed the work, and a boat (the
"Essayons") for dredging, or rather for'" stirring up" the bottom, designed by the late Major McAlester, U. S. Engineers, was constructed at
Boston under his supervision. The stirring instrument consisted of propeller blades extending below the keel.    This machinery, improved by
Maj. Howell, by the addition of the "deflector," which more effectually directs the stirred-up material into the upper currents, has since
been in operation.
It is stated by Maj. Howell, in an official letter to the President of
the first ship-canal Board (of 1873), that with full control over the use of
the channel and two dredge-boats, a channel may be made and maintained
with 20 feet depth, at a cost of $150 000 per annum   running expenses,
and $50 000 per annum for wear and tear; and this is reiterated in his
annual report for 1873.   His report for 1874 does not sustain these assertions. *
A wide " open mouth "-not a mere sub-aqueous "ditch" of uncertain depth and an irremediable narrowness of less than " 300 feet
width"-is evidently the kind of navigable outlet the great river needs.
To furnish an outlet which should evade the bar obstructions of the
mouths, a "ship-canal" had long before been proposed; first in 1832,
by Mr. Benj. Buisson, State Surveyor; and again, in 1838, by Maj. W.
H. Chase, U. S. Engineers; and the project was referred to by the Board
of 1852, as one to " fall back on" in case of failure of all efforts to
procure an open mouth; thus placing it in its rightful category of a
dernier resort.  This project was revived by the following resolution of
Congress, passed March 14th, 1871.
* Maj. Howell, in his last annual report, states that his former reports " do not show the
maintenance throughout any one year of a channel 18 feet deep and 300 feet wide." They certainly affirm that "natural causes effecting a blockade of the mouth of the Mississippi have
been overcome by the system of dredging adopted, so far as regards obtaining a 20-feet
channel across the bar of the South West Pass, is evidenced by my reports." "300 feet width"
may be a desirable width; but when a " 20-feet channel across the bar of the South West Pass"
is spoken of, a channel of sufficient width for the purposes of navigation is necessarily supposed, whether 300 feet wide or not.




114
"Resolved: that the Secretary of War be, and is hereby requested to
cause an examination and survey, with plans and estimates of cost, to be
made by an officer of Engineers, for a ship-canal to connect the Mississippi River with the Gulf of Mexico, or the navigable waters thereof, of
suitable location and dimensions for military, naval and commercial
purposes, and that he report upon the feasibility of the same to the
House of Representatives."
In the summer of 1873, a Board of Engineers, of which the writer was
President, was convened " to consider and report upon" a plan which
had been in the mean time prepared in fulfillment of the above. The
question submitted to the board was widened by the request of the Chief
of Engineers, at the suggestion of the President, to consider " the expediency of improving the navigable outlet of the Mississippi, by the
Fort Saint Philip Canal, as an alternative to, or a simultaneous measure,
perhaps, with, the improvement of the passes."
The majority of that Board favored the construction of the canal and
embodied their views as to jetties in the following paragraph.
" Upon a review of the practical difficulties which the adoption of
the jetty system of improvement at the mouth of the Mississippi would
entail, and a due consideration of the original cost of construction and
of annual extension, entertaining doubts, moreover, of the succcessful
issue of the attempt, the Board do not consider it advisable to recommend it."
The single dissenting member took the ground " that before resorting to an artificial work of the difficult and costly character of a shipcanal, a more attentive consideration of the superior advantages of the
natural mouths, and of the fair probabilities of utilizing them, is
needed."
And again: "the advantages of an open river mouth are inestimable.
The needs of a navigation so great as that which now exists, and which
in the future of the great Mississippi valley must be fifty-fold increased,
demand it."
" It is said that the'time has come' when the needs of commerce
demand the canal; but I answer that the time will come when there will
be the same cry for navigation unimpeded by locks-an open river
mouth-which we now hear for a canal."
The argument, as put in this last quotation, contains the very pith of
the question. No seaport in the world would substitute a ship-canal,
with locks, for an open sea entrance, if the latter be not unattainable.
Would New York, for example, accept as adequate a ship-canal,
for her sole channel of access to the sea? And what limit shall be
placed upon the magnitude of the freight-commerce which shall pour out
of the great Mississippi valley through its only outlet to the sea?




115
A ship-canal can be but a make-shift, a pis aller, only to be accepted
on proof that there is no reasonable hope of an "open mouth."    How
shall that reasonable hope be established save by reference to what
has been done elsewhere, and by an appeal, not to theories-for theories
simply mock the subject by their impotence to grasp the complicated
conditions-but to such simple elementary facts as may guide our
estimate of probabilities? Here all the burden of proof is thrown on
the opposer.
To say, in face of the highest authorities* on this subject, in face of
achieved success in numerous instances, that there are no probabilities,
or that there are none which justify an attempt which has for its end a
result so indispensable to a great navigation, requires not merely transcendent engineering abilities-it requires something like prescience.
The question is not, therefore, purely an engineering one. To fail
(if the contingency of failure be admitted) is far from being an engineering failure; for the undertaking is not merely justified, but demanded;
unless, indeed, the engineer can deny even the possibility of success.
While no engineer, it is presumed, would trust his prescience so far
as to deny a possibility of obtaining by jetties an "open river mouth,"
there are a few simple arguments on which to assert-first, the strong
probability of success; second, that the maintenance of the accomplished
open mouth need not be regarded as involving an unreasonable expenditure or an excessive amount of work.
As to the first point; no one, whether "engineer" or otherwise, has
yet denied the certainty of obtaining by jetties the desired result.
Hence the second point only, requires to be dwelt upon. In the
draft of a minority report, dated New Orleans, December 6th, 1873, this
matter is thus set forth:
" Now as to the application of this work. For reasons which will
appear hereafter, I should select Pass a Loutre for experiment. A glance
at the chart will show that from the point in the Pass where the depth of
25 feet ceases to obtain, to the otter crest of the bar, is about 21 miles (at
the South West Pass over 4 miles). The natural width of the Pass where
25 feet depth obtains is about 4- mile. * * *   By stopping the North
* In Europe, e. g., we have Sir Charles Hartley, engineer of the successful jetty construction of the Sulina; Col. James Stokes, Royal Engineers, British Commissioner for same
work; and Mr. P. Caland, engineer of the works at the mouth of the Maas; all advocating an
" open mouth" for the Mississippi; nor can the opinions of these eminent engineers be
impugned by the allegation that, being foreigners, they are unacquainted with the physical
peculiarities of the Mississippi. They have all studied it attentively with access to the most
authoritative information; including the work, the "Physics and Hydraulics of the Mississippi."




I16
Pass, and extending parallel, or nearly so,* jetties which, starting at the
cessation of 25 feet depth (above the North Pass), extend, a half mile
apart, 4 miles to points opposite the outer crest of the bar, the bar must
be excavated to 25 feet (i. e., the 25 feet channel will be extended at once
2~ miles); the velocity of current maintained unimpaired up to this point
will carry its sediment far beyond, into deep water. The present regime
of a shoal bar cannot again be restored, until the vast bottom area now
covered with deep water beyond, for a distance. of 21- miles, is raised.
That there will be a 25-feet, or even a 20-feet available channel all this
long time I do not contend. But this I do contend, viz.: * * * that
once extended out to the crest of the existing bar, jetties do not require
the incessant following ul) supposed; that they may ultimately require
extension I do not dispute."
The recent Board of Engineers (of 1874), to which the whole subject
was recommitted, used, in giving its decision in favor of jetties, substantially, but quite independently, the same argument, though more
fully and clearly developed.
"At present, the muddy water issuing from the South Pass spreads
out in somewhat of a fan-shape, the handle of the fan being at the mouth
of the pass and the ribs several miles in length."
"If the proposed jetties were instantly completed, and the new
channel scoured out, essentially the same amount of sediment would be
spread out in fan-shape, but, from the greater velocity of the issuing
water, the ribs of the fan would be longer, while the handle would be
narrower. More of the sediment would at first be deposited far out in
the Gulf than before,
" But with the present rate of advance, the 25 feet curve 120 years
ago was about 12 000 feet above its present position; and if the volume
of water carried by the pass is kept the same, neglecting the slight difference in slope of the Gulf bottom outside the present bar, in about 120
years a new end for the pass will probably be formed of the same general
shape as the lower 12 000 feet of the present pass. It makes little difference, in the whole time required to accomplish the work, whether the
same volume of water flows out at starting over the present shallow bar
or from between two dikes which force the water to take a depth of 30
feet. In an average of many years, the rate of progress must be about
the same as now, namely 100 feet per annum, the volume of water being
kept as at present; and it is on this basis that the average annual cost of
extension, namely, $130 000, has been computed."
" It has already been stated that it is proposed to obtain a depth of
30 feet between the jetties, in order that some years may elapse before
the shoal which will form beyond the jetties can have on it less than the
* " Perfect parallelism is not necessary; large deviations may be made to select the best
location."




117
required depth of 25 feet in the channel through it. There are no precise data for estimating this period. Going seaward from the upper end
of the proposed dikes, the slope of the bottom of the South Pass is about
—.   This slope doubtless depends mainly on the velocity of the water
flowing through it and on the lifting of the fresh water by the salt. As
the causes remain essentially the same, it would seem natural that the new
end of the South Pass to be formed by the sediment passing through the
jetties should at least have the same bottom slope. If this assumption
were true, the bottom would at least shoal from 30 to 25 feet in a distance
of 5 X 440 = 2 200 feet, and the time required would be about twentytwo years."
Mr. Eads in his pamphlet entitled "The Jetty System Explained"
(printed in 1874), pages 6-7, dwells on and develops the same principle, justly reasoning from the facts of nature, that between the lowest
point at which there is 60 feet depth in the South West Pass (whether
that final 60 feet be in the unmodified pass, or suddenly realized, by
means of jetties, on the site of the present bar) and a bar having only 15
feet, there must be a distance of 7~ miles-and the time for the creation
of a new bar (if by jetties the present bar be deepened to 60 feet) must be
equal to the number of feet in 73 miles divided by the amount in feet of
the present annual advance; a quotient which he assumes to be 178 years.
The arguments are essentially identical in all the three independently
exhibited forms. These reasons are in themselves the broad and indisputablefacts of observed configuration.  All that has, in a different and
conflicting sense, been arrayed is-where not pure assumption-based on
theories* which have themselves, no sufficient inductive basis.
But the matter may be somewhat otherwise stated, thus-the passes
can prolong themselves no faster than they can build up their base of
formation; the broad bank on which their bed is laid and that serves
as the solid trough (as it may be called) which conveys them to the sea.
The front on which the fan-like expansion over which the great delta arms
are laid, has an extent measured on the 100 feet (depth) curve of about 40
miles.  The material if all laid within (and we know it is transported far
beyond) the 100 feet line, cannot advance this front much more than 300
* The two most prominent and sharply defined " theories " of bar formation are those
of Mr. Ellett, (the vertical eddy theory,) and of the "Physics and Hydraulics of the Mississippi," pages 445-6, 7, 8, refuting the former and attributing the bar formation to the heavier
sediments "pushed along the bottom," but which also calls for "vertical eddies" and a
" dead angle, where the river-water meets and rises upon the salt-water." Were this latter
theory proven, it would be far from sustaining the inverse rule-of-three computations of
enormous bar advance, under influence of contraction by jetties, based on it in Ex. Doc. 220,
page 31.




118
feet per annum.  Or if we take the South West Pass by itself (forming a
kind of salient on the western extreme of the general configuration) its
base measured at the same depth is 10 miles, (even on the 18 feet curve
the base is 5 miles,) and this same configuration obtains whether we refer to Talcott's chart of 1838 or the Coast Survey chart of 1867, during
which interval there has been a general progression of 1~ miles. The side
banks, from the pass to lateral deep water, are quite pertinently compared
by Mr. Eads to completed natural jetties; and as such, they must
complete themselves as they advance. There is every reason to affirm
that the advance, whether the material be projected with high velocity
over a bar crest 2 000 feet long, or with low velocity over one 11 500 feet
long, cannot be made on a base of less width than we observe to be essential
to the natural advance.*
Stress is laid on this exhibition of the case because it discards all theorization, and all subsidiary and disputable questions of "littoral currents," "tide," "winds " or "waves."   If in the final minority report of
the writer stress is mainly laid upon precedent, it is because successful
precedent is enough to prove the claim that the "open mouth " should be
striven for at the Mississippi; and because the greatest possible brevity
was aimed at.
It will be noticed doubtless, that in the passage quoted from the minority report as first drafted, preference is given to Pass A Loutre, not because the analogy between the South Pass of the Mississippi and the
" Sulina " (both bearing about 8 per cent. only, of the total discharge)
had not already attracted the attention of the writer; but because in
sketching out, under great pressure as to time, some outline of an application of the jetty system to the Mississippi delta, it was deemed-inexpedient to select an obscure pass-one which has never been known to
navigation-for the purpose.   Pass a Loutre is, on the other hand, in
some degree a rival of the South West Pass, while the application there
would involve very much less expense than to the latter.  Preference was
subsequently given to the South Pass for reasons fully set forth in the
minority report.t
In the foregoing, I have endeavored to give a concise history not only
* The following from the " Physics and Hydraulics of the Mississippi," page 449, is but
anotherform of the above affirmation: "The oscillating motion of waves, when meeting bottorn, is changed into a motion of translation, and this tends to arrange the deposit made by the
river into the same gentle slope at which it disposes similar material at corresponding depths along
the shores."
t Ex. Doc. 220, page 124.




11.)
of actual operations at the delta mouths, but of the various projects
for providing an adequate outlet, whether by operating on the mouths
themselves, or by canalization. I have, at the same time, endeavored to
make plain the grounds on which an " open river mouth " may be confidently hoped for.
In this matter the recent Board (of 1874), in all essential matters, is
in unison with me. Its plans or its estimates are doubtless able and
thorough, and, in absence of any precedent for the kind of work it
proposes, are naturally founded on the best foreign precedents, which
are those furnished by the practice of Holland, and more especially by
recent works at the mouth of the Maas.*   These works were visited by
me in 1871, and fully described with full details of the fascine and ballast
construction (" zink stukken," &c.) in Professional Papers, Corps of Engineers, No. 22, and the methods indicated as likely to " prove especially
available in Louisiana."  But at the same time they can but serve as
models. In practice and adaptation to the peculiar locality, I believe
they may be much simplified and the desired results of the jetty application be arrived at more speedily and economically than by following
rigidly the board's model.
In all that relates to the rival project of a canal, the recent Board has
given incomparably the best solution of that problem yet offered. Nevertheless, while always admitting the practicability so far as the mere
question of construction (a difficult one though it be), I am far from believing it to offer a reasonable assurance of success, if by success is
meant the furnishing of an outlet at all adequate; I doubt much
whether in its working, it will fulfill even that degree of success which
would be justified for similar canals in other localities, and I feel sure it
could not be maintained for the small annual sum estimated.
Before closing this paper I would advert to the case of the Rhone
mouth, upon which much stress has been laid as an unsuccessful attempt
to apply jetties. I have before me a copy of the " Rapport sur les Projets presentes par les Ingenieurs charges de la Navigation du Rhone," t
made by a " Commission " of the "Ponts et Chaussees."
* These, even in Holland, were the first applications of this kind of work to an open sea
exposure. So successful has it proved that the lower legislative body of Holland reports that
(' the complete success of the works at the Hook of Holland has removed all doubts as to thepossibility
"of making piers at sea " (i. e., jetties) " on our coast."
The construction, wholly of beton blocks at the sea entrance of the new North Sea Canal,
has not proved so complete a success.
t Furnished to Gen. Wright, President of the Board of 1874, by M. Malezieux.




120
Stating that the General Council (Conseil General) of the Ponts et
Chaussees had already been engaged with this subject in connection with
two projects presented by Mr. Surrell, it mentions the first as that for a
"canal maritime," &c., &c., and the second, as one for concentrating the
ceaters of the Rhone in one single arm, that of the " Grau de l'Est;" the
Council referred to, stating the principle upon which rested this project
to be,' " that the depth over the bar of a river mouth emptying into the
sea is the greater, the greater the quantity of water which is discharged through it."
Thus we find laid down authoritatively, at the very outset of this matter, not the jetty method but (see enumeration, page 107, made by the
Board of 1852) a distinct and very different one, i. e., that, specified in the
words of the board just cited, as "closing the useless passes." And, indeed,
through the forty closely written manuscript pages of the report before
me, this, and this only, is dwelt upon and discussed in all its bearing,
without even an allusion to the very distinct notion of jetties or, as
more generally known in Europe, that of "parallel piers."
That the actual construction assumed somewhat the semblance (as far
as they went) of "parallel piers," is due simply to the fact that it was
deemed indispensable to connect the dams, which stopped the " useless
passes," by dikes running over the low, intervening islands (theys)
-that, on the east side it was found easier and more economical to make
a continuous dike, riverwards of the islands, than isolated dams.
Finally, this Commission, concludes as follows: "Whereas," * * *
" 3d. The delta arms (grals) of Piemanson and Rouston, which are
the first derivations, up stream, from the main river, receive more than
half of the total discharge.
" 4th. That the. closing of these arms will constitute a valuable experience from which may be derived useful information.
" Therefore, the works to be executed for the improvement of the
Rhone mouth should be confined to the closing of the Piemanson and
Rouston arms."
The government, however, decided to carry out the plan of Mr. Surrell in full; and now let us hear in the language of M. Malezieux (September 10th, 1874) what was done.
The works constructed constitute a continuous dike (" endiguement ")
upon the two banks of the Rhone, from the Tower of St. Louis to the
bar; that on the left has a total length (" develolpement ") of 7 kilometres (4~ miles), and stops at 1 531 metres (a mile less 80 yards), inside the




121
crest of the bar; that on the right has a length of 6 500 metres (4 miles),
and stops at 1 460 metres (-Al mile) inside the crest. * * *  "The result
of these works was to concentrate the waters of the Rhone in one arm,"
&c., &c.
And again, the official journal, " Annales des Ponts et Chaussees"
(1863, 2d semestre), refers to this matter in these words (translated):
" After having attempted to improve the mouth by concentrating all the
water in a single channel, the administration" (i. e., of the Ponts et Chaussees) "renounced their fruitless efforts and had recourse to the construction of a maritime canal," &c., &c.
It seems scarcely necessary to remark that the jetty system has never
been applied to the Rhone mouths.    The method of closing useless
passes having resulted in failure, the work was abandoned without
even an attempt to apply the true jetty system, and the alternative project of Mr. Surrell-a " canal maritime " (ship-canal)-made.  This, in
the words of the report of the recent board, "is more than adequate
to the wants of commerce." Inasmuch as at that locality there are comparatively trifling wants to be supplied, in our sense of such needs, one
might suspect a vein of irony in these words, when we consider what a
"canal-maritime" would be to the wants of commerce of the port of
New York. Even those wants should not rival.the exigencies of the now
enormous, but yet undeveloped, freight-commerce of the Great West
through its only outlet where the Father of Waters bears to sea the
spoils of half a continent!




