■<:««^- 
 
 
 
 <SI2J^<.m 
 
 ^^. 
 
 at: mZiA-w:.. 
 
 
 
 
 ;^^~^r^:. c<: 
 
 '.V'- 
 
 ' ,-^^^^^'^"^ 
 
 <' ';, ■- 
 
 " -S-S^' ■ <- 
 
 
 '-)>SS^^"i jfif ■■'- 
 
 ■^" 
 
 r^-^^^^~ ci^"''' 
 
 _■*- ^ 
 
 >^^'' i"*:"^-'- 
 
 i--. . v'". 
 
 ^^--r c£?: 
 
 o<c^ 
 
 |r-— - ^ <-tr^5 <]t< 
 
 '- cc?^ 
 
 J^" c-CX. c<5? 
 
 ':. ■ "<""C.'.«i 
 
 G^- G>c:;PV cc 
 
 ^ •■" ' <r«i 
 
 G'T^" -.£S^--' '^< 
 
 " -',C«GI 
 
 :/-«S!gSl.CCC 
 
 "< • 1 C" 
 
 
 ■ - -«^ ■■ < 
 
 ■<;:"*C^ -WsSci*;^- " 
 
 ::<:■£>:«£-- «ac<sr < 
 
 «>li^ " " " 
 
 . <_«Ci;.. ■ •sot- 
 
 c.6;^ 
 
 :«eaSL.,^CSfc:: ■ 
 
 
 
 
 
 
 
 
 
 r.::?i<i£jmM^ 
 
 gg:&- ';* 
 
 
 PiLIBRART OF CONGRESS. r 
 
 Ui — ^ I 
 
 '^ [SI/aTESONIAH BEPOSIT.] 
 
 ^1 
 
 J UNITED STATES OF AMERICA. !l 
 
 
 
 
 ,<< cot 
 
 «" ftC«CE3S^^"" 
 
 \^^L<^ 
 
 
 
 -Mar: 
 
 <^-,^?^ 
 
 .^^-<<^ 
 
 
 
 
 '5^'*?, 
 
 
 
 
 
 
 
 

 _ _ c-«L <^' t <i -"Cat ^ 
 
 
 
 c<a^"^'^^ 
 
 - ■ _<;.?r.<r a-e<- ear" <;■ 
 -<^ ' '^^^^ ■^•- -^^ ■ ' «< ' -^^ 
 
 
 
 
 
 <T«r < ci<r 
 
 J '■■eye : «J< 
 ;. Car '^'C . 
 
 
 
 
 ''-^t<z '-*^i^r < 01 
 
 
 
 h.*m,: 
 
 ? t cere 
 
 *C,_^?»^M 
 
 r®:cc ■ 
 
 mt-m3m-. 
 
 
 
 3S*te- -,-^-«- r-TC-istC ^j^-^j*'- 
 
 -<cst crc i< 
 
 
 
 
 <S»:i:<«2ac: 
 
 
 ;.c:_- - cr ^^'^Si 
 
 
 1: erffii^-acsor: ■ 
 
 a^S 
 
 -<^fe; G-5K 
 
 
/I. 
 
 <^ /t^ 
 
 <^^ t:^ 
 
 PRACTICAL DESCRIPTION 
 
 ^, _„_ HEREON' S^ 
 
 PATENT TRELLIS RAILWAY STRtfCTURE, 
 
 EMBRACING THE MOST APPROVED MODIFICATIONS; 
 
 ALSO, 
 
 THE PATENT WROUGHT IRON RAILWAY CHAIRS, NEW AND IMPROVED MODE OF JOINING 
 THE ENDS OF RAILWAY BARS, SCARFING TIMBERS, AND IMPROVED FASTENINGS: 
 
 ILLUSTRATED BY FOUR LARGE PLATES OF WORKING PLANS, 
 
 ACCOMPANIED BY EIGHTEEN ACCURATE ESTIMATES. 
 
 V 
 
 TOGETHER WITH 
 
 A COMPENDIOUS ACCOtJNT OP THE PROCEiSS OF KYANIZING, IN USE ON THE ENGLISH RAILWAYS, 
 FOR PRESERVING THE TIMBER FROM DECAY: 
 
 RECENT DISCOVERIES OF M. BOUCHERIE BY MEANS OF THE PYROLIGNATE OF IRON. 
 
 PRECEDED BY 
 
 PRACTICAL OBSERVATIONS ON THE DEFECTIVE NATURE OF THE RAILWAY STRUCTURES IN USE ; WITH AN 
 
 INVESTIGATION OP THE PRINCIPLES AND STRUCTURE ESSENTIAL TO THE STABILITY AND 
 
 PERMANENCE OF RAILWAYS, IN WHICH THE OPINIONS OF MEN EMINENT 
 
 IN SCIENCE AND' ENGINEERING ARE COLLATED. 
 