122
CIX.
ON THE MEANS OF AVERTING BRIDGE ACCIDENTS.
Reports     by  JAMES     B. EADS, C. SHALER         SMITH, THOMAS
C. CLARKE, JULIUS        WT. ADAMS, ALFRED         P. BOLLER,
CHARLES MACDONALD and THEODORE G. ELLIS,
Members of the Society.
PRESENTED     MARCH    3D, 1875.*
REPORT I.
To the American Society of Civil Engineers:
The Committee appointed, under the resolution of May 21st, 1873,t to
enquire into the " most practicable means of averting bridge accidents,"
begs leave to report as follows:
After a careful examination into the causes of the more disastrous
accidents which have occurred during the past few years, it finds that
they can readily be divided into three different classes. First, where
bridges are erected by incompetent or corrupt builders, and accepted
by incompetent or corrupt railway or municipal officials. Second, where
bridges of good design and sufficient material fail from absolute neglect:
on the part of their owners, or from injury to the material during transportion or erection. Third, where bridges, good or bad, are knocked
down or destroyed by derailed trains moving at a high rate of speed, or
where the growth of a neighborhood has brought a class of traffic on a
bridge, which it was not originally designed to bear, either by the builders
* At the regular meeting of the Society, March 3d, 1875, ordered printed and set.down
for discussion at the Seventh Annual Convention to be held at Pittsburg, Pa., June 8th, 9th
and 10th, 1875.
t At the Fifth Annual Convention, held at Louisville, Ky., May 21st and 22d, 1873, it was:" Resolved: in view of the late calamitous disaster of the falling of the bridge at Dixon,
Ill., and other casualties of a similar character that have occurred and are constantly occurring, that a committee * * * be appointed to report at the next Annual Convention the
most practicable means of averting such accidents."
The committee appointed consists of Messrs. James B. Eads and C. Shaler Smith of St.
Louis, Mo., I. M. St. John of Quinnimont, Va., Thomas C. Clarke of Philadelphia, Pa., James
Owen of Newark, N. J., Alfred P. Boller, Octave Chanute and Charles Macdonald of New York,
Julius W. Adams of Brooklyn, N. Y., and Theodore G. Ellis of Hartford, Conn. Mr. Alfred L.
Rives of Mobile, Ala., was appointed on the committee but resigned




123
or the owners. How to treat each of these classes of causes will now be
considered in the order above stated.
Accidents occurring from the first class would certainly not have taken
place had the wrecked structures been correctly designed and had they
possessed the proper sectional areas in their different parts-but failure
from faulty design, is not nearly so frequent as failure from insufficient
material. One great difficulty in the way of protecting the public from
the results of imperfect design or scanty material lies in the absence
of a fixed legal standard of loads and stresses for all classes of these
structures, and another is the negligence of those controlling public
works or those engaging in their construction, in securing skillful professional aid.
It would seem, therefore, to be our duty as a Society to establish in a
few general terms-such as can be readily embodied in a law-a standard
of maximum stresses and a table of least loads for which bridges should
be designed, and to add thereto a practicable suggestion as to the necessary legislation required to give the public that protection which an
adherence to this standard would afford. First, as to the standards for the
least live loads to be used in proportioning bridges; a law which would
provide that all railroad bridges should be built to carry not less than
the following loads, would be well within the mark of safety.
For highway and street bridges the standard loads should not be less
than as in first table on next page; for city and suburban bridges and
those over large rivers where great concentration of weight is possible, as
in column A; for highway bridges in manufacturing districts, or on level,.
well ballasted roads, as in column B, and for country road bridges,
where the roads are unballasted and the loads hauled are consequently
light, as in column C.
POUNDS PER SQUARE FOOT.
SPANS.
A.          B.          C.
60 feet and under.         100         100           70
60 to 100 feet.             90          75          60
100  200 "                   75          60          50
200 "400 "                   60          50          40
With the highway bridge the floor-beam strength is especially important, because of the great concentration of weight which may be carried




124
on a single pair of wheels, therefore the floor system of each class of
bridge should be-per floor-beam for each wagon-way-for city bridges,
6 tons; turnpike bridges, 5 tons, and county bridges, 4 tons.
POUNDS PER LINEAL                POUNDS PER LINEAL
SPAN OR PANEL.                    SPAN OR PANEL.
FOOT OF TRACK.                   FOOT OF TRACK.
Under 12 feet       6 000         Under 75 feet.     3 000
15 "            5 500            100 "           2750
20"            5 000          "  150 "          2 500
25 "           4500         150 to 175 "        2500' 30"             4 000         175 to 200 "       2400
50              3250         200 " 300 "        2250
The panel weights for railroad bridges are obtained by using the
standard weight per foot for short spans. In computing all web members, one panel of panel weight is to be considered as preceding the
standard span load. The proposed law should also provide that with
the foregoing loads, the stresses on materials shall not exceed the following:
For wrought-iron in tension, long bars or rods, 10 000 pounds per sq. in..."   "     "       "     short links, (for
floor beams)............... 8 000   ".. "        "  against shearing force..... 7 500    "     " 
and for wrought-iron in compression, as in this table:
POUNDS PER SQUARE INCH.           POUNDS PER SQUARE INCH.
DIAMETERS.                        DIAMETERS.
SQUARE ENDS. ROUND ENDS.          SQUARE ENDS. ROUND ENDS.
10        10 000     7 000      30 to 35     6 000     4 000
10 to 15      9 000     6 500      35 "40       5 000     3 500
15 " 20       8 000      6 000     40 "50       3 800     2 500
20 " 25       7 500      5 500      50 " 60     3 000     2 000
25  30        6800       5000.............................................
Where one end is square and the other end is rounded, a mean is to
taken between the two.
Cast-iron to be used in compression only, in lengths not exceeding
22 diameters, and at the same stresses as those prescribed for wroughtiron.




125
The shapes under compression in the above are assumed to be hollow
struts either square or cylindrical in section; other shapes than these to
have the stresses varied as actual experiment may dictate.*
For wood, the greatest allowable strains shall be as follows:
For oak in flexure..............1 200 pounds per square inch.
i" pine "    ".............1 000  " 
and in compression as in this table:
POUNDS PER SQUARE INCH.
DIAMETERS.
OAK.               PINE.
10                    1 000              900
10 to 20                 800               700
20 "30                   600               500
30  40                   400               300
The above standard should be changed or elaborated more fully, from
time to time, as future experience and experiments on material suggest.
In order to secure to the public the full measure of benefit from the
adoption of this standard, the law in question should provide for the appointment by the governor in each state, of an expert whose duties would
consist in having cognizance of the construction and maintenance of
every bridge intended for public travel in the state or states for which
he was appointed. The law should also make it imperative that the expert so appointed shall pass an examination as to his mathematical and
mechanical competency, which, it is suggested, should be by a standing committee of this Society, regularly constituted for the purpose,
and that the appointment of any such expert who fails to receive
the endorsement of this committee shall be null and void. Under the
proposed law it should be the duty of all railroad, city, county or state
officials having charge of the letting or construction of bridges to call
upon this expert-first, to examine the strain-sheet of the proposed structure before work has been commenced, to certify to its correctness if
correct, or to make such alterations as may be necessary if it is faulty
in design or scant in material according to the legal standard; next, to
be present on the completion of the bridge, and then and there to make
* Further experiments can alone determine the values to be used for other than square
or cylindrical cross sections.-A. P. B.




126
a critical examination of the work in all its details, comparing and verifying the sections on the strain-sheet with those of the actual structure,
and if these last are insufficient, to forbid the use of the work until the
law is fully complied with; and, lastly, if the bridge is up to the standard in all its parts, to obtain from him a certificate to that effect, copies
of which certificate shall also be given to the builder, and filed on record
in the proper department of the state government.  This officer shall
also see that a tablet or plate is placed on a conspicuous part of the
bridge, bearing the names of the builders, his own name, and that of
the officer of the railway or corporation who accepts the work, together
with the strength of the bridge as designed, and the year of its erection.
Accidents arising from the first class of causes would be nearly, if not
quite prevented by the general enforcement of the foregoing provisions.
Against accidents occurring from causes of the second class, the law
should further provide that all railway or other corporate bodies, when
having a bridge built, to be used for public travel, shall be compelled
during the erection of the work, to keep on the spot a competent
inspector, who shall have power to reject any piece of material which
may have been injured in transportation or while being placed in
position. Also that all railroad and city bridges shall be inspected once
every month by a competent person in the employ of the corporation
owning the bridge, for the purpose of seeing that all iron parts are in
order, all nuts screwed home, that there are no loose rivets, that the iron
rails are in line and without wide joints, and that all wooden parts of
the structure are sound and in proper condition.
It should also be the duty of the state officer before mentioned, upon
any bridge being reported as in a neglected condition-whether the report be an official one or made by one not connected with the corporation-to proceed to the spot and examine for himself, and if he finds the
bridge in a neglected or dangerous condition, he should cause the owners
to put it in safe order without delay.
In relation to the third class of causes-destruction by derailed trains,
high winds, or by concentration of living weight owing to the growth of
cities or neighborhoods-prevention is less easy, but much can be done
by carefully designing the structure.  In most of our railroad bridges
the floor system is the weak point.   The cross-ties are short, the
stringers are proportioned for a train on, not off the rails; and the
guard-timbers are too low, and are insufficiently bolted.  A derailed
engine on such a floor as this, plunges off the end of the cross-ties into
the open space between the stringers and the chords, and generally




1.27
wrecks the bridge. To obviate this, the law should provide that, first,
-all cross-ties shall extend from truss to truss, they shall be placed
so close to each other that if supported at the proper intervals it will
be impossible for a derailed engine to cut through them, and the
stringers shall be so spaced as to give them this support. Next, the
guard-timbers shall be scantlins not less than 9 X 10 inches, and
they shall be strongly bolted or spiked to at least each alternate crosstie. And lastly, the clear width between the trusses on through bridges
shall be so great that the wheels of a derailed train will be arrested by
the guard-rail before the side of the widest car can strike the truss.
Where switches are placed at the end of a bridge, the Wharton or some
other form of safety switch should be used.
Against the majority of accidents from high winds, a provision in the
law requiring that all lateral bracing shall be sufficient to resist a pressure of 30 pounds per square foot of truss and train, will be sufficient.
Lateral bracing can be proportioned at 15 000 pounds per square inch
against this particular strain, as it is of very rare occurrence.
The last case in the third class of causes of accidents is where a
bridge built originally for a neighborhood or country road becomes too
weak for the requirements of a growing community or possibly of a newly
established manufactory; also where a railroad bridge, intended only
for that class of traffic, has a highway floor subsequently added to it.
Against the first contingency, the vigilance of the state official and the
chance that some one of the users of the bridge may occasionally notice
the tablet setting forth its strength, would seem to be about the only
safeguard; but in the second case, the law should provide that-except
by permission of the state officer in charge of bridges-no corporation
or other bridge owner shall add to the dead weight on a bridge without
at the same time making the proper addition to its strength.
The foregoing provisions, if embodied in a law, will afford the public
about all the protection which is readily obtainable in the case.
No mention is here made of the quality of the material, as the proposed officials engaged in carrying out the law will be men who have been
passed on by the Society, and the very fact of their surveillance will be
apt to produce care in this regard. In addition to this, the standard
stresses have been placed so low that the use, whether accidental or
fraudulent, of low grades of iron will hardly endanger the work. A
provision in the law that all bridge details shall possess the proper proportional strength to that of the main members of the bridge, and a
series of instructions from the examining committee of the Society to




128
those who pass their examinations for appointments under this law, in
reference to these proper proportions, will protect the purchasers of
bridges from insecure details of construction.
In addition to his duties, as above defined, the state officer in charge
of bridges should also visit the scene of any accident in his district as
soon as possible after the occurrence, and remain during the removal of
the wreck, or until he is able to ascertain the true cause of the failure.
The facts in the case should then be reported by him to the examining
committee of the Society.
In conclusion, it is here advised that a committee be appointed to
draft such a law as is outlined in this report; that a resolution be
passed by the Society recommending the adoption of this law by the
different state legislatures, and that printed copies of this report, the
proposed laws and the accompanying resolution, be sent to the members
of the Society with a request that they move actively, each in his own
state, towards procuring the passage of the specified law by the various
state legislatures during the coming winter.
JAS. B. EADS, Chairman.
October 30th, 1874.                             C. SHALER SMITH. *
* In advocating the views presented in the foregoing report, the undersigned is actuated
by the following reasons:
First-the resolution under which the Committee is acting requires from it "the most
practicable means of averting-i. e., preventing-bridge accidents "-rather than the mode of
sitting in judgment on them after they occur.
Second-as the national legislature has for some time been passing laws for the protection
of life on the navigable waters of the United States, prescribing qualifications and standards
for engineers and pilots, the proportions of safety valves, &c., for boilers, and appointing
examiners and inspectors under these laws-so, sooner or later will the question of the proper
construction of railways be taken up and legislated upon.
Third-many mistakes have been made in these laws, owing to ignorance on the part of
those passing them, and the undue influence of interested inventors and manufacturers, and
each succeeding Congress has had amendments to make in order to repair some injustice or
supply some omission.
Lastly-as laws regulating the construction of railroads and bridges will certainly be
enacted, and official positions will assuredly be created by them-it is far better that this
Society should take time by the forelock, dictate a law which will be just and equitable,
and hold control of the appointments under it, than that it should stand in the background,
until an aroused public opinion compels legislation which may be injurious to the profession,
especially if enforced by political appointees who may be utterly unfit to fill such positions.
All laws are written by some one, and the greater the knowledge of the subject matter on the
part of that person is, the more probable is the production of a good and wise statute.
Hence the undersigned believes that the fixing of the standards as proposed, the preparation of such a law as suggested, and the professional surveillance of the appointees under it,
are eminently the province of this association, and that all legislation on the subject should be
both inspired and dictated by the most competent authority in the premises-the American
Society of Civil Engineers.
April 18, 1875.                                            C. SHALER SMITH.




129
REPORT II.
The undersigned differ from the views expressed in the foregoing
report, and present the following as an expression of their own:
I.-They agree with the report, that it is desirable the American
Society of Civil Engineers should publicly declare what it considers
to be a standard bridge, anything below which is not to be deemed
as a safe or durable construction.  But they do not think it is desirable to go much into detail, as they believe it to be impossible to
construct a specification that will meet all cases. Incompetent engineers
cannot be prevented from building bad bridges by any specifications
however elaborate; they therefore are content with laying down general
principles, leaving the application to others, and offer the following
standard specification for bridges of iron and wood:
1. Every highway bridge shall be capable of carrying, in addition
to its weight, a moving load per square foot of roadway and sidewalks
as follows:
SPANS-FEET.        POUNDS.        SPANS-FEET.        POUNDS.
100 and under.         100           300 to 400            60
100 to 200            80          Over 400               50
200   "   300         70..............
2. Every railway bridge shall be capable of carrying on each track, in
addition to its own weight, 2 locomotives coupled, weighing 30 tons on
drivers in space of 12 feet, and whose total weight, including loaded
tenders, is 65 tons each; said locomotives to be followed by so many
loaded coal cars weighing one ton per lineal foot, as will cover the remainder of the span.
3. Bridges shall be so proportioned that the    above loads shall
not strain any part of the material over one-fifth of its ultimate
strength.
II.-The signers of the foregoing report, propose to cause future
bridges to come up to the standard by a system of inspection, the inspectors to be passed by the Society before being appointed.  The undersigned believe that in the present state of public opinion this is impracticable. If any inspectors are appointed, it will be by political influence,
and the results will be worse than at present, as the inspection will be




130
inefficient, and yet, to a great extent, relieve the owners of bad bridges
from legal responsibility.
The undersigned consider that the most the Society can hope to do,
is to provide means in case of the fall of a bridge, by which the responsibility of imperfect construction (if this was the cause of the accident)
may be fixed on the designers and builders, and iron manufacturers.
It is therefore recommended that the Society prepare and present to
the state legislatures, a petition embodying the following data:
1. That the standard of the American Society of Civil Engineers shall
be the legal standard, and in case it shall be found that any bridge is
of less strength than this, it shall be taken as prima facie evidence of
neglect on the part of its owners.
2. That no bridge shall be opened for public traffic until a, plan, giving the maximum loads it was designed to carry, the resulting strains,
and the dimensions of all the parts, sworn to by the designers and
makers, and attested by the signature of the proper officer representing
the municipality or corporation by whom it is owned, be deposited in the
archives of the Society, and that the principal pieces of iron in the bridge
be stamped with name of maker, place of manufacture and date.
The result of this will be, that in case of the fall of a bridge, the
responsibility can be directly and easily traced to the right party, which
at present cannot be done, and the Society should willingly aid to such
a purpose. This, it is recommended, should thus be done; the Society
to appoint a committee-with compensation to be fixed by law-which,
upon the call of the executive of any state, should visit and report
upon any fallen bridge, care being taken that no parties interested in
the construction of the bridge be upon the committee.
It is believed by the undersigned, that the knowledge all bridge
builders would have that their misdeeds, if any, could, by this process,
be traced home to themselves, would make them very careful in the
future, and eliminate all failures from imperfect design or material.
As to the inspection of existing structures; if the Society assumed
the first duty, this would soon fall under its jurisdiction, and if it would
volunteer the duty-in case any plan was deposited obviously unsafe-to
protest against it, that also would be well; such would have prevented
the fall of the Dixon bridge, and the lamentable loss of life and limbs
there occurring.                              THOMAS C. CLARKE.
February 1st, 1875.                       JuLrUS W. ADAMS.




131
REPORT III.
The undersigned differ from the views expressed in the foregoing
reports, and respectfully present the following:
1. The members of the committee agree that it is meet and proper
the American Society of Civil Engineers should determine the standard
of strength for all bridges to be built in this country, and they further
agree in the main, what this standard should be. The differences in
opinions grow out of the methods for incorporating this standard in the
every-day practice of the country. Two general modes present themselves
for so doing; the one legislative and compulsory, and the other looking
forward to directing public sentiment to right conclusions by a thorough
dissemination of the adopted standard.
2. The undersigned advocate the latter method as the true policy of
the Society, believing that any attempt to influence the enactment of
laws that would be so far-reaching as the ones proposed, is impracticable,
if not contrary to the genius of the Society itself. They further believe
that when once public sentiment is aroused by the publicity which should
be given to the adopted standard, it will compel the passage of laws covering the question.
3. The undersigned therefore suggest that the report to be accepted
be simply one covering a standard strength for all bridges, in as general
terms as possible, with a recommendation that such standard be widely
disseminated by circular and the public prints, and that copies be distributed among the legislative bodies of the several states.
The following standard, culled from the foregoing reports, is proposed
for adoption:
4. For highway bridges, as submitted in Report I, page 123.
5. For railroad bridges:*-the structure shall be at least capable of
carrying on each track, in addition to its own weight, 2 locomotives
coupled, weighing 30 tons on drivers in space of 12 feet, and whose total
load, including tender, is 65 tons each. Said locomotives to be followed
by as many loaded coal cars, weighing one ton per lineal foot, as will
cover the remainder of the span.
Bridges to be so proportioned that the above described loads shall not
strain the several parts in excess of one-fifth or one-sixth of the ultimate strength. In determining the strains produced by the above stand* Being the same as submitted in Report II, page 129.




132
ard, it is to be understood that the chord system is to be computed for
a uniform loading, while the web strains must be based upon the irregularly distributed or concentrated loads produced by the above described
train, in its passage from one end to the other.
The following table represents the uniform distributed moving load
for different spans:
SPN OR P L.  POUNDS PER LINEAL.     PAN      POUNDS PER LINEAL
SPAN OR PANEL.   FOOT OF TRAC.    SPAN OR PANEL.  F      R
FOOT OF TRACK(.                  FOOT OF TRACK.
12 feet.         5 250             75 feet.       3 000
15 "             5250              100 "          2750
20 "             5 000             150 "          2 500
25 "             4 500             175 "           2 400.30 "            4 200             200 "           2 300
50               3 250        200 to 300 "        2 250
The extreme panel weight for all spans is obtained by using the
standard weight per foot for short spans.
6. Under the standard loading, as expressed in above table, materials
should not be strained in excess of what is submitted in Report I, page
124.
All of which is respectfully submitted.
ALFRED P. BOLLER.
March 1st, 1875.                         CHARLES MACDONALD.
REPORT IV.
While agreeing in many important particulars with the report of the
Chairman of this Committee, the undersigned holds the views expressed
by some of the other members regarding the expediency of compulsory
legislation on the subject. It is believed that the opinions of this Society as a body, advanced for its interest and benefit and that of those
who should choose to be governed by them, would have more weight and
influence than though the Society should assert itself as the only competent authority upon bridge construction. If this Society adopts a well
defined standard of strength for bridges, it is believed that the public
generally will wish to conform to it, and engineers even who are not




133
members will be glad to avail themselves of the united opinion of so,
many of the profession.
There seems to be a unanimity of opinion among the members of the
Committee as to what constitutes the ordinary load upon a railway bridge,
and but a slight difference of opinion as to its amount.
From an examination of the weights carried upon many of the principal railways in the United States, it is found that the heaviest engines
weigh about 2 830 pounds per foot; and that three, and sometimes four,
are coupled. The heaviest weight on one pair of drivers is from 21 000
to 24 000 pounds, and the weight on all the drivers, generally not exceeding 12 feet wheel-base, is from 72 000 to 84 000 pounds. The heaviest
trains may be assumed to weigh 2 250 pounds to the running foot, exclusive of the engines. As the coupling of more than two engines is mainly
upon snow roads, it is not believed they should be included in a general
rule for proportioning bridges, but should be classed among those exceptional cases for which a general provision cannot be made.
In view of the above, it is believed that all railway bridges should be
proportioned for a rolling load of 3 000 pounds to the foot for the total
engine length, and for 2 250 pounds to the foot for the remainder of the
bridge; that the bracing on each system should be proportioned to sustain 84 000 pounds on any 12 feet of track, and that any point on the
track should sustain 24 000 pounds.
It is not believed that the system of expressing the loads that a bridge
should carry, by so much per foot with a varying amount for each length
of span, is the best; but if such a standard is to be adopted, the table*
in the report of the Chairman is believed to be the best of those given,.
although it is somewhat below the loads actually carried by many roads
in this country.
The floor-beams of railway bridges should be proportioned for not less
than the following loads:
SPACES.         POUNDs.        SPACES.         POUNDS.
4 feet apart, or less.  28 000  12 feet apart, or less.  42 000
6..             31 500    15    (             45 000
8                    35000                          3 000
More than 15 feet apart 
10..       38 500                         to the foot.
* Page 124.




134
The following table is offered as embracing the foregoing loads when
reduced to so much per lineal foot:
POUNDS PER LINEAL                 POUNDS PER LINEAL
SPAN OR PANEL.                    SPAN OR PANEL.
FOOT OF TRACK.  I                 FOOT OF TRACE.
12 feet or under.     7 000         50 feet.           3 000
15 feet.             6 000         100                 2 800
20 "                  4800          200 "              2600
25 "                 4000          300 "               2500
30 "                 3 600         400 "               2 450
40 "                  3 200        500 or over.        2 400
For intermediate lengths of span the proportional number of pounds
per foot should be taken.
These loads do not include the extraordinary weights that are sometimes drawn over railways in this country; such as heavy pieces of
machinery, blocks of stone, or a locomotive of different gauge on a truck
car, nor more than two engines coupled.  These are exceptional cases
which can be provided for when they may be expected to occur, and the
weight can ordinarily be distributed so to cover a sufficient length of
track as not to exceed the loads above given.
For the effect of wind, the maximum strain is believed to be about 40
pounds per square foot horizontal, and about 20 pounds per square foot
vertical.
For highway bridges, the following table is offered as a substitute for
that given in the report of the Chairman* for the three classes of bridges
named.
POUNDS PER SQUARE FOOT.
SPANS.
(Intermediate lengths in proportion.)  A.     B.            C.
100 feet and under.             100           75           60
200                              80           60           50
300 "                            70           50           50
400 "                            60           50           50
500 " and over.                  50           50           50
* Page 123.




135
The floor-beams and flooring should be of sufficient strength to sustain
the following loads on four wheels:-Class A, 24-B, 16-and C, 8 tons
respectively. These do not include the extraordinary loads sometimes
taken over highways. They are exceptional cases and the weight can
generally be divided.
With regard to the factors of safety to be used, it is believed that a
less factor is required for the permanent and unchanging dead load, than
for the vibrating and uncertain live load, which may, by accident, be increased beyond the limit for which it was computed. This, together with
the fact that a larger factor for the dead load gives no additional strength
to the bracing near the middle of the span, but only at the ends, leads to
the following substitution for the factor offered in the report of the
Chairman.
For wrought-iron and steel in both compression and extension-for
the dead load including snow, — for the live load including wind, l the
ultimate strength.
For cast-iron in compression only, and for lengths of not more than
20 diameters-for the dead load i, and for the live load, o0 the ultimate
strength.  For large masses, as in arches, a factor of i the ultimate
strength may be adopted.
Bridges should be tested with the maximum loads which they are
intended to sustain.  A less load would seem to be of but little use,
and a much greater one might unnecessarily strain the structure.  The
load should be applied gradually, and the moment any undue deflection
or crippling is observed, or the slightest diminution in the transverse
section of any bar is occasioned, the load should be immediately removed
and never repeated. If no actual rupture occurs, the bridge will probably be safe with 0.4 of the test applied. The acceptance of all bridges,
after being constructed with proper proportions for the material used,
should be subject to such a practical test.
THEODORE G. ELLIS.
April 20th, 1875.




136
Cx.
ON THE FORM, WEIGHT, MANUFACTURE AND LIFE OF RAILS,
A second Report, by ASHBEL WELCH, C. E. (Chairman),
Member of the Society.
PRESENTED MAY 5TH, 1875.*
The members of the Committeet on Rails desire, when they make
their final report, to present some definite results, which a wide experience alone can establish. They therefore earnestly beg their professional brethren to state their experience on the subject, in response to
the interrogatories issued by the Committee last year,+ or in any other
way they may think best.~ In the meantime, a few points now settled
will be presented, and a few questions raised for the consideration of
the Society.
Forms of rails on the same principles as those recommended by
the Committee have been adopted on three of the great railroad systems
of the country-the Erie, Lehigh Valley and Pennsylvania.          The Pennsylvania has also adopted the plan of a deeper head for main lines, on
nearly the same base as a shallower head for branches; the base for main
lines is 41 inches.
The Committee considers it settled, that with the present weights on a
wheel, a thickness of the middle of the stem of seven-sixteenths inch
* At the adjourned regular meeting of the Society, May 12th, 1875, presented, ordered
printed, and set down for discussion at the Seventh Annual Convention at Pittsburg, Pa.,
June 8th, 9th and 10th, 1875.
t Appointed January 8th, 1873, to determine the "the best form of standard rail sections
of this country; the proportion which the weight of rails should bear to the maximum loads
carried on a single pair of wheels of locomotives or cars; the best methods of manufacturing
and testing rails; the endurance, or, as it is called, the'life' of rails; the causes of the
breaking of rails and the most effective way of preventing it, and the experience of railways
in this country in the use of steel rails."
The committee consists of Messrs. Ashbel Welch, of Lambertville, N. J., M. A. Forney
and Octave Chanute, of New York, and I. M. St. John, of White Sulphur Springs, W. Va.
$ Referred to in Proceedings, Vol. I., page 76.
~ The Chairman of the Committee is instructed by his colleagues to give "particular
Jesse" to the members of the Society who have furnished no answers or information;being
utterly unskilled in handling that terrific weapon (if it is a weapon), he asks each recusant to
consider himself the recipient of a tremendous poke, and keep the fear of a similar punishment before his eyes hereafter-if he continues in default.




AMERICAN SOCIETY OF CIVIL ENGINEERS.
INCORPORATED      1852.
T'lRANSA.CTIONS.
NOTE.-This Society is not responsible, as a body, for the facts and opinions advanced in any
of its publications.
XCVIII.
UPRIGHT ARCHED BRIDGES.
A Paper by JAMES B. EADS, C. E., Member of the Society.
RT.\1) JUNE 10, 1874.
In a report on the Illinois and St. Louis Bridge, in 1868, I advauced the proposition that, for railroad purposes, an upright arched
bridge could be more cheaply constructed than was possible by the
suspension system.  This postulate called forth soon after, some dissenting views from engineering journals to which I had not, at the
time, the leisure to reply.
In this paper I will endeavor to show that upright arched bridges can
l)e more economically constructed than is possible by any other method
whatever, no matter what length of span may be required. As the cost
of the span (excepting the cost of roadway) will, by any system, increase
at least as rapidly as the square of the distance spanned, the length of
span will be limited in all cases, by financial difficulties, before grave
engineering ones are encountered.
I will first point out some of the disadvantages of the system adopted for the superstructure of the St. Louis Bridge, which was the best
method of arch construction then known to me, and will respectfully
venture to indicate such improvements as have suggested themselves, in
consequence of these difficulties being so forcibly thrust upon my attention in the progress of that work. An earnest wish to make these
and their remedies thoroughly understood, will I trust, excuse in me
much that may seem commonplace or didactic.
15




196
In the St. Louis Bridge the upper and lower members, which constitute a single rib of the span, are tubular in form, the tubes being each
18 inches outside diameter. The rib thus formed is 12 feet deep from
centre to centre of the tubes, and the two lines of tubing are braced
together by a single triangular system of vertical bracing, formed of flat
bars secured to the tubes at points about 12 feet distant from each other
throughout the length of the rib, the bars being placed in pairs, one on
each side of the tubes, and secured to them by pins passing through the
ends of the bars and the couplings of the tubes. An upper and lower
line of tubes thus braced constitutes one rib of an arch or span. The
curve selected for the ribs was, for greater convenience of manufacture,
the segment of a circle. Each individual tube is straight, the curvature
of the rib being accomplished at the junction of these individual members, each of which is about 12 feet long. The curve of the rib differs
from a parabola but a few inches.
If the effects of temperature could be avoided, and the curve were a
parabola, an equally distributed load on a rib would, of course, be borne
by the upper and lower tubes equally, that is, half on each; and the
only strain on the bracing would be in transferring half of the imposed
load to the lower tube. In this case the tubes could be together and the
bracing dispensed with. We should then have, in the two lines of tubes,
the least possible section requisite to sustain this equally distributed
load. To brace the arch against unequal loading, the tubes were placed
12 feet asunder.
The medium temperature was assumed at 60 Fah. The effect of temperature (ranging from - 20 to - 140-' Fah.) increases the length of the
rib about 6 inches.  This extension causes the crown to rise, which
relieves the lower tube of compression at the abutments, and hence that
tube does not then support any portion of the weight of the rib or the
load. These are borne wholly by the upper tube at the abutments,
and hence its section there must be increased     accordingly.  At
the crown, the strains are likewise changed. There, however, the lower
tube does all the duty, as the upward bending of the rib, relaxes the
compression in the upper tube at this point, hence the lower one must
be increased at the centre of the rib to enable it to do this double
service.  It is thus seen that the upper tube at the abutments, and
the lower one at the crown, must, when the temperature is at its
maximum, perform the whole duty of sustaining the rib and its load.
The strains in the respective tubes decrease from these points toward the




197
haunches of the arch, where the centre of pressure passes through the
middle of the rib. At the haunches the compression is borne in nearly
equal portions by the upper and lower tubes. The extra section required
by temperature, therefore, becomes smaller in the tubes as we leave the
crown and abutments.
These are the effects of expansion. As the ribs are formed for a
medium temperature, contraction by cold reverses these strains. The
crown of the arch falls, and the upper tube at that point must be reinforced, because it must then sustain the entire compressive strain,
which, at medium temperature, is equally borne by both; while, at the
-abutments, the lower tubes have now to carry all the load. Thus it is
seen that, both at the abutments and the crown, the upper and lower
tubes must be each of greatly increased section on account of the
-changes of temperature.
These changes, in very extreme cases, will probably tend to withdraw the one tube or the other from the abutments, as extreme heat
or cold affects the rib. At the suggestion of my chief assistant, Col.
Flad, to give greater resistance to a change of form of the rib under
unequal loading, the skewbacks in which the tubes are fastened are
rigidly held to the abutments by anchor bolts.  Thus, at times, a tube
at the abutments may not only be relieved of all compression, but may
bear tension and pull upon its skewbacks, by which an additional compressive strain m# be transferred through the braces to the other tube.
For this contingecy additional section had to be provided at the abutments in the firstfwo or three tubes.
Greater depth of the rib would have increased all these strains of
temperature, and would thus have involved larger sections at the crown
and abutments. By lessening the depth, the strains would have been
diminished, but a more flexible rib, under unequal loading, would have
resulted. Between these evils on either hand, after various calculations,
the most economic and satisfactory results were obtained with 12 feet
depth of rib.
A diagram of the strains of the central arch, on which my remarks are
based, is appended, (Fig. 25). From these it will be seen that effects of
temperature have compelled the use of a much larger quantity of metal
in the ribs, than would have been necessary if these effects could have
been avoided.
No one will, I think, deny that the cheapest possible form in which
steel or iron can be used to span an opening and support an equally dis



198
tributed load, is the catenary or suspended arch. It is equally evident:
that if an upright arch has the same span, weight and curve of any
given catenary, the strains in the one form would be no more intense
than in the other.  In the upright arch they would be compressive
strains, and in the suspended arch they would be tensile ones. In the
suspended form, the section of metal may be solid; in the upright, it
must be hollow, tubular, or of some construction that will resist buckling; but to be of equal strength it must not necessarily be more weighty.
If either form have an equally distributed load placed on it, no bracing
will be requisite to preserve its normal curvature. The same load, if
within the elastic limit of the metal, can be borne by either with the
same amount of material. Hence, the only difference in material required
for the two systems for any given span and load, must lie in the difference of bracing them to resist distortion when - uequalbly loaded. I refer
here simply to the span between the supporting towers in the one case
and the abutments in the other. Between these points, with the same
curvature, the sectional areas of each may be the same, under equal
loads.  In this I, of course, assume that the iron or steel can be
trusted with equal strains in tension or compression. In compression,
tubular or other sections are quite capable of sustaining, in wrought-iron
or steel, all that it is safe to apply in tension if the lengths of the compressive members do not exceed 8 or 10 diameters. In the St. Louis
Bridge however, they are but 8 diameters.
The section of arch is doubled by doubling the span, and in the suspension cable it is evident that a long span involves equal sectional areas
also in the shore spans, or portions between the towers and anchorages.
Whereas, when the stream is spanned ly a long arch, the need of such
extra section in the shore spans does not exist, because they can almost
always be short spans. Mr. Peter W. Barlow, C. E., in 1860, published
a pamphlet on the Niagara Railway Suspension Bridge, from which I
quote the following remarkably clear explanation of the fact that in all
known systems of girders, with any given depth, fozer times as much
metal is required to obtain the same rigidity as is needed with the arch
or the cable.
" Let A C B, Fig. 1, represent an arch supported on abutments A and
B, and let the deflection produced by a given weight, loaded equally, be
represented by unity. Now, let us consider the effect of making this
arch into a self-supporting structure, or bowstring girder, by removing
the abutments and substituting a tie, A B, Fig. 2.




199
" Assuming the same weight, w, to be placed equally all over, the deflection will be 2, the points A and B being no longer rigid, because the
tie A B will extend as much as the arch A C B will compress. Therefore, to produce the same rigidity in a bow string girder four times the
metal is required as compared with the arch. The same result arises
in a cable A C B, suspended from the two fixed points A B, Fig. 3.
" If the back chains are removed, and a compression tube A B substituted, the metal is doubled and you have a structure with only half
the rigidity. The Chepstow Bridge on the South Wales Railway is an
example of this arrangement.
" The mechanical combination in the Saltash Bridge is represented by
substituting the arch A D B for the tie A B, Fig. 4, forming a combination of a suspension chain and an arch. The arch A D C will not perform the duty of compression unless it is connected with the chains by
the ties (ta, a a. When thus connected both the cables and the arch
assist in supporting the weight of the load. The points A B now become fixed points, and as both the arch and the chains assist in supporting the weight, the deflection will only be half that of the simple suspension cable with double the weight of metal. It therefore appears:
"1. To convert an arch supported on two fixed abutments into a bowstring girder, four times the metal is also required to support the same
weight with the same deflection.'2. To convert a cable susplended from two fixed points into a Chepstow girder four times tlhe metal is required to support the same weight
with the same deflection.' 3. To convert the same cable into a S:altash combination (which consists of a bowstring and Cllepstow girder combined, so that the horizontal tie in one case neutralizes the comlpression tube in the other, by
which they are both avoided), the deflection is reduced one-half, with
double the weight of material, or the same weight of material will. produce tlhe same deflection with the same load, as in the case of the simple
arch or cable. But this is obtained at the expense of double the depth;
and if the arch or suspension cable was of the same depth as the Saltash,
only one quarter of the metal would produce the same stiffness.
* In the preceding illustration, the bowstring and Saltash girders
are referred to-parallel girders are more commonly used-but they present no economy over tle simple bowstring, and however perfect their
arrangement and proportions, they will still require not less than four
times the metal of a simple cable of the same depth and span to produce
the samle deflection."
Enough is shown by Mr. Barlow in these extracts to prove a far
greater economy in the metallic arch or catenary than is possible with
any truss system known. In practice, however, the chief causes which
have conspired to lessen this evident economy are: 1st, the bracing requisite to preserve the form of the arch when unequally loaded; 2d,




200
the effects of temperature, and 3d, the necessity of heavier masonry to
resist the pull of the cables or the thrust of the arch.
I cannot better illustrate the notably small amount of bracing or
"girder-power" required to resist the effects of unequal loading in the
suspended arch or cable than by again quoting from the same distinguished author. Mr. Barlow says:
" If the beam A B, Fig. 5, be divided into two beams by being supported at C, the two half beams, A C and 3 C., will deflect one-eighth of
the amount of the entire beam A B with the same weight. Let us
assume this to be a girder attached to a chain, and a load placed at D,
the effect will be to distort it into the shape shown in Fig. 6. The deflection by the weight at D will cause a corresponding elevation at the
point E, and the girder will assume the shape represented by dotted
lines in the figure, to produce which a force equal to double that for a
given deflection on half the beam is required, from which it is evident
that the wave produced by a given weight at D will only amount to onesixteenth of the deflection the same weight will produce on the entire
beam resting on its two ends.
"In the above proposition it is assumed that the beam is supported at
its centre point only; in practice, when attached to a suspension cable, it
is supported at various points of its length, the difference between the
wave of a supported girder and the deflection of an unsupported girder
will therefore be greater than one-sixteenth."
"In order to arrive at the result by experiment, I had a model of the
proposed Londonderry bridge on a scale 3l' of the actual span, the length
being 13 feet 6 inches between the bearings, a length exceeding that of
the average of the models used by the Iron Commissioners in their
experiments, and is amply sufficient-due allowance being made for
the scale-to determine the accuracy of the deflections on the actual
girder.  The principal object of these experiments was to ascertain
the deflection of the wave of a girder attached to a chain, as compared
with the deflection of the same girder detached, which being obtained, it
is perfectly easy to arrive at the deflection of the wave of any required
suspension girder, because we have sufficient experiments on actual girders of various dimensions to obtain the deflection from a given load on
the same girder not attached to a chain. These experiments gave a mean
result of 2~,, so that, it being first determined what amount of deflection
is to be the limit with a given load in a given bridge, you have only to
arrive by calculation at the sections of metal of a girder of the same
depth which would deflect 25 times that amount."
The economy of a continuous girder, resting on a central support, depends upon the several supporting points remaining absolutely stable, or
in line. The sinking of one or the elevation of another, creates extra strains.
in the girder which may destroy all the advantages of continuity. The great




201
economy of girder power theoretically shown by Mr. Barlow is not attained in any suspension bridge of which I have knowledge, because the
effects of temperature have not been avoided in their construction. The
extension and contraction of the cables cause the central part of the
girder to fall and rise, whilst the ends of the girder remain at a constant
level. This is not avoided even where the towers are of iron also. Additional strains are thus created in the girder, for which increased sections are required. These strains can be avoided however, by jointing
the cables at the centre of the span, and cutting the girder in two at the
same point. By this means each half length of girder trusses the half
length of cable to which it is attached. If the end of each half girder
be securely fastened to the central joint, the economy of a continuous
girder may be attained, because the effect of the load in bending the one
half girder down is resisted by the other, just as it would be if the two
were continuous. The loaded one cannot deflect without distorting the
cable above it, and the cable cannot deflect, without rising where it
is unloaded, which movement could only occur by bending up the unloaded half girder beneath it. In this case the cable must be likewise
jointed at the towers, as each half of its length is rendered rigid by its
connection with the half girder, and the half girders must also be allowed to expand and contract at the towers.
Fig. 7 illustrates this method of eliminating the strains of temperature
from the suspension bridge, and securing the fullest benefit of " girder
power" with the greatest economy attainable by this system of bracing.
If Fig. 7 be reversed, as in Fig. 8, the dotted lines will show that exactly
the same economy of bracing or "girder power" is attainable in the upright arch by this method, except that in one case the vertical connections act as ties, and in the other as struts, and inasmuch as the compressive strains on the long struts will necessitate greater sections than the
tension strains require in the ties, a slight advantage is found in the
suspended arch. Owing to the weight of the cables, this advantage will
be increased by a difference in the expense of wind bracing. In this
last item, the difference will be found to be less, however, than is
generally supposed.
This fact may be thus illustrated; if a column of metal be suspended by its upper end, its weight will resist the deflection caused
by a current of wind that would blow it over, if standing on its
base.  It must be remembered, however, that if suspended, the
resistance of the weight is only effective after the centre of gravity




202
of the column has been moved.   When vertical, its resistance is nil;
the maximum    effect being exerted when the horizontal position is
reached by the column. Therefore to preserve the lower end absolutely
from movement by the wind, would require as strong a brace or guy as
would be needed to preserve the same stability at the upper end if the
column were standing.  The case is identically the same in the suspended
and upright arches. The bridge should be braced against all movement
by wind as far as possible, and hence the stability of the cables due to
their weight, will not, on investigation, be found to possess as much importance as has been claimed for it. The plan of spreading the arch at
its bases, as proposed by Telford in his 600 feet span for the Thames,
has been applied in its reverse form to the suspension bridge, as a means
of securing greater stability against wind pressure. Either application
of this feature, suspended or upright, involves the necessity of increased
section in the cable or arch, the supporting power of either being lessened as the plane of the curve inclines from the vertical. The increased
section therefore involved in this means of securing stability should be
charged in the estimate of wind bracing.
The illustration I have given of the suspended and standing column,
may likewise be referred to here, to show that for rigid structures, the
weight of the cables in preserving the suspension bridge from a change
of forim under unequal loading is not an element of so much importance
as advocates of that system claim.  Let A 1B, Fig. 9, be a suspended
arcll loa.led at I), anl  C be the point to which the greatest curvature has
been removed by the load from the centre E, of the span where it was
when the cable was in equilirium. By reversing the figure, we see at
once that in the upright arch, the flattening from A to C, increases the
thrust of tlhat p-oitionl aginst the remainder of the arch, which being
without load, and of inferior weight, can only be sustained by the
strength of its bracing or girder power. This change of form in the arch
lessens its ability to resist the effect of the load, and even increases the
power of the weight, to further distort it by the additional horizontal
pressure against the portion ( B, due to the flattening of -A C, whilst
the curvature of C l becomes less favorable for resisting this horizontal
pressure. In the cable, these conditions are exactly reversed.  The
effect of gravity upon the flattened portion tends to resist a further distortion of the catenary. The tensile strains in this portion increase with
the straightening betw-een;1 and C, and thus equilibrate the effects of
the load at D.




203
If we apply the illustration of the column in explaining these
phenomena, we see that if a force shall have already deflected the suspended column, the application of a load to its lower end will tend to
restore its vertical position, and will lessen the strain on any horizontal
brace or guy employed to resist the deflecting force; but if the column
be placed on end and deflected, the imposition of an additional load
upon its upper end will tend to overthrow it, and will increase the
strain on the brace or guy, resisting the deflection. These effects are
precisely the same with the arch, when suspended and upright.
If we desire to prevent any appreciable deflection of the suspended
column from its normal or vertical position, it is evident that we must
apply the same amount of horizontal bracing to resist the deflecting
force that would be required to preserve the column vertically against the
same force if the column were standing on its end, for so long as the suspended column remains vertical, the resistance due to its weight or that
of the load is nil, as before stated, and so long as its vertical position is
maintained when standing on end, its own weight cannot be added to the
deflecting force, and hence it will need no more bracing than if suspended. It is fifter distortion has occurred that the effect of the weight
of the cable or the arch is felt; in one case to restore equilibrium, and
in the other to increase its disturbance.
For railroad purposes, as well as to insure durability in the structure,
this distortion, should be prevented as far as possible, and I have no
doubt with a llproper degree of stiffness in the roadway of both systems,
the bracing of the upright arch can be quite as economically accomplished as that of the suspension cable. Just in proportion, however,
as we permit greater undulations in the roadway of the bridge. the
ecolnomy of bracing will incline in favor of the suspension system.
Recurring to Fig. 8: if half the sp)an of the upright arch be loaded, a
horizontal impulse is given to the arch at the crown, tending to move
its central point towards the unloaded half. To prevent this horizontal
movemnent of the centre of the arch, is one of the most important
problems with which we are dealing.   If tile unloaded half be kept
from curving llupward, tlhe movement of the center horizontally cannot
occur, and then the undulation of the roadway will be reduced to the
minimum. Evidently the slightest bending of the girder will result in
a horizontal movement of the centre of the arch and cause a wave in
the roadwaly, hencle depth of girder is an important element in stiffening
the arch.  This movement naturally suggests as an expedient that the




204
ends of the lower chord of the girder be firmly fixed to the masonry, by
which means one-half of the girder would resist this movement by tension, and the other half by compression, which would certainly reduce
the wave.  But owing to the effects of temperature, the ends of the
girder, at the abutments, must be left free to move horizontally, or
they will push or pull the masonry to its injury or destruction.
Another effect will be noticed as a result of the load being on half of
the span; see Fig. 8. The top chord of the loaded girder, as well as the
arch itself, is in compression whilst the lower chord acts in tension to
resist the thrust from the upper chord; therefore, on this side we
have two longitudinal compressive members, and as these strains are
reversed in the unloaded half, we find two similar tension members
there. It is evident that if the struts were run up to the upper chord
and tension diagonals were introduced from the top of the first strut at
the chord, to the bottom of the second one at the arch, and so on to the
centre, the arch would supply the tensional resistance for the upper
chord, and the lower one could be dispensed with in the loaded side and
the tension chord in the other half. This modification simply substitutes
spandrel bracing for the girder, and when jointed at the crown, forlms.
the system used in the jointed arch bridge at Szegedin. From an examination of Fig. 10, which is a skeleton drawing of that bridge, the
strains of the various members of one of its ribs will be seen, as given by
M. Ritter, from whose work the drawing is copied.
The resistance to deflection by this and the preceding form  of
bracing would evidently be increased by greater depth of girder. In
the Szegedin bridge the large sections of the braces and chords near
the centre of the structure are due mainly to the shallow depth of the
rib, while this depth has evidently been kept shallow, because of the increased length required for the struts and braces. These being alternately in compression as well as tension, are by economy limited in length.
The span of the bridge is but 135 feet. In a structure of 400 or 500 feet
length of span these members would be of great length, and hence
the advantage of depth would be lost by its greater proportionate cost.
The half girder system of stiffening the arch will be found more
economic than any method of spandrel bracing, because vertical members
only are required between the arch and half girder, whilst greater proportionate depth is practicable without involving such long compression
members.   But for another important reason, greater economy and
greater rigidity can be attained by it than by spandrel bracing. In resist



205
ing the horizontal movement at the central joint by spandrel bracing, the
diagram of the Szegedin bridge shows that the chord over the unloaded
half arch acts wholly in tension.  Its large sections in the central panels
of the half rib, being about one-third of those of the arch, indicate the
great pull brought by it upon the unloaded half arch, while stiffening it
to resist the buckling impulse from the loaded one. This pull increases
the compression in the unloaded arch, and thus increases the tendency
in the central joint to move towards the unloaded side of the structure.
When it is remembered that the deflection of the loaded half depends
almost wholly on the horizontal movement of the central joint, it will be
evident that any system of bracing which, while tending to stiffen the
arch increases the impulse to move its centre in the direction of the unloaded half span, must be in conflict with economy, at least in theory, if
not in practice. As the loaded half produces compressive strains at the
same time in the chord over it, and these by the braces at the central
joint also tend to increase its horizontal movement in the same direction, resistance to these strains can only be had by increased material.
These objections do not exist in the girder system, whilst all spandrel
braced arches are open to them, whether jointed or not.  Therefore
spandrel bracing will invariably prove more expensive, theoretically,
than girder bracing, for the arch.
If the chords in the spandrel bracing referred to, be placed beneath instead of over the arch, as in Fig. 11, these objections vanish, because the
strains are reversed. Under the loaded half they become tensile strains
and resist the horizontal movement at the central joint by pulling directly against it, and this serves to prevent the spread or flattening of
that half. By a system of struts, between this member and the arch
above it, the chord becomes virtually a suspension cable and acts in
unison with the arch to sustain the load, in the manner of the Saltash
girder.
This brings us at once to the most economic solution of the problem,
of preventing the horizontal movement of the centre of the arch, and as
a sequence, to the most economic system of superstructure that is
possible.
If we examine the effect of this arrangement on the unloaded side of
the span, we find the chord here has become a compressive member and
also resists the horizontal force at the crown. The intermediate strut
work acts in tension and prevents the rising of the unloaded arch, as well
as the fall of the chord. Exactly such strains in tension as are borne by




206
the loaded chord will be the sum of those borne by the unloaded one in
compression.   On the unloaded side, the arch and its chord or counterarch act simply as a strut to transmit the horizontal and vertical forces at
the central joint directly to the abutment at the unloaded side. The
necessity of preventing, as far as possible by any method of bracing the
arch, any horizontal movement at the crown, should not be lost sight of,
as it is of the first importance in insuring rigidity.  By no method can
this be so economically accomplished, as by the counter or inverted
arches shown in Figs. 14 and 15, which give also diagrams of strains of
a 500 foot span on this plan.
By reference to these diagrams it will be seen that a notable uniformity
of section is obtained in all of the members of the structure, a circumstance very favorable in construction.  Perhaps the most remarkable result, however, that is developed by the diagrams will be the fact that with
a steel arch, with maximum    compressive strains of 20,000 pounds per
square inch and all the other members of wrought iron, with 10,000
pounds maximum strains per square inch, and with 10 per cent. added
for joints, it is practicable to sustain on a 500 feet span, a moving load of
2,500 pounds per linear foot by a superstructure weighing (including
rails, floor bed and everytlhing) only 1,500 pounds per foot.'"
This remarkable result is due1. To the fact that every strain fro(m temperature is completely eliminatel.
2. By combining tie arch and cable, great depth of girder or bracing
is obtained.
3. No struts longer than one-half the versed sine of the arch are requred and but few that long.
4. No great strains, such as are incident in almost every other system,
are thrown on the struts.    When the roadway is suspended beneath
the arch as in the ease of the diagrams of strains (Figs. 14 and 15), the
compression oi the struts is much reduced.
5. Long struts being unnecessary, a greater versed sine or more economic depth of arch is practicable.
6. The least disturbance of equilibriumn brings every member of the
structure into play to resist it.
* The correctness of this statement has been ascertained by carefully estimating the weights
of the span from detail designs, and calculating the strains induced by the weight of the
structure and its variously proportioned loads. The result given in Figs. 14 and 15, as well as
the requisite sizes and weights of road-bed, rails, &c., have been all carefully verified by my
friend Col. Henry Flad, and may be accepted with confidence.




207
I will not stop to compare the relative economy of this method of
stiffening an arch, with that of any girder system as at present applied
to suspension bridges, but will call attention to some of the facts disclosed
by the diagram.
To span an opening 500 feet long with a bowstring girder, and attain
equal stiffness would involve, as is shown by Mr. Barlow, the use of four
times as much metal as the arch would require. The average strains in
the arch in the diagram equal 1060.6 tons of 2,000 pounds, hence a bowstring girder would involve four times as much, and would require an
average sectional area in its two members, equal to the resistance of
4,242.4 tons. To stiffen the arch in the diagram, an average section in the
counter-arch is required equal to the resistance of 184.8 tons, or about,t part of the weight of the bowstring girder. Mr. Barlow obtained by
his experiments with models a result of,, as the weight of his stiffening
girder.  The similarity of his experimental results, with those of the
careful scientific deductions shown in the diagram, seem to me quite remarkable. It will be observed that his proposition as to the necessity of
quacdruplicating the weight of the girder to get the stiffness of the arch
relates to the sustaining members only, and omits the bracing or web
needed in all girders between the upper and lower chords. In his experiments this was of course included.
Respecting the relative strength required in a stiffening girder for a
suspension bridge, as compared with the strength of a girder for the
same span suited to bear an uniform load of the same intensity, Rankin
says (in a note, pag'e 375, "'Applied Mechanics," 5th edition): "Hence,
it appears that if the chain be supposed inextensible, the proportion
borne by the strength of the stiffening girder to that of a simple girder
of the same span, suited to bear an uniform load of the same intensity
with the traveling load, ought to be as 0.138 to 1; while if the chain is
supposed very extensible, as in the approximate solution, that it is found
to be.A or 0.148 to 1; so that in the intermediate cases that occur in practice, no material error will be committed if that proportion be made as
1 to 7 or as 0.1.43 to 1."
The wide difference between the proportion of., obtained experimentally by MIr. Barlow, and the I mathematically deduced by Prof.
Rankin, arises chiefly from the fact that the one relates to stiffness, while
the other refers to strength.  Mr. Barlow's illustration, shows that
with double the metal and the same load, the deflection of the bowstring
will be twice as great as that of the arch, but as the strains would not
b)e increased thereby, it follows that double the metal in the girder should
(theoretically) give the same strength of the arch, whilst four times the




208
metal is required to give the same stiffness. After making this allowance,
however, the proportions of Piof. Itankin are nearly twice as great as
those of Mr. Barlow. This may arise from difference in the assumed
proportions of depth and length of girder and chain on which the Professor's calculations were based, and those of Mr. Barlow's models.
Leaving this discrepancy to be explained by others, I will point to the
fact that by the diagram the average strains in the counter arches are only
184.8 tons of 2,000 pounds, while those in the arch are 1060.6, or 5; times
as great. Hence, as it requires at least twice the metal of the arch to
convert it into a bowstring girder of equal strength, it will be seen that
the girder would weigh eleven and a half times as much as the counter
arches by \which the l)roposed arch is stiffened, or 65 per cent. less than the
weight of stiffening girder required for the suspension bridge according
to Rankin. While the stiffness of the arch over that of the bowstring
girder is maintained under the whole load, the form and depth of the
counter arches give far greater resistance to undulation under the movement of lartial loads, than is possible by any practicable depth of stiffening girder yet proposed for a suspension bridge.
I am confident that a careful investigation of the system suggested
and the facts stated, will convince those interested, that it is entirely practicable to brace the upright arch more effectually and with equal, if not
greater economy, thlan is possible by any known method of stiffening
suspension bridges; and that the proposed system avoids all the disadvantages resulting from temperature.  These two difficulties have hitherto
prevented the most lerfect economy of superstructure from being attained.
By overcoming them, the cost for long spans is wonderfully reduced,
compared with the most economic truss systems yet devised. By any
method of girder construction hitherto known, it is impossible to span a
clear opening of 500 feet,'with less than three times the dead weight of
the arch on the proposed system, with equal strength of girder and with
the same material and allowable strain. More than twice the quantity
would be requisite in any case, but when the span becomes so great, a
less economic depth of truss must be taken, and the length of truss must
considerably exceed that of the arch, because the girder must rest upon
the masonry, whilst the arch rests against it. In addition to this great
excess in its cost, the girder will have twice the deflection of the arch
under equal loads.
The arch, as hitherto constructed, being still much cheaper for the
superstructure, it is evident that a great saving in the substructure must




209
have existed in the girder systems, to enable them to be introduced
during the last thirty years in all parts of the world, almost to the exclusion of the arch.  The difficulty of stiffening the arch, and the inconvenient effects of temperature, together with the greater cost of masonry,
have given the different girder systems a degree of public favor which
must disappear when these objections to the arch are removed.
Having shown how the cost of the arch in superstructure may be brought
to the lowest possible point, by economic and effective bracing and by
the avoidance of the effects of temperature, I will proceed to suggest
such methods as will in almost every case, render in combination with
the arch the economy of the girder slbslrucltu'e available. Evidently if
the cheapest possible form of superstructure can be combined with the
cheapest methods of substructure, we shall have attained the most
economic system of construction that is possible.
The greater masonry required for the arch arises solely from the
horizontal forces resulting from the weight of the arched span and its
load, and from those which are induced by temperature. By reducing
the weight of bracing to a minimum, and by eliminating the strains of
temperature, we not only arrive at the greatest possible economy of
superstructure, but by thus lessening the horizontal pressure upon the
masonry to the lowest possible point, we also reduce the cost of the substructure to a minimum, so far as the arch per se is concerned. Therefore, to reduce still more the cost of masonry and approximate, if not
equal, the economy of girder bridges in this item, we start from an
already advanced point in our problem.
Where timber is abundant, an economic method of saving masonry in
the piers of arched bridges may be employed where the bridge consists
of two or more arches, by introducing wooden chords in them against
the skewbacks or piers and abutments. These wooden chords would
act in compression only and form a series of compri'ession members instead of a line of tension> members or chords from abutment to abutment,
as in the case of bowstring girders. They need not, however, be in
compression unless the bridge is loaded. If there be a series of long
spans together, however, the loading of an arch at one end of the series
would produce compression throughout the entire line of chords in the
other arches, and this might shorten those chords so much in the aggregate as possibly to allow the loaded arch to spread too much, and thus
produce objectionable deflection in the roadway of that arch. In this
case it would be desirable to make the abutments stronger and put an




210
initial compressive strain in all of the chords of the system by means of
screws or wedges against one end of each line of chord timbers.
In a series, for instance, of five arches of 500 feet span each, where the
maximum horizontal force of the load is 500 tons on the chords, if an
initial compressive strain of 400 tons be produced in the entire system,
from abutment to abutment, when the bridge is unloaded, then this
initial strain will be taken out of the chords of the first arch so soon as
it has its maximum load on it, while the compression in those of the
unloaded ones will be only increased 100 tons, and therefore the
shortening of those chords would only be one-fifth part as great as if
they had no initial colmpression; hence the deflection by load on any
one arch in the system would be reduced accordingly, and would be
really less than what it would be in an ordinary bowstring girder. In
this case, the maximum stress on the abutments, when all the arches are
loaded, would be 500 tons load + 400 tons initial compression - (the
force from unloaded arches say) 300 tons, making a total of 1,200 tons;
while the piers would be subjected to vertical pressure only, and hence
they could be as cheap as if for ordinary girders.
If the timbers were secured together to resist tension, of course the
compressive strains would be so much lessened, and the abutments proportionately reduced in cost. With such a system of wooden chords used
only in compression, the repairs of the timber would be very simple and
easy. The sticks should be squared at each end and butted against each
other throughout the span, vertical movement being prevented by the
connections of the floor beams with the arch, and lateral movement, by
the wind bracing of the floor. To remove any defective stick it would
only be necessary to withdraw the wedge, or slack the screw at the end
of the line in which it was located, and by which the initial compression
was created, and every piece in that line would then be released and any
one easily removed.
The initial compression could be so great that no tension could be
produced in either chord by wind pressure, and hence no jointing of the
sticks together would be necessary to resist wind.
The track stringers and every longitudinal timber in the floor-way
could be tlhus utilized to resist the thrust of the arches, and in this way,
where timber is cheap, a very durable and economic structure can be
erected. As no thrust in such a bridge can come on anything but the
abutments, and as these can generally be located on the high banks of
the stream, the cost of the entire substructure would exceed very slightly
that which would be required in a truss bridge with spans of equal




211
length. It would probably be best to make the arches of such a series,
uniform. The thrust at the abutments will be the same whether there
be but one or many arches in the system. The stress on the chords
(except initial) would be due entirely to the unequal loading of the
various arches.
This method is applicable to parallel truss bridges, and by it the iron
lower chord or tension member may be omitted and wooden compression
members substituted therefor. The objection to the combination of
wood and iron in bridge construction, owing to the difficulty of repairing
the bridge, does not exist in this method. In all others, the wood is
either under tension or compression and tlerefore difficult to be removed.
In this, the entire chords of any one arch could be removed without endangering the stability of that arch or of any other one of the series;
for it is plain, that if any temporary weight were placed on the floor
beams, which would equal the weight of the chords to be removed, the
equilibrium of the whole series would be undisturbed by their removal,
so long as the whole bridge remained unloaded. In repairing it would
never be necessary, however, to remove any one chord entire at once,
but only to replace such pieces as were found defective.
Where the use of timber is found objectionable the cost of masonry
in arched bridges may be greatly reduced by the following method also.
Suppose A B, Fig. 2, be a bowstring girder, whose arch produces a
horizontal force of 500 tons; and that the load increases this force to
1,000 tons. The iron chords of such a girder will require a sectional area
of 200 inches if we allow a strain of 5 tons per square inch. Suppose two
such girders, Fig. 11, constitute a bridge with abutments capable of resisting a horizontal force of 1,000 tons, and that each arch abuts against
the other at the central pier, then it is evident that the section of the
chords may be at once reduced one-half or to 100 square inches; for the
thrust of one arch will balance that of the other, and the only strain which
can come on the chords of either will be from the effect of the load, and
this cannot exceed 500 tons.  No horizontal strain can come on the
central pier, for any unbalanced thrust will be borne by the chords.
If both arches be equally loaded, or both be unloaded, no strain whatever will be on the chords of either.
If the arch A has its maximum load imposed, its chords will be
strained in tension to 500 tons, but if the chords of both arches be
formed to resist compression as well as tension, 250 tons of this strain
will be transferred in compression to the chords of B, and by it to the
16




212
abutment of B, while the tension in the chord of the loaded arch will be
reduced to 250 tons. Instead of a central pier which would have to sustain
a horizontal force of 500 tons if no chords were used, we now have a pier
subjected to vertical pressure only; and instead of bowstring girders, with
chords of 200 square inch section, we need only one-fourth of that. As
the chords of a bow-string girder are in great part supported by the
arches, if we can reduce the chords 75 per cent., we at the same time
lessen the requisite section of the arch itself, by relieving it of this
much dead weight.
The only difficulty in carrying out this suggestion is the expansion
and contraction of the chords by heat and cold. Suppose that to the
end A of the lever A B, Fig. 12, we attach one end of the chord C of
one arch, and to the end B we attach one end of the chord D of the
other arch, and that the fulcrum E be secured to the central pier, whilst
the other ends of the chords are attached to the abutment ends of the
two arches. Now, if the inner ends of the arches be made to abut against
the fulcrum of the lever, it is evident that the two chords may expand or
contract without moving the central pier, or in any manner disturbing
the arches. Such expansion or contraction of the chords will simply
cause the lever to turn upon its fulcrum. A force against the fulcrum
from either side of it will not, however, tend to turn the lever at all.
Such force would be resisted equally by both chords, one in tension and
the other in compression. One half of the force-500 tons due to the
load on one arch-acting in the direction of the arrow, would be received
on C in compression, and one-half on D in tension, whilst the only
movement of the lever or fulcrum would be that due to the stress on C
and the tension on D. The lever could be made in the form of two
circular disks fitted in circular rests in the skew-backs, the axes of
the disks being coincident. The end of the chord from one arch would
be fastened between these disks by a suitable pin passed through that,
chord and the upper part of the disks, and the other chord would
be fastened b)y a similar pin through it and the lower part of the
disks.  The effect of heat or cold on the chords would simply cause
a rotation of the disks in their rests without strain and with no horizontal movement of them or the skew-backs.   This lever arrangement
is illustrated by Fig. 13, wlhich shows a skew-back in perspective,
seated  oln a series of rollers resting on the cap-plates of the
pier. The object of the rollers is to avoid any possibility of horizontal
strain on the pier, arising from the extension of one set of chords and




213
the compression of the others in two adjacent spans, when the load on
one span is greater than that on the other. This movement will be so
slight, however, il spans of ordinary length, that if the central pier were
proportionately high, the rollers would be unnecessary.
The chords of the arches are, in Fig. 13, attached to the disks by
links pivoted to them. One end of a link is inserted between the two
disks, and a pin is passed through its end and through the two disks
above the centre of the disks, while the link from the other chord is
similarly secured by a pin passed through the disks and link below the
centre of the disks. The other ends of the links are pivoted to the
adjacent chords of the two arches. This, in effect, constitutes a vertical
lever attachment for the ends of the chords, of greatly stronger design
and more compact than would be )po.sibll with a vertical lever having a
central fulcrum pin through it. The links must, of course, bear compression as well as tension.
This plan of disks is best used in two-span designs, but can be used
witl great economy in those of but one, or of more than two spans.
When both arches of a two-span bridge are loaded, the abutments
will of course receive as much thrust as with an ordinary arch, for the
chords are only strained when a greater load is on one arch than the
other. But the central pier has nothing but vertical pressure on it, and
may be greatly reduced in cost, while the chords are only one-quarter the
weight that would be needed in ordinary bowstring girders. If four spans
be desired, the central pier need then only be strong enough to resist the
thrust due to unequal loads on the arches, as the thrust from the dead
weight of the arch on one side of it will be balanced by that of the one
on the other side of t.
In the case of a single arch, let us suppose that one end rests upon
one abutment capable of holding but lhalf the thrust of the arch, and at
the other abutment we locate a skew-back, designed substantially as in
Fig. 13,. Let a chord be secured to the end of the arch, at the first
abutment, and to the inner link at tle other albutment. Now, if'the end
of the second link be secured in this albutment so as to resist compression, it will throw against this abutment one-half of the thrust of the
arcl, whlile the chord will receive the other half in tension.  By
such a modification one-half the weight of the chord of an ordinary
bowstring girder and one-half of the masonry that an ordinary arch
would require will suffice.  Thle skew-back with the disks in it would
have to move on rollers in this case, and would make one-half of such
movement as would be due to the elongation of the cihord from  )temp'er



214
ature and from the load also. By very simple modifications of the system
this movement could be divided so as to occur equally at both abutments. This movement would, however, be no more objectionable than
that of the ordinary iron girders on their piers, while the deflection, by
preventing the arch from spreading, would be 25 per cent. less with a
single arch, and 50 per cent. less if two arches are used, than with any
girder bridge of equal depth and strength now known.
By combining with this modification of the bowstring girder, the
counter-arch method of bracing the arch, before suggested, and with
central and abutment joints in the arch, we have a system of bridge
construction from which the effects of temperature are absolutely eliminated, and which will be found to greatly surpass in economy of superstructure anything yet devised, and which admits of such reduced cost
of substructure as to almost, if not quite, equal that applied to the
various kinds of girder bridges.
The construction of the arches in half-length ribs, with the counterarch bracing, enables the ribs to be easily erected, even if the spans
be enormous in length. Each segment or half of a rib could be easily
erected by a temporary tower placed in the stream, midway in the span,
either on floats or piles, to support the inner ends of the segments, and
on this, and on the abutments, should be placed machinery sufficient to
lift a segment. When it is understood that such a segment, with its
counter-arch and bracing for a span of 500 feet to carry 2,500 pounds
per linear foot, weighs less than 50 tons, if the arch be of steel, and that
the hoisting machinery needs only to lift each end of such a segment, or
25 tons, the ease with which such arches can be put together will be at
once manifest.
No suspension bridge system, yet devised, possesses anything like the
resistance to change of form which this does, owing to its great depth of
bracing; while for equal length of span it possesses greater economy.
The catenary to span the same opening, must be longer than the arch, by
the diameter of one tower, as it extends from centre to centre of towers,
while the arch will spring from the faces of the masonry. This advantage possessed by the arch, will quite compensate for the joints required
in its construction, and which are not needed in wire cables.
Where great height of span is desirable, the cost of masonry to hold
an ordinary arch may become so great as to forbid availing of its unquestioned economy. In such cases the chord with its compensating lever
or disk attachment, at one or both ends of the arch, will relieve the piers
of any desired amount of the thrust, just in proportion to the relative




215
lengths of the lever arins, or relative distances from tie centre of tile
disks, at \which the pins are located.
It is, how-ever, when we compare this system with other parts of the
suspension system, that its great economy over that system is seen.
Whlen we leave the great central span, the chief feature of nearly all suspension lridlges, and examine the large sections that must be maintained
in its cables over the reduced spans between the towers and anchorages,
and compute the necessary weight and great cost of the masonry required
to resist the tension of the cables, and compare these features with the
shorter antd more economic spans and light piers which this system admits of in the approaches, we perceive its remarkable economy over the
best suspension system yet devised.    To the economy and rigidity
secured by tlhe systelm prol0sed, we must add in its favor also the important elements of safety and durability, which are secured in a higher
degree by using the material of the chief sustaining member of the
structure in compression, instead of tension.
There is no limit except a financial one, to the length of span which
may be safely constructed by this system, and spans of 1,500 or 2,000
feet will be found to be entirely within a practical or profitable limit of
expenditure.
Figs. 16, 17, 18, 19 and 20 represent a bridge with two spans of 400 feet
each, on the proposed system, for a double-track railroad.  Table I contains a sumlmary of quantities and weights of materials for its construction. Table II contains the quantities and weights of a bridge with five
spans of 250 feet each, all the piers being of masonry. Figs. 21, 22, 23
and 24 illustrate the five-span bridge. Both tables have been carefully
prepared from complete designs for both bridges. (See pages 235, 236.)
Mn. CHARLES MACDONALD-I would not presume to discuss the paper
just read without opportunity for full consideration, nor will I at this
time attempt an examination of the argument presented in favor of the
economy of the arch over thle suspension system.
We, as engineers simply, too often loose sight in bridge construction of a very important element, the cost per pound of the material
and the facility for erection, while instead we are consideringalso an important element-the reduction of weight.       Those who
examine tlhe St. Louis Bridge, must be impressed with the great
accuracy in workmalnship required to fully carry out the theory of
the design. The most expensive material was used, to withstand the
strains which was said in Capt. Eads' first report to exist in the membels.  All tltis increased the cost per pound of the structure, and I




216
doubt whether in tile construction of another bridge, he will again
adopt this design for the purpose of saving a few pounds of metal.
In my judgment, and from a practical point of view, great increase in
lengths of spans generally seems unnecessary. In most cases the problem
of covering a given space is best solved, and with greatest economy,
when the cost of superstructure about equals that of its supports. I
will refer to a recent instance; bids were asked for a bridge on the
Cincinnati Southern Railway, over the Kentucky River, where the depth
from the rails to the water was 275 feet, and the opening was 1,225 feet.
Ten or twelve proposals were received, only one of which was for a
suspension bridge, and that was the most expensive offered; the others
were for spans varying in length up to 270 feet only. In this case the
engineers who bid, did so with reference to cost, and without regard
to "glory "; each in designing his structure endeavored to get the most
economical arrangement of parts and length of span.
More attention should be given to the cost per pound, and a
preference to shapes of iron most easily and cheaply made, and which
are most economical in the end.  There may be a little more or less
difference in weight in consequence, but where the cost of the whole
bridge is taken into account, it is hardly necessary to make so nice calculations as have been just read.
MR. ASHBELn WELCH-I wish to endorse as strongly as possible what
Mr. Macdonald has said about cost of structures, and to remind this
body of Civil Engineers of a cardinal principle to be regarded always in
practice-that engineering is best which most fully answers its purpose
at the least cost.
MR. J. DUTTON STEELE-I will nention a historical anecdote relative
to the Suspension Bridge built by Telford over the Straits of Menai,
which will be found recorded in the history of that structure. It was his
intention to construct a cast-iron arch, such as he had put up in many
places in Great Britain; to support the voztssiors during erection, lie
designed a suspension system, similar to that employed at St. Louis but
more elaborate, which when complete, was found sufficient in itself
without the arch, hence he abandoned his first design and put up a
suspension bridge.
MR. WILLARD S. POPE. *-In this paper is stated the broad and sweeping proposition, that " Upright arched bridges can be more economically
constructed than is possible by any other method whatever, no matter
* This and the following discussions were presented at the meeting of the Society, held
September 16th, 1874.




217
what length of span may be required;" the general reasons are given on
which this opinion is based, and in explanation and corroboration thereof are appended diagrams and estimates of materials in upright arched
bridges of various spans, in accordance with the designs.
This proposition is a startling one. During the past quarter of a
century the necessities of commerce have stimulated bridge construction
utterly beyond precedent, and consequently great and earnest attention
has been paid to this specialty of engineering. All over the civilized
world engineers have devoted their best capacities to the study of this
subject, and the concurrent action of almost all has been in one direction. Without consultation with each other, and certainly without undue
bias, reasoning out their conclusions from different data and by different processes, they have, with surprising unanimity, departed from the
simple arch and gravitated toward other forms of structure. Of the
multitudinous bridges for railroad purposes that have been constructed
in metal during the past twenty years, those that are of the pure arch
form are insignificant in number. The bridge at St. Louis is almost the
sole modern example.
If the proposition so loldly affirmed is correct, it shows that the
general tendency of modern bridge engineers is in the wrong direction;
that they have been misled and are all astray. It is surely proper, therefore, to examine with care the reasons on which this revolutionizing
statement is based.
Action and reaction are equal. In a structure of the kind known
distinctively as the truss, the compressive action of one member is met
by the tensile reaction of its corresponding mate, and so the superstructure is of itself in equilibrium. In the suspended or the upright
arch, the direct reaction is taken by the substructure, which must be enlarged accordingly. Therefore, both substructure and superstructure
must be considered together in making any comparison between the
economic merits of different forms of construction.
Here let me commend the characteristic fairness with which the
writer has met the case. He recognizes the fact just stated, and very
properly considers the entire bridge as a unit. Whatever may thought
of his conclusions and the reasoning on which they are based, he certainly confronts the problem boldly and squarely.
First, as to the difference between a suspended and an upright arch.
Here, as the reaction in both cases is borne by the substructure, that
element of cost may, for the purposes of the comparison, be disregarded.




218
In the suspended arch, the cables which carry the structure must be
extended full size over and beyond the supporting towers to the anchorage behind them. Thus the actual length of the bridge is considerably
greater than the effective length. This item of cost is saved in the upright arch, and so far, therefore, the comparison is in its favor. In
many cases, however, short spans are required at each end of the main
span, inl which case the extended cables answer for such support, so that
such extension is not necessarily a total loss.
But the cables or chains of the catenary are in tension, while the ribs
of the upright r      ar ae in compression.  It is doubtless the fact, as
stated in this paper, that iron or steel can be trusted with equal strains
for equal sectional areas, whether in compression or tension, although
the almost universal practice is to tax the metal less with the former than
with the latter duty. But in compression the ratio between length and
diameter is an essential element, while in tension this is not considered.
To get the full value of the compressive capacity of the materials named,
the length of the member must not exceed ten or twelve times its diameter. This at once introduces the necessity of bracing the upright
arch to secure its permanence, a necessity which does not at all apply to
the suspended arch.
That we may get a clear insight into this matter, let us leave entirely
out of view for the present the lateral strains induced by wind or by
moving loads, inasmuch as these are common to both forms of structure,
and material must be provided to resist them alike in each. Let us also
disregard for the present the necessity (also common to each) of providing for a variable load, and assume that the weight to be carried is definite and uniformly distributed.  The data being the same, the strains are
alike in each, only of an opposite character; and it may be granted for
the sake of the argument that the net sectional area of metal and consequlently the net weights are also alike.
What, then, are the conditions respectively? Each single cable of the
catenary is complete in itself. To perform its functions it needs no other
or auxiliary connections. It is entirely independent, self-supporting,
and carries its load without outside help. It is of simple construction,
compact, solid, without objectionable joints, and contains within itself
no element of destruction.  It is easily and safely erected in place. Application of reasonable load, and indeed its own weight, serves only to
increase its stability. All the forces of nature act with it, and it is in
entire harmony wittl the immLutable laws that govern all nmtter.




219
In the upright arch, most of these conditions are reversed.   Each rib
is by itself utterly helpless, even to carry its own weight, and must,
therefore, be connected with othela ribs by an elaborate and expensive
system of lateral bracing.   Its construction must be tubular, or rectangular, or of some similar form    suitable to resist flexure, and all such
forms are expensive.    It abounds in joints, each one of which must be
carefully and perfectly fitted.  The lateral bracing must be so complete
as to thoroughly secure the various ribs at intervals of not greater than ten
or twelve diameters.   To erect it in place is a difficult and dangerous
task.  The centre of gravity is high relatively to its base, and, consequently, the forces of nature act ag'ainst rather than with it.  No human
work can be perfect, and defects of workmanship, that would be almost
immaterial in the catenary, become serious when in the upright arch.
It is impossible, even if it were desirable, to make a structure which will
be ablsolutely rigid; oscillations and undulations that would be harmless
in the catenary might prove ultimately fatal to its rival.
These are some of the inherent and broad differences between the
two forms of construction.   Other matters, such as provision for variable
load, for side strains from wind, c&., &c., are common to both, and so
need not enter into the present computation.      And, in the opinion of
modern engineers, it is these differences that have so generally influenced
the substitution of other forms in place of the arch.
Indeed, the fact is not to be disguised that, on general principles,
most engineers prefer to avoid compressive strains in metals as much as
possible.   It is easy to know perfectly our material and all its capacities
in tension; but, in the lpresent state of our knowledge so many elements
of uncertainty enter into the use of iron in compression, that such use
must l)e attended with much risk.   It is one of those subjects concerning
which we feel that a little knowledge is a dangerous thing, and prudent
engineers are careful to err lesignedly on the side of safety. *
If an upright larch is to be used at all, the suggestion of a counterarch for variable loads has much merit.     The difficulties that were encountered in the designs of the St. Louis arch-bridge, from     the effects
of chlllanges of temperature, are fully and forcibly detailed.     In this
* The great desirability of more information in regard to the practical compressive capacity
of metals is illustrated by the action of the Society in appointing a committee (of which
Capt. Eads himself is a member) to procure the assistance and co-operation of the Government in lmaking a thorough and elaborate series of tests and experiments on the capacities of
metals. No more important subject than this can engage the attention of the Society and its
nmembers.




220
structure, the two memlers composing one rib are intended to act
together, each bearing its proportion of compression, and at medium
temperatures they do so act, but in extreme cold weather the rib deflects,
changing of course its curvature, and then at the crown the upper
member and at the springing the lower member, does all the duty,
which duty is still further increased by an actual tensile strain brought
upon the opposite member at those points. In extreme hot weather this
action is reversed. Thus the novel experience is had, of certain portions
of an arch subjected to tensile duty. This tension was found so considerable as to render it necessary to firmly bolt the skew-backs to the
masonry. Doubtless the counter-arch jointed at tihe crown and at the
skew-back would have obviated the difficulty, and indeed, it would seem
that, even with the present construction, if the two members composing
the rib had been brought together at the springilng and at the crown, the
trouble would not have occurred.   As the bridge is designed, it appears impossible to tell how much duty each skew-back is at any time
actually performing; just as it is impossible to tell precisely how much
weight comes upon each leg of a four-legged table.
Accompanying Capt. Eads' paper are plans and estimates for a
bridge on his design, with a span of 400 feet, and a rise of 50 feet, the
floor being suspended below the arches flush with the springing. A
thorough system of lateral and diagonal bracing between the respective
arches and counter-arches is shown, so thorough indeed, as to quite
block up the roadway, the lateral strut between the coulter-arches nearest
the springing being only about 6 feet above the floor, while the lateral
ties seem to extend quite down to the floor. -
But assuming that a clear headway of 20 feet above the floor is needed
for the passage of trains, no lateral strut or tie could be placed between
the main arches for a distance of 40 feet from the springing, or between
tile counter-arches for a distance of 120 feet from the springing. For
these distances therefore, the arch and counter-arch could only be stiffened laterally by knee braces from the floor; inasmuch as the entire
floor is simply suspended from the arches this would not prove a very
reliable security, and for just this locality where the lateral bracing is
the weakest, the need of it is tile greatest, as it is exactly here that tle
strains are the heaviest.
Indeed, the (lemands of an upright arch for continuous and effective
lateral bracing between tile ribs are so imperative and absolute that it
* This, however, is evidently an oversight of the (Iraftsman.




221
is almost a necessity that the flooring and passage way should be placed
either entirely below the springing or above the crown. The vertical
space between the springing and crown is fairly interdicted for such service, and any intrusion therein is at the expense of the stability or
economy of the structure. The grade line of the floor being generally
the governing consideration, if the arches are placed above it sufficiently
for the necessary head-room, they are raised very high, which of course
adds to masonry, and so to cost. If the arches are placed below the
grade line, they not only take up what may be valuable headway for
navigation but also necessitate a heavy and costly system of struts for
the support of the roadway on top of the haunches.
It is furthermore pertinent to consider that so far as bridges are or
may b)e built over navigable streams, an arch placed below grade lusurps
more of the head-room for vessels than does any other form of structure.
A law decreeing a certain specified clear space between high water and
the crown of the arch may prove very deceptive as a protection to the
interests of navigation.
Leaving, now, the comparison between the upright and the suspended
arch, we come to a consideration of the devices suggested to secure
economy in masonry. These are ingenious, and so far as I know, novel;
they can be best understood by a reference to the paper. That the line
of timber chords suggested for a series of arches would answer the purpose I have no doubt; to secure them firmly however, would require a
large amoumnt of bracing, both vertical and lateral. They would not be
especially ornlmental, and would be liable to decay and to fire. To decide
as to their ultimate economy wouldl require a calculation for each particular case.  I imagine that engineers generally will hardly urge their
adoption.
The other andl more elaborate plan provides for dividing the thrust
into parcels, amnd apportioning a certain definite part to the abutments and
a certain other part to a system of longitudinal tie-rods or lower chords;
in other words, the arch now becomes partly an arch, and partly a bowstring girder. In the case of two adjoining spans, if they are both empty
or both uniformly loaded, the abutments receive the entire thrust and the
chords carry nothing. But if one arch is loaded and the other empty,
then the chords are strained to the extent of one-half of the thrust from
tie live load, and the arrangement is such that of this thrust one
half is borne in tension by the chords of the loaded arch and the other
half in compression by the chords of the unloaded arch; the chords




22 2
being so constructed as to resist both tension and compression. In any
case the pier sustains nothing but vertical load. The device is ingenious in the extreme and well worth careful study.
The reflection presents itself however, that possibly there may be
some practical difficulty in harmonizing three distinct agents, viz., the
abutments in gravity, the lower chords of one arch in compression and
of the other in tension, so that each will always do exactly its allotted
duty, for if one does less, the other must do more, and thus be overburdened. It is somewhat similar in effect to a continuous bridge resting
upon several supports, so long as each does its work fully and fairly, so
long is the structure effective; but if one settles or in any. way shirks its
duty, new, unexpected and dangerous strains are at once developed in
the girder. Another pertinent illustration of the difficulties attending a
divided duty, is found in the practice formerly so common of uniting an
arch and a truss in the same bridge. Experience showed that it was
impossible to harmonize them-one did all the work, and the other,
instead of being a help, was really an incubus and a nuisance; the better
modern practice has, I believe, entirely abandoned the plan.
Even assuming that it is entirely practicable to parcel out the thrust
of the arch, as it is proposed to do, yet the plan will not be without
cost; inasmuch as a lower chord 400 feet long, to resist even limited
compression, must be well and thoroughly braced both vertically and
laterally. If it were simply suspended from the arch as shown in the
drawing, it would be as limber as a whip-lash.
After all, it is largely true that in such matters each particular locality
forms a law unto itself.  A structure that is admirably adapted for one
place may be utterly unsuitable for another, and he is the best engineer
who best fits his plans and designs to the peculiar dellands which they
are to serve, and who thoroughly accomplishes the desired purposes with
the least expenditure of money.
The entire paper is exceedingly interesting and suggestive. It is a
bold and vigorous attack upon the labors and opinions of the great majority of modern bridge engineers, and both for the brilliant professional
repute of its author, and for its own merits, it will command and receive
earnest attention.
MII. THOMiAS C. CLARKE-It must be a source of gratification to the
Society that a memler, so mmuch occupied with public business as Capt.
Eads, should find time to present such an elaborate paper as this.
During the last seven or eight years the proper construction of iron




223
bridges has been investigated with great ability.  It will be observed,
however, that almost the whole of this effort has been directed towards
one point-elucidating the action of the strains caused by a load upon
the main trusses or girders.  Very little attention has been given to the
bridge as a whole-that is regarding it as a complex piece of framing,
subject to impact, vibration and the swaying from side to side, caused
by a rapidly moving train; unless all these points are carefully attended
to, the bridge cannot be considered a perfect structure-time will develop
its defects.
It is with some hesitation that a criticism of the plan here presented
is ventured upon, from this point of view.  Assuming for the present,
that the main arches do what is here claimed for them-require less material than any other system-yet it will be seen that a platform merely
suspended from them will not possess much lateral stability, and must
sway from side to side under at rapidly passing load. In what is termed
a "through" bridge, the floor is held in place not only by its iron system
of horizontal bracing, but also by the posts and knee or angle braces
which connect it with the upper chord. Horizontal bracing alone will
not prevent lateral motion, for the lower chord being in tension will
vibrate like a stretched string.  It is necessary to connect the floor by
rigid angle bracing with the upper chord, which being in compression,
never leaves a straight line.  If this were done in the bridge proposed
by the writer, some of the economy of material claimed for it would disappear.
Moreover, practice has shown that a bowstring girder is the worst
possible form to brace against horizontal motion, on account of the very
acute angle at which the ends of the arches meet the chords. Practice
has also shown that the deflections on a bowstring girder are very great
at each end, owing to the want of height of the truss at these points.
The locomotive seems to fall upon it with a blow or shock on leaving the
pier, instead of gliding on over a curve gradually increasing toward the
centre, as in a parallel girder.
In its present shape, the bridge proposed is not well suited for a
rapidly moving load; it may possibly be modified, but this will add material again.
In the next place, the economic merit of this system does not appear
to be borne out by the figures presented. The weight of five double
track spans of 250 feet each, in iron, is estimated at 3,010,300 pounds.
The weight of five equally strong parallel girder spans would not exceed




224
3,300,000 poulnds, and on account of their greater simplicity of construction, would cost enough less per ton to more than compensate for the
difference in weight.  If we add to the plan as it now is, enough iron to
make it laterally stiff, the cost would somewhat exceed that of the parallel
girder.
But when we come to consider the quantity of masonry required by
the two systems, we see the great economy of the parallel girder.  The
nmsonry in these two piers (which, like those under a parallel girder,
receive no pressure except a vertical one) is estimated at 1,410 cubic
yards, while those two which receive the semi-thrust of the arches
require 3,630 cubic yards, or more than double as much.  It is difficult
to see wherein this lever-disk arrangement, which by the paper nnder
consideration seems to save half the iron in the chords and double the
amount of masonry, is so economical.
Finally, tle conclusion to which we arrive is this-economy in
bridge construction is to be attained by saving in the cost of piers and
abutments so as to allow of shorter spans, rather than in the cost of
iron in order to build very long spans.  There are of course, other
questions than those of economy which come into consideration, such as
exceptionally long spans required in navigation, or because good pier
foundations cannot ble obtained, but this is foreign to the subject of the
paper, which challenges the parallel girder system on the ground( of
ecoiiomi/ aloae. We say that the case has not been proven; because the
cost of the piers land abutments which should make up at least one-half
of the cost of the -whole-but which in the case of this system, would absorb considerably more than one-half-has been omitted.
MR. Louis NICKERSON —In discussing the bridge before us, it would
be well to examine somewhat the short review of arch bridges by Mr.
Barlow, which has been used as a postulate. He therei-n claims, that
for certain reasons, an arch bridge is more economical than a truss
of triangular or trapezoidal elements.  I think that this is an error.
For even as a theory, it is founded on the assumption that the deflection
of a truss of any kind is the manifest point, from which the investigation should be made.  Now the Fink bridge, though rather more costly
in original construction, than some other well-known forms, is one of
the cheapest as regards llaintenance, and one of the safest as regards
stability, that is used. Yet the Fink bridge deflects more than any other
known bridge; I think twice as much as the Warren girder improved by
Fink. But the same cause which makes its entire deflection great-the




225
advance of the stress towards the center, and the uniform flexure of the
chord from compression-entirely eliminates the reverse curve common
to other bridges, which it is miost our duty as engineers to get rid of,
and which, so far as my knowledge extends, is much worse upon arch
b)ridges, metal for metal, than upon any other. It is this partial and
wave-like deflection preceding and following the engine in its course,
which breaks the back of the bridge, destroys its fastenings and abutting surfaces, and which engineers have attacked since the substitutiwn
of more elastic material for stone. The design before us really strives
against this, with an effect which we will attempt to measure hereafter.
As regard(s practical economy, the case seems to be as follows:
place a bowstring bridge on piers and a similar simple arch between
abutments (their stiffelling webs may be sul)posed equal), then if the
chord of the bowstring costs more than the extra abtting lmasonry, the
open arch is the cheapest; if the chord cost less, then the open arch is
the dearest. This refers to the amount of materials alone.
In constructing an ordinary triangular or trapezoidal truss bridge, it
is not necessary that the handicraft be very perfect. An error of ( -inch
in length is not perceptible in practice, and as regards safety more may be
allowed with impunity.   By bringing to bearing little imperfections
it will rather improve than otherwise during a year's service, after
which they seem not to exist.  Now, to arrive at this o4 -inch costs but
little per pound of material, because it is easily approximated by ordinary
machine tools, but to finish with the same perfectness the abutting surfaces in the long and by itself flexible curved member and also the
bracing miembers of an arch bridge, requires much more care, and even
when not carried to scrupulous excess, involves a much greater working
cost per pound; whence a rectangular bridge can be built at less first
cost, and all things considered, with a really more economical result than
can an arch bridge.
If you were asked "what is the lowest priced bridge, so far as amount
of material is concerned, that can be built?" I think the unanimous answer would be " that rectangular bridge which, with the greatest height
of truss is least liable to'buckle' the same area of web "-the Whipple
bridyge, and we know (whatever faults it may have) the Whipple bridge
is not only cheaper in cost, but, metal for metal, has much less haunch
deflection than an arch bridge. Arch bridges, as hitherto constructed,
can neither be considered economical nor very rigid.
Notwithstanding his use of 3Mr. Barlow's text, the author of the




226
bridge before us realizes the difficulty of the reversed curve in the chord;
this is shown by the fact that he has made no special attempt against the
bridge deflection, but has struck strongly and well against the haunch
deflection, and whether successful or not, he deserves our praise for the
essay. Fortunately for your patience, the best way to arrive at the mleasure of this is the most simple and expeditious one.
First, in regard to stiffness, let us remember a fact too often forgotten-that no system of bracing, however complicated, can have any
other influence upon the effects and directions of the forces in a girder,
than the most simple plan that an engineer could imagine.  If the
braces be removed from the "loup" girders of this bridge and replaced with plain plates of the same resistance, the status of the girders or the bridge has not been in the least altered.  Take an ordinary
bowstring bridge of the same span and height-a similar bridge-and
replace the web of that with a corresponding plate; we could mark
upon it with chalk the exact saving of metal with but little computation,-and from this deduce the form of the two "haunch " girders.
Remember, that the rigidity or flexibility of a bridge does not at all
depend upon its height of truss, but on the depth of web wcithin the tress,
and as no form of bracing within the limits of a truss, however ingeniously conceived, can vie in depth with the web that fills the truss,
therefore none can be so stiff.  With arch and bow-string girders thenwhich are already in fault with respect to flexibility-the change is not
an improvement.
Second, in regard to strength;-this quality and that of stiffness are
fully separated in the design. As we have only been called on to treat of
one in discussing the arch arrangement, so we are almost exclusively
confined to the other when we come to examine the lower chords and
lever attachment.  That the arrangement is highly ingenious all must
at once see, yet it seems to me it must fail from the same cause that
several attempts of the same kind have failed-notably, the construction
of a trussed girder of which the beam should bear one-half the load and
the rods the other portion, or with the same idea the attempted combination of arch and truss.
For, supposing first only tensional chords, and that the loaded span
would elongate and its arch move outwards against the heel of the other,
as described in the paper, the unloaded arch would then move slightly
inwards, thereby being relatively weaker to resist thrust than before,
and an additional strain would thus be thrown upon the chord of the




227
loaded truss. Then this must further elongate, the loaded arch spread,
and the unloaded arch be again forced inwards, becoming still less
capable to resist thrust; so repeating until ihe security bein sufficient,
the engine would cross the first arch, strike the second and cause it to
recoil with a jerk; the same action to be partially reversed upon the
other when the load had entirely passed onwards. Safety then would
probably depend upon there being sufficient seceurit/ to allow the chords
to resist the frll horizontal strain.
But supposing the chords capable for compression (and they would
then probably cost as much as full-sized tensional chords), we must
remember the inequality between the elongation and depression of material within elastic limits, and even at low strains, so that to shorten one
chord a certain distance, and to lengthen the other the same distance (a
sine qtac non to the attempt), would require very different amounts of
force (per square inch), a fact at once seen to be inadmissible with equal
chords. Hence, at no moment could there be a mutual assistance.
In the two cases mentioned, the trussed girder and the arch and truss,
the first was too obviously in error to hold a place in mechanics long, and
in regard to the second, it has been found necessary and more economical, to make one or the other member sustain the load, because by no
combination would the materials act together in such unison as was required. It seems to me, with all diffidence, that in the case before us,
either a sufficient chord or a sufficient abutment is necessary for safety.
and economy.
MR. SAMUEL H. SHREVE-The paper before the Society appears to
be an acknowledgment of " the disadvantages of the system adopted
for the superstructure of the St. Louis Bridge," which the writer says
were forcibly thrust upon his attention in the progress of that work.
He says of the bridge: "If the effects of temperature could be
avoided, and the curve were a parabola, an equally distributed load on
a rib would, of course, be borne by the upper and lower tubes equally,
that is half on each"-but the effects of the temperature are inevitable,
and consequently the above supposition, if correct (which it is not), is
without value. He further says: "This extension (from the temperature)
causes the crown to rise, which relieves the lower tube of compression at
the abutments, and hence that tube does not then support any portion of
the weight of the rib or the load.  - *  At the crown, the strains are
likewise changed.  There, however, the loirer tube does all the duty, as
the upward bending of the rib relaxes the compression in the upper tube
at tiais point."
1t




228
Suppose the crown to rise, the lower tube if relieved of compression
at the abtments, would not be strained at any point. Bracing beneath a
parabolic arch unformly loaded can receive no strain and is useless, and
if the curve of the tube considered is not a parabola, it is so near it, that
the strains which are transferred by the braces to the lower tube are too
small to be noticed.  The conclusion is unavoidable, that the upper tube
at times must bear the whole load, and even more, for the tubes "are
rigidly held to the abutments by anchor bolts," and a large extra amount
of strain is in this manner brought upon the loaded tube. Now consider
the opposite condition: "As the ribs are formed for a medium temperature, contraction by cold reverses these strains-the crown of the arch
falls," and the lower tube must just as surely take the whole load and the
upper one be useless.  Suppose a still different case; when the tubes are
affected by the temperature, one may, and probably does, expand or contract more than the other, and the condition is still worse.
The result is simply this: each tube must contain a quantity of material sufficient to resist the strains caused by a full load, and more (on account of the bolting down of the abutments) or be injuriously strained.
This doubt as to which members the strains will affect should not
exist in regard to any truss, and the axiom may be laid down, that a plan
in which such ambiguity of strains obtains is sufficiently defective to be
xejected without further examination.@  This complication may be prevented in a simple manner, and the strains confined to their respective
members which then need not be proportioned to bear all that can possibly come upon any part of the truss.
Suppose that of a framed arch the lower are member at each abutment, and the one of the centre are removed, the structure will then be
of the form  shown in the figure (next page), divided into two parts,
jointed together at the centre and similarly joined by the same tube or
arc to the abutments.  Expansion and contraction will not affect this
truss, and the upper arc has its work well defined, which work cannot be
added to by any action on the part of the lower arc.
The horizontal component of the strain in the upper arc is
wet  S
H  =-8                                (1)
where w is the maximum load uniformly distributed, s the length of the
* There is no difficulty in calculating the strains in a correctly planned arch truss; they
are readily determined by simple mathematical reasoning as in any horizontal truss. The
labored and uncertain theory of arches as applied to masonry, has no more bearing upon
arched trusses than it has upon a Howe truss or a king-post bridge.




229
span and h the height of the centre of the arch above the points where
it is joined to the abutments.
The segments of the lower arc are sometimes subject to compression
and sometimes to tension, according to the position of the load. The
braces are needed only under partial loads. That the greatest tension,
in the lower arc of either segment is wl llen that segment alone is fully
loaded, is self-evident; to determine the amount of strain and the
requisite quantity of material, some calculation is necessary.
-—.. —------         --
Let the arch be loaded uniformly (horizontally) from A to C, then
since t', represents the full load of equal density, the weight borne by
a'i
the abutment B, or its vertical reaction, is 8  The load upon A C
produces a strain which must pass through C and B, and which may
consequently be represented by the strain line B C, of which force I7
will represent the vertical, and 2 the horizontal component. This vera.  l{.S     wll  S'n S
tical component is  8.'' 2  1-7'
" 8'   16 h
w.s
hence -16j is the horizontal component of the force through C B, or
the horizontal reaction of the abutment B; it also is evident that there
are the same forces acting at C as at B.
Take moments at any point (, where the braces meet in the upper are
in the loaded part, distant x from the centre (C, and we obtain for the
longitudinal strain in the opposite member of the lower arc, L X d, (d
being the depth of the are) - the vertical reaction at C(X x. which 
WV x    w s
8  +  16 h Xy (the term being the horizontal reaction at the same'I. X 2
place, X y the vertical distance of C above the point a) -
2 s
(the load on x); collecting these we have
IV x   ws y 1,X /2
L      d   --  -       2s                  (2)
8     16 It    2 s
Since the curve is a parabola, y may be expressed in terms of x. thus:
from the properties of the parabola, x' = 2 p y, and -i- - 2 p h




230
s2         4  i',-2
2 p --..   - -', -?hence
V'X   w X< 2  IV X       W xt   W  2
d     =   +  4.  — 2s' L =8 (-  4'         (3)
From this equation we can find the strain in each member of the
lower are of one segment when that alone is loaded. It is seen that
when x =' L reaches its maximum value and becomes -6-4 7
4                                          64 d
For convenience in calculating, let d bear a certain proportion to h,
say cd = -4, as this is somewhat near the facts in the St. Louis Bridge,
tw s
and then                  L    1                            (4)
As the load passes on and gradually covers the whole arch, this tension
diminishes until when the arch is fully loaded, it entirely disappears.
When the arch is half loaded, as in the case supposed, the upper arc in
the loaded part is subject to much less compression than when it is
fully loaded. When one segment is loaded, the upper arc of the other
is subject to tension and the lower to compression; a glance at the
figure will show that the pressure from C in the direction of B will tend
to cause a bending upward of the segment C B. Here again we can
easily find the strain; take the moments around a point distant x to the
right of C, the left segment being loaded, and we have the same equation
x    W s y
as before,             L d =   - - 1i6 h
the forces at C acting in opposite directions.
As the load comles upon this segment, these strains gradually disappear, and unlike the previous case, are diminished by the weight of
the arch. The tension in the upper arc can be entirely prevented by
giving sufficient depth to the arch.  To do this in the St. Louis Bridge,
12 feet will be found to be ample; that is, a depth of 12 feet in this arch
will prevent any tendency in the unloaded segment to rise when the
other segment is loaded.
These investigations have been made to obtain formunla whereby
the quantities of material may be compared.  The cubical quantity
of material in any member of a truss is the area of its cross section
multiplied by its length, and the area of the cross section varies directly
as the strain upon it, so that if the strain is multiplied by the length
of the member supporting it, the product will be all that is needed for
the purposes of comparison, and a quantity from which the weight may




231
be directly calculated. For compression members a certain percentage
to prevent buckling or bending must be added, which generally may be
put at, say 25 per cent.; in tension members this addition is unnecessary, and in comparison no extra material need be allowed for joints, as
this may be assumed to be the same in the different cases.
In the centre arch of the St. Louis Bridge, the span is 515 feet, the
rise at the centre 51.5 feet, and the length of the arch in even numbers
is 529 feet, or about 2- per cent. longer than the span. The horizontal
component of the strain throughout one arc or tube when fully loaded.w s
is -8' and bears about the same proportion to the longitudinal strain
that the span does to the length of the tube, or the longitudinal strain,
1.025 ws
L     ------,                     (5)
8 7(
multiply by the length of the tube, 1.025 s, and we have the quantity,
1. 0625 w si2
Q=; — 08-2;                     (6)
8k h(6)
next add 25 per cent. for stiffening, and
1.0625?w s2  1.0625 iw s  5.3125 w s2
Q      8 h    +    32      -    32              (7)
Now, as each tube of the St. Louis Bridge is liable to be subjected to
the whole load, the quantity of material required for the two tubes in
5.3125  s
one rib, is             Q                                  (8)
Compare this with the quantity of material in an ordinary horizontal
truss of equal length and depth. The quantity in either chord is represented by the contents of a parabola, whose base equals the length of the
truss, and whose height is equal to the strain at the centre of the truss,
iws   2 s   w s2
or,                  Q   Sh      3   =   12h;
add the 25 per cent. for the upper chord, and
5 ws2
Q     48 h;                     (9)'t   S2
for the lower chord,       Q = 12h                        (10)
9ws2
Adding (9) and (10), and   Q-   48h                       (11)
Comparing these quantities we find that the chords of the horizontal truss require less than 57 per cent. of the material needed in the
ribs of the arch. I do not know how much material the tubes contain,




232
but if they are properly proportioned to bear the loads, there must be
that difference, which will hold good if the arch and truss are subject to
the same strain per square inch.  This enormous difference being considered, it is hardly worth while to discuss the comparison of the quantities in the braces. No argument further is needed as to the merits of
these two systems.
To compare the arch of the St. Louis Bridge with that of the figure,
we have, in the upper arc of the latter, the same quantity as in one of
the tubes of the former, or
5.3125 w s2
Q —    32 h
In each segment of the lower arc the strains vary as in a horizontal truss,
S
and are equal to the contents of a parabola whose base =   and whose
W s             w s    2   s    w s2
height =   -   ], hence Q = 1  X 3 of 2 = 48h
t   S2
and for two segments,      Q =   -24;                       (12)
add this to the quantity for the upper arc, and
19.9375 w s2
96 h
Q=      96   - A'(13)
which is 62 per cent. of the material in the St. Louis arch. The horizontal truss has still the advantage even of the arch properly constructed;
the latter does not need so much material in the braces, but the roadway
is to be supported, and other expensive items added that will probably
more than compensate for the difference.
The form of arch which the writer now recommends is certainly a
great advance upon his previous one, but it is doubtful if it possesses
great advantages over an ordinary bowstring. His views in this particular
seem rather extreme, and will hardly be accepted by engineers, who are
likely to be satisfied, for the present generation at least, with what has
already been done in the construction of steel arches.




233
TABLE I.
ESTIMATE OF WEIGHTS AND QUANTITIES OF DOUBLE TRACK RAILROAD
BRIDGE OF Two SPANS, EACH 400 FEET BETWEEN SKEW-BACK
PINS; WITH LEVER ARRANGEMENT.
Weight of structure for a main arch of steel-1.28 tons per foot, and of wrought-iron-1.93
tons per foot. Live load —2.5 tons per foot. Versed sine —.
1ST. MAIN ARCH OF STEEL. (Ten per cent. having been included for weight of joints.)
DESCRIPTION.                         Steel.   Wrought Cast iron. Timber.
iron.
Pounds.    Pounds.    Pounds.    Pounds.
Weights in one span-Main arch..................         253,630.........................
Counter  arch........................................    110,880....................
Main bracing......................................    147,550..............
Lower chord and track stringers- 
Chord in centre............................!..........       48,720...................
2 outside chords.......                                      8,800................
4 track stringers........................                  137,600..........
Cross beams.........................................        60,523....................
Wind bracing of main archStruts...............................................        20,000...................
Rods................................................         46,466...................
Diagonals...........................................   10,000................
Wind bracing of tracks...............................       27,048...................
Suspension  rods......................................    21,000 
Cross-ties and planks.....................................                25,440
Rails.................................................       32,000.....
Joint in centre................................................          7,000.........
Total................................     253,630    746,587      7,000     25,440
Total weight of span 1,032,657 lbs.=1.29 tons per
lin. foot.
Weights of both spans............................      507,260   1,493,174    14,000     50,880
Castings above piers...................................        63,000..........
abutm       ents.....................................        36,000........
Ends of main arches....................................     20,4001..........
Levers and pins......................      4,620.........
3 columns of centre pier....................                      141,000..........
W ind  bracing.........................................    20,000.........i.......
Lanterns.....................................                      4,500..........
Total................................     511,880  1,513,174    278,9001    50,880
Masonry (estimated from 50 feet below superstructure)-2 abutments, 6,506 cubic yards.
2D MAIN ARCH OF WROUGHT IRON:
Weights in one span................................ 1,520,587           7,0001    25,440
Total weight of one span 1,553,027 lbs. = 1.94 tons
per foot.
Weights of both spans...................................         3,041,174    14,000     50,880
Castings on pins and abutments.........................     99,(00..........
Ends of m ain  arches............................................     20,400.........
Levers and pins...................      4,620 
3 columns of centre pier....................................           141,000..........
Wind bracing.............................          20,000...................
Lanterns.......................................................    4,500..........
Total.............................          4,620  3,061,174    278,900!    59,880
Masonry (estimated from 50 feet below superstructure), 2 abutments, 7,386 cubic yards.




234
TABLE II.
ESTIMATE OF       WREIGHTS AND       QUANrTIrES      O1' DOUBLE       TRACK     RAILROAD
BRIDGE OF FIVE SPANS, EACH 259 FEET, BETWEEN SKEW-BACK
PINS; WITH      LEVER ARRANGEM.ENT.
Weight of structure of main arch of steel-1 ton per foot: and of wrought iron-1.2 tons
per foot. Live load-2.5 tons per foot. Versed sine-)-.
1st. ONE SIDE SPAN. (Ten per cent. being allowed for joints.)
MAIN ARCH OF WVROUGHT          AI  A   H OF ST EL.
IRON.
DESCRIPTION.             -___ _                                   _.     __
Wrought    Cast   Tim-    Steel Wrought Cast        Timiron.    iron.   ber.            iron.    iron.   ber.
Pounds. P'nds. P'nds. P'nds. Pounds. P'nds. P'nds.
Main  arch........................  183,000..............  6,000.........
Counter  arch....................  44,500.............               44,540
Main braces. Struts.............     24,0..........     24,000
Rods..............     29,880....................   29,880
Joints............     5,000...............     5,(i0
Lower chords and stringers......   169,250.................   169,250
Cross beamss....................     41.700..............      41.700
Wind bracing....................    18,      2880......      18 644.......  28,380
Diagonals and wind bracing of
counter arch...................    4.00..;               4,000
Wind bracing of arch. Stays.....      6,076..........           6.076
Rods.....     6,992...................        6,992
Coupliugs and eyeplates for rods..    2,000....................     2,000
Suspension rods..................    6154....................         6,154
Bolts and nails...................     688..................          688
Rails (60 pounds per yard).......    21,00.....1,000
Joints in centre............................ 6,0001.........  6,000
Total............  562,924' 6.000  28,380  86,00(0  379,924   6,0001 28,380
Total weiht of ones ie span         597,304 pounds.           500,300 pounds.
Total weight of ones ide span  -_=  2,4)100 pounds per    - 2,000 pounds per lineal ioot.
lineal foot.
2d. CENTRE SPAN: * 
Total weights....       5:33)(1,6  6,000 28,380;  86,000  350,600  6,000  28,380
( 5681,0ft)) pounds.         471,000 pounds,
Total weirght of cesntre spfan.....:,-.2723 po unds per      1,884 pounds per lineal foot.
lineal foot.
3d. FIVE SPANS:
Centre span......................  533,60)   6,000  28 380, 86,000   350,600  6,000~ 28.380
4 side spans............ -.,251,710)  24,000 113,520' 344,000 1,402,400  24,000 113,520
Castings on top of piers and abut-'
ients.......................    0,000.................     150,80 0
Ends of main arches................ 30,000....30,000
Levers, pins and connecting rods.    12,000................
Lanterns...........................................  3,000S
Total........... 2,797,300 213,000 141,900, 436,000 1,753,000 213,000! 141.900
ft.B. M                          ft.B.M.
=37,300                          =37,300
Masonry (estimated from    50 feet
below superstructure.)
2 Abutments............... 4,910 cubic yards.          4,636 cubic yards.
2  Piers.................... 3,60                    3,6303, 0 
2  Piers.................... 1,410'        1.40 "..1,4
Total...........9.950 cubic yards.' 9,726 cubic yards.
~ The centre span is not a bowstring girder wittl chord, but purely an arch; the thrust
of which is received )y the piers between which it is placed. Hence it is lighter than the
other spans.




-..._!',.-                      ": —'7 = =
i   -..._ -'.' -"
A..                 -— *_ — -.. —--
A         _            __ i >
-' - --
I3  AI
t.5.2  lf~re k!~.~8-L25.8  t 36.29  -_43_87~., 9__i C~)   3)0 








*j^-w  - -  - "'_ _ j r^'i.  ^~?*/g? ^s>  w   *^  ^'~i-'  -_  _ * _  _  _ __ - _ _  _  _  _ 7~fsi -  -- \  -  -   _- __ -  -  -   -  - j -  -  - --  j -' - \-  — T-l~o'
*h ^  13 ^^ " ^^ ^'^"                  *^ -      -  -   -;4 =::j::R: _ ~ "-   -ZZZ.Z ^ Z Z.8^,,  __ _ _   _ _ _ _ _ _: _, _ _ _ ___ _ Z E ^ _ _ _ _ _ _ _ _ _ _ -_   __ _ _ - - --.- - - - - - - _ -_ 
9QT~~~~rz _ _____________________________________~^^ -^ __ __ __i __ __ __ __ *-~ —^..__ _ __ __ __ __ __ ___ __ __. __L _ __ __ __ __ -- __- __- -- _ - -- -- -- -- -- — ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~v?(cs,7 ~ ~.^ r - - - - - ab! A,-  - -^-2y^: - Ok O  91~  l, a~.  04' 0~9   5r o'  - - o —-l'-tv~. o*'~~~~ ~ ~."~~~" 0  OTFi (Y X  izPz iz^_ ~zzR 
70W +                                           0471=.+== z^ = ^E ^=::=:^
~-~__ — ~ __ "^ __ ~ZZZ __~^ ^:y  ^~'^^ — ~ —.~''r~.-~~-~   -- __ __ __ __ i  -- _ -- —.   -— j ___,~"~"_  ",  ~~"~~"~~ __ ~__ZZ3Z __ _ ~ZZZ_~ZZIL __I____.   s?"!~~~~4-- -- -' — - -- -- -'   -- - -- - t - - - - ^: ^ - -- -- - -- - - -- -- ------ - ~ ~ ~ ~ ~ ~: " Z ~ Z Z: Z Z Z: Z I I I _ _ ~Z Z ~Z Z Z Z Z Z Z:. _ __   I  
y —/6 r ot^ ('f  j^ 0^ ff  ti 0^ ~ r t~ o~_ off QQ />e gggp o^ sc?g  0 f ^'!^~''j_'- ^. Q.J_.^'- ^S 1)<'  o'  A'f 6^ ^ ^^ O^ 0.^_ ~^ ^I - ^ _06_^~_. _~9_S~. ~t__M_~j _'TL _ 2
-- - -- - - -- - - -1- -- - - -- - - -- - -- - - -- - - -- - -- - - -- --- --- - -- -; —- -::^ — ^;- - -- —.- - -- - VI1
0-9^^              ^   ^   ^   ^    ^^    ^    ^   -^-      ^    ^    ^^    ^   ^   ^   ^:^:       t o
- - - - - -   - -: - - - - - --,  I ^ - - - -  j   - -- - - - -''" " ^ """l  ^ ^ ^ i' " " - ~ ~ - - - - -'"' "' " " T - ^   M ~ u'i  5
OP__  _ __ _ _ _  __ _ - - - - - - -- _ __ _ _ - -- - - - - -- - -- - -- -- ^ - 0!- -- -- - - { - -- -- - - ^ - -- - -- -- _'_ ". -^:"'= -^   ""1
I 097       }   ^ ^ ^ -J ^ I  *' Z ^ ^   ^ U  ^-:^  - ^ ^j!'j -^^ l ^  1\   I'/;;'"^^'l/^              i~^^          \,;fj; i'\;'.j7 z! 
0,+                                                 4 -'                    - -  ^   7 -   u 2 ^'* 9
^^^..1 —--— *~~                              ~~ 7A                                      /    sa
*    fwl/'m*yi                               f^^/~ l i/f








'reve0.-'^.  ^77
^/'^^^^^*^'^^c-^^^^^^//^^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ IX~1:;"
^===================^:^^======^===^^^===^^^^T'rn so "^ pi _., — - - - ^ -  ^ ^   ^ ^ ^ ^ _ _ ^   _ _.,^. -^ ~7'0 y t ~   ^ ^:: = ^:::^ - -  ^ -   ^ *   -   ^   j *  __ _ ~   * _   _ __ ^ "lMm C
~*0ft ~ "   ^^   ~   ~   ~ "   \ ^ 1~ __''~ _|__ -J - " ~ -^ -- -  ~ ot   ^ j   ^   ~   - * ~  ^.  ~_   *.".. ~  _ __ ___r  S W
*^ -7 /^z' 72       ~                      /^'' - o y^ t   j~i- -  - ---         -  -
| ^ -- ~ ~ ~  -------: = = -  ^ ^ ^ = ==:==: = =:= = - = === = =:^ = =: = ====:=:= = = == = = = =:==== = =:
do  _ ___   -  _ _-   g g  __  _ _6 —    ___     _ _ —--           -----
t ======^^====^              ===^===^            ==============.1                      -de^^ ^^ ^^^ — ^^  -^   — ^^^         ^^^^^^^^
-900^  ^- - "" " ^                ~  ~ ~ 9 o ge _   ~   * __ __ __   __ _ _ __  __ __ ___ __ __ __ __ __ _ ^
^.4   - - * *' ---- --- --- --- ---- - -- --- --- - --- --- --- --- ---- --- --- --- ---- -— j -- *^ ^ -- r " --- --- ---- --- --- --- ----------- --- --- ---- --- --- --- ---- -- - --- --- ---- --- -- - --- ---- -— 4
zzz -- - zrzzz -rr - - ~~~^ - - - -  - -^r^- -   - -- - -- - - -- - - - --   * - - - -^ - ^ -^=-:^ Z:ZZ: _ _ Z _ - ^^^ ^:::^ ^ g;'[ |^-_ _~ 1   ~ -~^~Z _ Z^^ ^::r - ^ ^ ^ I:   Z- - _ _
-- -- -- _^ ^ ^.^ -^^,. ^ ^ — ^ -. -- -^.^ - -- -- -- - -- ---— _ -- --. -- --   --.  j  ^^ -- -- -- -- --   -- -- -- r^^^^ -- -- -- -
__ __ ^-^ ^ ^ — nr:: __ __ 4 ~ ^: __ -,^ 2 - __ __ __ __ __ __ __ __ -- -L -- __ - __ -- -- --'- -- -- -- -- -- -- -- -- -- -- -- - - -- -- -- --. __ -- -- ^ ^ ^ ^ --:^ __a,. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~8 9_0   100   110_ _1207_____  ZZO_230  4_ _250'  Z60   27_zs_ %_ _ __00  a_ _3_o  3-O 3)+_ 35
m 0'   10  20  30  40  5____0 60___ 70  1___ _ _ ___ __ ____1___0_ _ _ __ _ ___160_ ___170_ __ _ __ _ _ __ _ __ _ __ _ - 1








9^~-.9j         Jzs~zcv~~rc^-r^. ~ St rain o/s o  (.eie.,S'ay,,..redoe nc ~t ~'. Ollg.._  /r~r,
]oz f....?yvsseoo tgo?, -— /-7az' e7s o v: — e —./ —-- ----------------------- ~ _ -----------—.._?- -, O Jn.e. o             - \  L, i  o I
9,,~ ~\,I            -I i' i!            /'       "
1       o, t.... _ t.                         A -— = —-r-  F  ~
60.A-. -  l-___ __ I _ —   t  _ 4 —- --- t                   -.::
6e                         I..',       I      I
s. voao  ------- --  ~-  t  -i- - — 1 ~-i - - - --  t-  " —----
4 -4. 06   ___ I_____ __ ___ __-_ __ _ J —':- _  -~!  h_ |  +__ to 6.
1 oo.l X_ _      __       _          _____.- \ 
7o  _ -| —- _ -_ —-- - -.. ----  - __ ----- I ---------— 6
*   *   8o<>'  ^  i   i  i   i   i   i   *~ ~ ~~~~~~~~~~', I a_.  _ _   _ _a;"~e  L t  -, I'gstsouc  __gbe   I t..................   I'..2yo *. e/ ___e__  ____-..-. —-_[X7 -,, _ _  -... ——,, 0-.-_ 6 t___ __  _____  __-_-._  ___ -fiv_____ L-.. ft-............ 
F -    j,_   -.  ____ _______.  _ _lS ( / c   ^ f   ^ ff   ft.' ^   ___ -- --








161
Railway grade crossings will be free from accidents if guarded at the
intersection by signals which block one road when they open the other,
and which are repeated when necessary at suitable distances on either
side. No sane engineer will run past a red light in full view, into an obstructed crossing.
We have laws requiring trains to come to a full stop before passing
such points, which is well, but well devised signals are better, on the principle that an engineer in a hurry will be more likely to take risks by
breaking the laws of the land than by breaking his own neck; we hear
much of the carelessness of engineers as the cause of accidents, but as a
class they have so high an appreciation of their lives and of the risks of
their calling as other men and will not be found running into known
danger. In a majority of cases the want of care will be found more with
those who have charge of the roads and running arrangements, and they
should take full share of the responsibility.
Tunnels cannot be safely worked except by the full block system.
Hoosac, with her five miles of length, cannot be safely worked on any
other principle, although such an amount of business may be conceived,
as would make a block station desirable inside of tunnels so long as this.
The Committee is without necessary information to treat intelligently
what may be termed the railway vocabulary of the semaphore, but American ingenuity will find no difficulty in supplying the deficiency. This
report is closed by repeating the recommendations of the Institution of
Civil Engineers, that uniformity should be obtained:
First.-In signals of danger, caution, stop, go ahead, and go back,
Second.-In the character of the fixed signals used upon our several
roads, for distinct purposes, such as switch, block and station, and
Third.-That the signals should be as few and simple as possible.
For the Committee,
April 24th, 1875.                       J. DUTTON STEELE,
Chairman.




162
DISCUSSIONS.
ON UPRIGHT ARCHED BRIDGES.*
MR. SAMUEL H. SHREVE.-In replying t to the criticisms on his
paper on "Upright Arched Bridges," Capt. Eads enumerates so many
errors in my remarks upon the St. Louis bridge, that I seem liable to a
charge of inaccuracy.  It is true his assertions are unaccompanied by
proof, yet the time and thought he has devoted to this subject give his
positive declarations more weight than they would otherwise be entitled
to. At the time of making my previous remarks, I had not sufficient data
for numerical calculations of the strains affecting the arches of the St.
Louis bridge, and was, therefore, obliged to give only general formule.
I have now the information I then needed, and while briefly noticing
some of the points of difference between us, I propose to devote this
paper chiefly to a careful analysis of the strains affecting the ribs of the
centre span of this bridge, basing it solely and entirely upon Capt. Eads'
statements of the condition of the arch at different times, the weights given
by him and the actual dimensions of the members of the tubes, as the
upper and lower chords of the ribs composing the arch are termed. If
these are found not to possess sufficient strength, it is unnecessary to examine the braces, joints and other parts of the structure. and if it were
necessary, I should be compelled to limit myself, for want of space, to
the matter I have chosen.
The St. Louis bridge is a fair subject for criticism. It differs from
any other metallic bridge in this country, it has cost a vast amount of
thought, labor and capital, and if it possesses one-half the merits Capt.
Eads claims for it, it is time that bridge engineers should examine it more
closely than they have yet done, and compare its strength and economy
with the other styles that are now so much more popular throughout the
world. Engineers are by no means a bigoted or prejudiced class, and if
it can be shown that the St. Louis bridge has any great advantages over
the more common types, they will not be slow to follow the example
it presents them. It is purely a question of form and not one of material.
* Continued from page 85.         t Vol. III, page 319.




163
But first they require a clear and complete analysis of its strains and the
proportions of its different members; these have not yet been given. The
pamphlet of investigation published some years ago undoubtedly does
credit to Capt. Eads' ability as a mathematician and shows him expert in
the use of the integral calculus; but it is incomprehensible; I am inclined
to think, to the majority of engineers, and is, consequently, unsatisfactory. This investigation gives no rules or formulae for determining strains
in an arched truss, and is only an illustration of the fact, which must be
remembered, that the calculus will not determine strains affecting a truss,
whether arched or horizontal.
Before proceeding with the analysis, I wish to allude, in as few words
as possible, to some of the points in my former paper* which Capt. Eads
has termed mistakes, and has numbered so that there may be no difficulty
in referring to them.
The reply of Capt. Eads has these words: "Mr. Shreve says,'the
paper before the Society appears to be an acknowledgment of the disadvantages of the system adopted for (the superstructure of) the St. Louis
bridge.' " t In his paper under discussion on " Upright Arched Bridges,"
Capt. Eads' language is, "I will first point out some of the disadvantages
of the system adopted for the superstructure of the St. Louis bridge."+
My first error seems to be in assuming that Capt. Eads acknowledges
the disadvantages which he himself "points out." If he will not go this
far, we can hardly expect him to acknowledge the "disadvantages" others
may "point out." The next three errors may be summed up in the
fourth, which is, that I declare "that each tube must contain a quantity
of material sufficient to resist the strains caused by a full load.". These
errors may be better judged of when we come to the analysis.
Again Capt. Eads writes: " Hence Mr. Shreve makes a fifth mistake
when he asserts that' bracing beneath a parabolic arch uniformly loaded,
can receive no strain and is useless.' " i For this error, I think I am pardonable when I offer as an excuse, that Rankine, Stoney, Weisbach and
every other author I have read, who treats the subject, has made the same
mistake. It is, too, an error one is likely to commit, since the most simple methods of determining strains, the method of sections and the
graphic method, lead one so easily into it. Capt. Eads himself has not
escaped it; in speaking of the upright and the inverted arch, his words
are: "If either form have an equally distributed load placed on it, no
bracing will be requisite to preserve its normal curvature."  In other
* Vol. III, page 227.  t Do. page 323.  t Do. page 195.  ~ Do. page 228.
1 Do. page 324.             ~T Do, page 198.




164
places in the same paper he gives evidence of his falling into the same
mistake which he attributes to me.
The eighth mistake is in making the assertion " that the greatest tension in the lower arc of either segment is when that segment alone is fully
loaded." * Capt. Eads then declares that "greater tension will occur in
the lower members when the maximum load per foot has advanced so as
to cover either three or five-eighths of the entire span,"t and claims that
this is shown graphically in one of the figures accompanying his paper.
This assertion has been made, I think, rather hurriedly, as there is
nothing shown graphically in any of his papers concerning the bridge. A
diagram of strains is not a graphic solution and proves nothing whatever.
The remark just quoted is applied to a jointed arch, such as I have
previously suggested, as well as to the St. Louis bridge, and is almost
wholly incorrect. There is but one point, in the case alluded to, in the
jointed arch where a three-eighths load will produce the greatest strain,
and there is most positively, no point in the lower arc of the fully loaded
segment where an ounce more on the other segment will produce a greater
tension than the half load. When one segment is loaded, any weight
whatever, in any position on the other segment, will lessen the tension
on the lower arc of the loaded segment. I respectfully invite Capt. Eads
to prove, mathematically or graphically, that he is correct on this point.
He need not have recourse to the calculus, as there are three points
through which all strains must pass.
The greatest tension at any point in the lower arc produced by a uniform load is when its length extends a certain distance from the abutment
bearing a fixed proportion to the distance of the point; a formula for this
strain may be very readily determined.
There are other errors of which I am accused, which, however, are
not of much consequence in this discussion, and with which I do not
think I would have been charged had Capt. Eads read my paper a little
more carefully. Two of my errors were in assuming that each tube must
at times bear the whole load and must be so proportioned; upon this
assumption I made my calculations and comparisons, showing the
economy of the chords of an ordinary horizontal truss over the ribs of
the arch.
I propose to admit these errors and to make a new analysis and a
numerical calculation of the strains upon the ribs of one of the arches of
* Vol. III, page 229.         t Do. page 325.




165
the centre span, assuming nothing, but basing my whole argument upon
premises furnished by Capt. Eads' own statements contained in his papers
which have appeared before the Society.
Capt. Eads writes as follows: " The medium temperature was assumed
at 60~ Fah. The effect of temperature (ranging from-200 to + 1400
Fah.) increases the length of the rib about 6 inches.  This extension
causes the crown to rise, which relieves the lower tube of compression at
the abutments, and hence that tube does not then support any portion of
the weight of the rib or the load. These are borne wholly by the upper
tube at the abutments, and hence its section there must be increased accordingly. At the crown, the strains are likewise changed. There, however, the lower tube does all the duty, as the upward bending of the rib,
relaxes the compression in the upper tube at this point, hence the lower
one must be increased at the centre of the rib to enable it to do this double service. It is thus seen that the upper tube at the abutments, and
the lower one at the crown, must, when the temperature is at its maximum,
perform the whole duty of sustaining the rib and its load." " The extra
section required by temperature, therefore, becomes smaller in the tubes
as we leave the crown and abutments."
" These are the effects of expansion. As the ribs are formed for a
medium temperature, contraction by cold reverses these strains. The
crown of the arch falls, and the upper tube at that point must be reinforced, because it must then sustain the entire compressive strain, which
at medium temperature, is borne equally by both; while, at the abutments the lower tubes have now to carry all the load. Thus it is seen
that, both at the abutments and the crown, the upper and lower tubes
must be each of greatly increased section on account of the changes of
temperature."
Again-" Thus, at times, a tube at the abutments may not only be relieved of all compression, but may bear tension and pull upon its skewbacks, by which an additional compressive strain may be transferred
through the braces to the other tube. "*
Before proceeding further, attention should be called to the fact, that
more than once, as will be seen in the above quotations, Capt. Eads says
the crown of the tubes must be of greatly increased section to resist the
increased strain at that point, and the impression is certainly intended
to be given that the crown contains a greater quantity of material than
the haunches. The truth is, the two tubes are of uniform section
throughout, with the exception of the five members at each end.
" The total dead load of the (centre) span is 8 000 pounds, and maximum live load, 6 400 pounds, per lineal foot. "t I give below, other data
* Vol. III, pages 196, 197.                t Do, page 326.




166
for the centre arch, which I have obtained from personal investigation:
span, 520 feet; actual rise of the arch, 47.05 feet. Each tube has an
envelope of wrought-iron, 4-inch thick. There are six steel staves in each
tube; outside diameter of staves and envelope, 18 inches; outside diameter of staves, 171 inches. The dimensions of the tubes are as follows:
the numbers begin with the abutment members in each tube.
THICKNESS OF THE  AREA OF ONE STAVE IN TOTAL AREA IN SQUARE
No. OF MEMBER.
STAVES IN INCHES  SQUARE INCHES.       INCHES.
1.             2l.               17.11.           102.66.
2.             178.              15.34.            92.04.
3.             1.               13.51.            81.06.
4.             1.               12.09.            72.54.
5..      T.              12.09.            72.54.
Thence to centre.   1-.                10.14.           60.84.
From this, it will be seen that only the five members in either tube
nearest the abutments, are of increased section. The uniform length of
the members of the upper tube is 144.965 inches. The length of the
skew-back members of the lower tube is 114.983 inches, of all the other
members, 142.622 inches; vertical distance between the tubes at the
centre, 12.05 feet. The total weight of the bridge and load is 7 488 000
pounds; the weight of one rib and its load is, therefore, 1 872 000
pounds.
Let w = the weight of one rib and load, s the span, and h the rise of
the arch and d the depth of the rib at the centre; then, according to
the well known formula, the horizontal strain at the centre,
H= 8W; substituting the values given above,
8h
H= 2 586 184.9 pounds.
If we consider each tube as taking half this strain, the compression upon
each will be 1 293 092.5 pounds; dividing this by 60.84, the area of the
centre member, we have 21 252 pounds as the strain per square inch-less
than the limit, 27 500 pounds per square inch, which Capt. Eads assumes
as a perfectly safe load for steel in compression.
This, however, is the strain only when both tubes are equally loaded,
a condition which by no means always prevails. When the rib is expanded by the temperature, and rises, the whole strain must be taken by




167
the lower tube, for the upper is entirely relieved, and the whole strain is
not simply 42 505 pounds per square inch, or the amount that was before
upon the two tubes; for this reason, that while the strain comes upon
the lower chord at the centre, the abutments react against the ends of the
upper. Hence, the actual rise of the arch is no longer h, but h- d, or
instead of 47.05 feet, we have 35 feet. The equation therefore becomes,
ws
8 (h-d),
Supplying the quantities and dividing by the area, we obtain 57143
pounds per square inch for the strain upon the central member of the
lower tube.
The formula given above is as correct as are the principles of the lever,
and the result is simply inevitable. The conclusion is apparent from the
figure. No matter what the intermediate shape may be, if it be uniformly
IXSS;:                                           Ca~ylS i?
loaded, the weight multiplied by the span and divided by eight times the
height of the centre point above the points of contact with the abutments,
gives the horizontal thrust at the centre and at the abutments.
This compression of 57143 pounds to the square inch upon the central
member of the lower rib is considerably more than double the allowable
strain, and makes an excess of 130 380 pounds over the test to which this
member was subjected.* Reasoning upon my erroneous assumption that
each tube at times must bear the whole load, the strain in this case would
be 6nly 42 505 pounds; an error, as is evident, entirely upon the unsafe
side. If the arch be in the same position and unloaded, its deadweight
* (Foot note,) Vol. III, page 326.-" The tubes are all of cast-steel, tested to 55 000 pounds
per square inch without permanent set, and the highest maximum strain they must bear will
not exceed 27 500 pounds." In the Appendix to Capt. Eads' Report to the Illinois & St. Louis
Bridge Company, for 1870, under specifications for cast-steel work, we read, "' The staves
composing the tubes will be required to stand a compressive strain of 60000 pounds per
square inch of section without permanent set." This latter test seems to have been abandoned as being too severe, though so slightly in excess of the actual strain.




168
will cause a strain of 31 746 pounds to the square inch, or the compression from the weight of the bridge alone is greater than the
maximum strain the member must bear. A strain of 57 143 pounds per
square inch is, even for steel, an unusual amount, and it may be questioned whether a structure subject to such compression is safe; surely an
iron bridge, with so low a factor of safety, would be immediately condemned; but it is probably by such experiments as this that we are to
learn the strength and durability of steel.
Formulae are next needed to determine the strains in the other members of the tubes; these may be found as follows. Let moments be
taken around a point in the middle of the centre member of the lower
arc when the arch is extended; a vertical section at this point will cut no
other member subject to a strain, consequently the moment of the vertical reaction of the abutment, W X S, in one direction must equal the
2    2
UW   S
moment of the load on the half span, -- X  -, and the moment of the
2   4
horizontal reaction of the abutment, IIX (h-d), (being at the end of
the upper tube,) in the opposite direction, or,
ws   w s
4=      8 +H(h- d),
w4 
whence    = 8 ( -d)                  (1)
Next, from moments around any point in the upper tube where the
braces intersect it, distant x horizontally from the abutment, we have,
for the moment of the longitudinal strain in the opposite member of the
lower tube, the difference of the moments of the load on x and the horizontal reaction of the abutment in one direction and the moment of the
vertical reaction of the abutment in the opposite direction, or,
_       W 2   Uw S y   Wx
2 s +   (h -d)-                   (2)
y being the vertical distance of the point above the abutment; it may be
expressed in terms of x and thus simplify the equation.
Let the curve of the arch be considered a parabola-from which it differs so slightly as not to affect perceptibly the results of the calculations
-and we have from its equation,
s=2   p8,.          2
=2ph,.' 2p-=;
4                  4 h




169
s  2s ~2                      2   s2   S2 Y
Again, (     x --  2 p (h -y),  *     s x   + x
Agai~, (2                ~i 4- 4              4:h'
whence, y= -    (sx —x2).
Therefore by substituting this value of y in Eq. 2, we obtain,'t X2     wh                 w x
L d= -2s   + 2s (- d) (s-x) -       2
which readily is reduced to
I wdx2      wdx
L   2s(h -d) +2(h -d,   T W X  W X2
therefore, L = 2   -   -
2 (h — d)  2 s (h —)'          (3)
from which the strains may easily be found.
In the same manner, by taking moments around a point in the lower
tube where the braces intersect it, we get,
jLd _    wx   w x2  w s (y-d)
L d==                                      (4)
2     2s     8 (h-d)                (4)
W X2       W7 X        7 S
which reduces to, L-=                                     (
2 s (h-d)   2 (hd) + 8 (h —  d)'    (5)
giving the strains in the upper tube. Moments which tend to produce
compression have been given the plus sign and the others the minus sign.
It is simpler to measure x from the centre of the arch than from the
abutment, and to do this, we have merely to substitute for it, in the
above equations, 2- x, when Eq's. 3 and 5 become,
2'W S       V X2
8 (h- d) ~2 s (h-d)'                 (6)
WX
and L — 7 \
2s (h-d)                          (7)
Substituting in these equations the values given above, we have from
Eq's 6 and 7
L = 3 476 571.43 - 51.43 x2.              (8)
and L - 51.43 x2.                      (9)
These apply only when, distended by the temperature, the arch is
resting with the upper tube against the abutments; in the lower tube it
then decreases, as the ordinates of a parabola, from the centre towards the
ends; in the upper chord, from the abutment towards the centre in a
similar manner. When the arch is compressed by the cold and sinks, so
that the upper tube at the centre and the lower at the abutments take the




170
whole amount of strain at these points, the equations vary somewhat.
Owing to the increased height of the arch, the strain is less, but it is
again increased by the fact that, while in the extended condition, the weight
of the rib as well as the movable load is acting against the effects of the
temperature, preventing, as Capt. Eads says, tension in the upper chord;
in the depressed arch they both act with the decreased temperature and
produce tension in the lower tube at the centre, and thus greatly increase
the compression in the upper at the same point and the other at abutments. The equations for this condition, found as before, with the quantities supplied, are for the upper tube
L = 2 973 943.47 - 30.46 x2.              (10)
and for the lower tube L =   30.46 x2.            (11)
To apply these, we use Eq. 8, for the upper chord, commencing at
the centre, until we reach a point where its value is exceeded by Eq. 10,
and then employ the latter to the end of the span; for the lower tube,
take similarly, first Eq. 9, and then Eq. 11.
The table on page 173 gives the maximum strains in each member of
both. tubes: in them x is measured from the centre of the arch horizontally towards the abutments to the point of intersection of the braces
with one tube, opposite that member in the other whose strain is sought.
The numbers of the members, counting from the centre, the values of x,
the amount of strain in pounds, the strain per square inch on the
present area, the area of section which must be added to the present
area of the different members of the tubes to reduce the strain per
square inch to 27 500 pounds, and the total area necessary, are given in
the respective columns.
It will be seen from the table that every member of the two tubes is
deficient in area; many containing much less than half the material that
is necessary.
Since the members in the upper chord are of uniform length (12.08
feet) we may add up the deficiency areas of the fifth column and their
sum will equal the area of one member 12.08 feet long. The sum of the
areas is 639.56 square inches; multiplying by 12.08, we have the area of
a body one foot long = 7 725.88 square inches; next, multiply by 3.4
pounds, the weight of a square inch of cast-steel one foot in length, and
the result is 26 267.99 pounds. This is for one-half of the tube; doubling
it, we obtain 52 536 pounds, as the extra weight of the steel required in
the upper tube to sustain the compression and not allow the maximum
strain to exceed 27 500 pounds per square inch.




171
In the lower tube there are 45 members; the uniform length of all
except one at each end, is 11.885 feet; the length of each end member is
9.582 feet. Performing the above operation with the quantities in the
tenth column, we shall find the deficiency to be 60 637 pounds for the
lower tube, and adding this to the other, we have for one rib a deficiency
of 113 173 pounds of cast-steel. To support this additional weight a still
further increase must be made in the section of the tubes.
To ascertain the amount of this, an equation may be easily found that
will give the weight of an arch which, with a given strain, shall support its
own weight and a uniform load.
Let w'= the load and w the weight of the arch, both uniform, I the
length, s the span, and h the rise of the arch. Then the strain at the
centre of the arch and the horizontal component of the strain at any
other point is given by the formulaH_ (w' +W) S
8h
The longitudinal strain in any member of the tube is to its horizontal
component as the length of the member is to its horizontal extent, and
it will be sufficiently correct to assume that the horizontal component is
to the average longitudinal compression as the span s of the arch is to 1,
its length, or
s:: (w' + w) s  (w' +  ) I:    8h        8h
which may be considered the uniform compression throughout the length
of the arch; and if this be divided by 27 500, the number of pounds
allowed to the square inch, we have (W'0 00 L)  for the area in inches.
220 000 h'
Multiply this by 3.4, the weight of one foot of steel, and by 1, the length
of the arch. and 3.4 (w' +,2 ) l = w, the weight of the arch.
220 000 h
3.4 w' 12      3.4 w 12          3.4 w' 12
ene 220 000hh   220 000 h      220 000 h  3.4 12
Substituting the values —w' = 113 173, 1  531.5 and h = 47.05-we
find, w = 11 573.
Hence, 113 173 + 11 573 = 124 746 pounds, is the additional weight
required in one rib, and the four ribs composing the centre span must
therefore be reinforced with 498 984 pounds of cast-steel before the
compression can be reduced to 27 500 pounds per square inch, or an
addition of over 52 per cent. of the present weight of the steel in the
tubes.




172
In comparing the St. Louis bridge with a horizontal truss of the
same span and depth, it is only just that the same strain per square inch
should be allowed; since, as has been before remarked, it is a question
of form and not of material, nor of the amount of strain per square inch
to which that material shall be subject.  Capt. Eads cannot, while
allowing a strain of over 57 000 pounds per square inch in his structure,
insist that other forms shall be confined to 27 500 pounds.
I have made an estimate of the weight of the two chords of a horizontal truss, of 520 feet span and 47.05 feet depth, bearing the same load
as one rib of the St. Louis bridge-1 872 000 pounds-and varied the strains
so that the members of one chord of one form shall have the same strain
per square inch as the corresponding members of the same chord of the
other. In the lower chord of the horizontal truss, subject to tension, I
have decreased the strains per square inch in the proportion of 27 500 to
20 000, and consequently increased the material, as this is the proportion
Capt. Eads gives between compression and tension members.
I have made the panels 13 feet long, and used the formula,
wl    w X2
8d    2d l'
-measuring x from the centre-to obtain the strains. In this manner, I
find the weight of the chords of one truss to be 145 265 pounds, or 61.48
percent. of 235 627 pounds, the weight of the tubes of one rib of the
centre span of the St. Louis bridge; which is slightly in excess of the
percentage (57) deduced from the formulae given in my previous remarks
on Capt. Eads' paper.
The great importance of immediately strengthening the ribs of the
St. Louis bridge can no longer be ignored, and probably the simplest
and most speedy method of doing this is to joint the upper tubes at the
centre and the abutments and to disconnect the lower ones at the same
points. Then the weight of the latter can be allowed to remain as it
is, while that of the former will have to be increased only to 173 730
pounds; the present. weight being 119 564 pounds, the addition will be
54 166 pounds instead of 124 746 pounds, the weight necessary to be supplied to one rib if the present form be retained.




173
TABLE OF STRAINS FOR ONE OF THE RIBS OF THE CENTRE SPAN                      OF
THE ST. LOUIS BRIDGE.
IN THE UPPER TUBE.                         IN THE LOWER TUBE.
| s i              ~         l,,~                            3s~~~ Ia
P4!i  i  t                                ir           11
1     2      3         4       5       6      7        8        9       10     11
0 o..... j......       0.0  3 476571  57143  65.58126.42
1     5.9; 2 972 868  48 863  47.28  108.12    12.1  3 469042  57 019  65.31 126.15
2    17.8 2 964 270  48 722'  46.96  107.8     24.2 3 446452   56 646   64.49125.33
3    29.7 2 947 092  48 440   46.33  107.17    36.3 3 408 803  56 029  63.12'123.96
4    41.6i 2921 356  48017    45.76  106.4     48.4  3356093   55162   61.2 122.04
5    53.4 2 887 083  47 453   44.15  104.99    60.4 3 288 947  54059    58.76 119.6
6    65.2 2884 337   46 751   42.49  103.43    72.5 3 206243   52 700  55.76 116.6
7    77.   2 793157  45 909   40.73  101.57    84.5' 3109348   51107   52.23113.7
8    88.8  2733590'  44930    38.56   99.4     96.4' 2998635   49288   48.21109.05
9   100.6  2665676   43 813   36.09   96.93   108.41 2872240   47 209  43.61]104.45
10   112.3  2 585392  42 495'  33.18   94.02   120.3 2732272    44 909  38.511 99.35
11   124.1' 2504 985  41173    30.25   91.09   132.22 2577738   42 369  32.89 93.73
12   135.8  2412 540  39 6541  26.89   87.73   144.  2410119    39 614  26.83 87.67
13   147.4' 2311 871  38 000   23.23   84.07   155.8 2228178    36623   20.19 81.03
14   159. I 2203883   36 2241  19.3    80.141  167.5 2033639    33426   13.121 73.96
15   170.6  2087839   34317'   15.09   75.93   179.2 1825018    29997    5.5  66.34
16   182.1 1,964209  32 285   10.59   71.43   190.8  1 706 285  28 044  1.24 62.05
17   193.4  1998912   32 858   11.85   72.69   202.4  1920065   31559    8.98 69.82
18   204.9  2141780   29 524!   5.36   77.9    214.  2146559    29591    5.52 78.06
19   216.3' 24971731  34 425'  18.27   90.81   225.5  2383351   32718   14.13 86.67
20   227.61 27649031  34109    19.85  100.91   237.  2622 741   32 356  14.31 95.37
21   238.9' 3044722   33080    18.68  110.72   248.5  2894328   31446   13.21105.25
22   250.3' 3336426' 32500     18.67  121.33   260.  3168 412   30 863  12.56115.22




174
MI. JAMES B. EADS. My review* of the comments upon my paper on
" Upright Arched Bridges " was written under the pressure of important
engagements, and I find that in my haste I expressed myself so carelessly
as to be misunderstood in reference to the bow-string girder bridge to
which I referred. It is so common to call an entire bridge span, a truss
or a girder, that I have unfortunately misled Mr. Whipple in reference to
the bridge design in question, by speaking of it as a "bow-string girder"
when suggesting that those who deny the " economy of my proposed system of bracing the arch under unequal loads," should "submit to the
Society the details and weights of a horizontal truss of 517 feet span on the
data given, to show how nearly it approximates to 3 100 pounds per lineal
foot in weight."t In a preceding paragraph I stated the moving weight
to be 2 500 pounds per foot, " including everything." I intended to say
that 3 100 pounds per lineal foot is the weight of this single track bowstring girder bridge, including road-bed, wind bracing and everything
ready for the rails.
Mr. Whipple has submitted to the Society, without.details, a theoretical calculation showing that a single trapezoidal truss of 517 feet span,
" capable (when secured transversely by connection or bracing with one
or more similar trusses) of withstanding the vertical action of 2 500
pounds of movable load to the lineal foot upon the whole or any part of
its length, with a stress of 10 000 pounds upon iron and 20 000 pounds
upon steel," per square inch will weigh, but 2 000 pounds per lineal foot.
His comparison is not complete, and hence gives no approximation to
the result of an investigation of the relative economy of the two systems,
inasmuch as floor beams, road-bed, wind bracing, &c., are omitted by
him. He has given the estimated weight of a single truss, or one side of
a bridge, instead of a complete bridge. Of course any comparative design should have regard to equal safety, so that if unusual lengths are
required in compression members, the proportion of diameter to length
should be regarded. The horizontal steel members of Mr. Whipple's upper chord, being 64- feet long, should be nearly 5~ feet in diameter, if
subjected to 20 000 pounds compression per square inch; as the bow-string
design, limits the length of compression members to 12 diameters. As
the end braces of his truss would be each over 90 feet long, they must be
about 7~ feet in diameter, or they would require additional sectional area
or additional bracing.
* Vol. III, page 319.  t Do., page 333.     $ Page 81 preceding.




i-             +  +   4-  4  4   4I
-- -- -. -- -- -.... 4........
\  14
r.....-4 -. —. -----... -... —..-..:.,. -..
-- I —- "~'f-'...... -V -... —"".....
VeI__ ___ ____ ___ _  _ - _\   ^ \ _____  \
s —-- -......... --.......
ie  /B\           \ ^~y Sf — --------  - - ---- -- ------- - ---   4 —
*'4.4e-  -j-'4 —--------
t          j - <    _   - 4- -----  - -- --  _ _ —-   --
-4. --  _. _:  ----- -V —
*;.-        _..: -' - "
4. I.....................-..-  
- -,  / -./. /.-..I.-.. ---
/...........-t  -.-' - -----------------... - I........ -—. - —..........
sff' __ _ __ _ _ ____ __i_ ___ _____ __ ___/  Ij' ",' "  __' __"''.':^''  _____-. - --,/   / /'        __
^  ^............-.........-................-^.........~ ~.......  -b1, _-  _' ~_ —-----—...-.'   -
---.IL — -   - - -----—..J...^::^.^ ^.......................:.............
_ _ _  I   47 4-'  ----— I-    4 4- "F  1
44 ----                             -I ---— J-7 —-; —L —r-T-rr- ----- - i- j^,,1- i
/__<-  _ _____  I/            ______ I________j /I
____   L,___  -  1- ________,,         L __________ ______
s~.,, _  I  I'  r ^  W  [.lt____ __'i._:__,s;'>,~.~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I,p







175
The longest of the braces in the bow-string girder are scarcely half so
long as Mr. Whipple's compression verticals; they resist only one-third
so much compression as tension, and hence, owing to the weight of the
roadway and moving load being suspended from their lower ends, bear
but 3 333 pounds per square inch of compression.
The desire to understand every argument advanced by an engineer
so deservedly eminent an authority on bridge construction, impels me to
ask Mr. Whipple's explanation of his phrase (page 84) " amount of action
the simple arch and chord alone of the bow-string truss under consideration undergo," and of the deductions following.
Herewith is presented the strain sheet of the design of the 517 feet
bow-string girder bridge referred to and estimates of weights and materials; from which it appears that the structure would weigh as follows:
Cast  iron...........................  6  600  pounds.
Wrought iron...................... 1 099 499   "
Steel...............................  310  650 
W eight of metal....................  1 416 749 
Add weight of timber...............  186 637'
Total weight of structure....... 1 603 386  "
Equal to 3 100 pounds or 1.55 tons per lineal foot.




176
WEIGHTS AND MATERIALS REQUIRED IN A BOW-STRING GIRDER BRIDGE.
OF 517 FEET SPAN, ON THE SYSTEM PROPOSED.
POUNDS.
DESCRIPTION.
CAST IRON. WRO'T IRON. STEEL.
UPPER   MEMBER.
Tubes...............................................................250176
Rings of T iron.................................................     17 504..........
Couplings.................................................18 240:
Centre  joint..........................................                  19 760..........
Centre pins.........................      2 076
Pins......................................................                              4 08
Nuts...................................................                     52...........  528..
COUNTER ARCH.
Links.............................................................  108648..........
Pi s....................................................................         12   464
Main braces........................................          127 000..........
Nuts........................................................      684
Stays  (gas  pipe)..................................                     2812..........
BOLTS IN CENTRE.
Upper wind bracing.................................................       96 
Couplings......................................................
Stays  (gas pipe).........................................               11    620..........
Eye-plates.1.............................                                3140..........
Rods....................................................  565
Pins.....................................................                 926..........
Diagonal rods.........................................      4264
Pins.................................................................   208..........
Pins..~~~~~~r~~~~~.~~~~~~~~~~~~.~~.~~~~~                                   208
Suspension rods....................                 37 352..........
LOWER CHORD.
Links.................................................    626 872..........
Pins....................................................                              20 202
Cross-beam s.........................................................  84  240
Castings for cross-beams........................      3 280..........
Bolts........................................................
~~B~~o~~~~lts.~~~~~~~~~~~~~~~r~~~~~~~~. 377
Rods for lower windtruss...3.............................  36 063..........
Pius...3 084
PillS............................................................................  3 084
Eye-plates.......................................................        5000
Total...........................................     6 600    1 099 499     310 650