 BY JAMES HEREON. 
 
 CIVIL, ENGINEER. 
 
 "The author of a discovery has always to contend with those wliose interests it may affect, with the obstinate partisans of all that is old, with the 
 jealous and the envious. These classes combined, form, we are obliged to confeSfe, the greater part of the public. This compact mass of opponents, 
 time alone can separate and destroy; but time is not enough, they must be attaclced boldly, they must be attacked without ceasing; the means of action 
 must be varied, imitating the chemist, who is taught by experience, that the entire dissolution of certain alloys, requires the successive employment of 
 several acids." — M. Arago. 
 
 PHILADELPHIA: 
 CAREY & HART AND J. DOBSON. 
 
 
 E. G. DORSEY, PRINTER. 
 1841. 
 
if7^ 
 
 CONTENTS. 
 
 Preliminary Observations on tiie Defective nature of the Railway Structures in use, explanatory of 
 ous expenditure of money annually sunk in the repairs and renewals of railways, locomotive 
 and carriages, ........ 
 
 Mr. Stephenson's Method of Seating Stone Blocks, . . . . . 
 
 Description of the Great Western Railway Track, England, I. K. Brunei, Esq., Engineer, 
 
 Extracts from Mr. Brunei's Reports, . . . . . . 
 
 The Joints of the Rails are the weakest points in a railway, .... 
 
 Patent Trellis Railway Structure, represented on Plate I., invented by James Herron, C. E. 
 
 Patent Angular Scarfing, ....... 
 
 Cast-iron Socket-screws to prevent the oxydation of the screw-bolts, 
 
 Prize offered by the London and Birmingham Railway Co. for the best form of Rail and Chair, 
 
 Effects of Inequalities at the Joints of the Rails, . . . . . 
 
 Effects of Expansion on the Rails and Fastenings, ..... 
 
 Baltimore and Ohio Rail-road Report, extracts from it relating to the temperature in America, 
 
 Comments on the above, and on the heat which iron acquires when exposed to the sun. 
 
 Patent Wrought-iron Joint-chair for securely holding and evenly joining the ends of rails. 
 
 Cast-iron Joint-chair, ....... 
 
 Hook-headed Screw-bolts, .... 
 
 Trellis Track laid on the Baltimore and Susquehanna Rail-road, 
 
 Proper Elasticity of Track explained. 
 
 Fastenings removable above, . . • . . 
 
 Middle-chairs, how fastened to the bars, 
 
 Timber-screw Fastenings, . . » . . 
 
 Packing and ballasting of the track. 
 
 Crossing bogs, quick-sands, and other bad foundations. 
 
 Estimates of the Trellis Railway Structure, Plate I. 
 
 Notes on the cost of materials, ..... 
 
 Plate II., description of the Timber-pinned Trellis, 
 
 Improved mode of Joining the Ends of Plate Rails, 
 
 Improved Cast-iron Rails, ..... 
 
 Estimates of the Timber-pinned Trellis Structure, 
 
 Plate III., Oblique-angled Trellis Structure, 
 
 Estimates of the Trellis Structure, Plate III. ..... 
 
 Plate IV., description of the Primitive, the Diamond, and the Iron Trellis Structure, 
 
 Primitive Trellis Railway, ....... 
 
 Lateral or flange friction on bad tracks frequently greater than the weight of the carriage and load, 
 
 The strength of the parts of a railway track, their power of resistance doubled when arranged and 
 ed as in the trellis track, ....... 
 
 Diamond Trellis Railway Structure, ...... 
 
 The Iron Trellis laid in Asphaltum Cement, ...... 
 
 Formation of a Permanent Road-bed, ...... 
 
 Estimates of the Primitive, the Diamond, and the Iron Trellis Railway Structure, Plate IV. 
 
 Summary of the Estimated Cost of all the Trellis Tracks, 
 
 On the Preservation of Railway Timber from Rotting, . . 
 
 M. Boucherie's Process for Preserving Timber by Means of the Pyrolignate of Iron, 
 
 A statement of the bearing surface per mile of track by which several noted railvyay tracks rest on 
 lasting, compared with the trellis tracks, ..... 
 
 the ruin- 
 
 
 engines 
 
 3 
 
 
 • % 
 
 
 . 10 
 
 . 
 
 13 
 
 
 13 
 
 . 
 
 14 
 
 
 . 15 
 
 . 
 
 16 
 
 
 . 17 
 
 ■. 
 
 17 
 
 
 . 18 
 
 . 
 
 19 
 
 
 . 21 
 
 . 
 
 22 
 
 
 . 23 
 
 ■ 
 
 %3 
 . 24 
 
 . 
 
 24 
 
 
 25 
 
 
 25 
 
 
 26 
 
 , 
 
 26 
 
 
 27 
 
 . ■ 
 
 27 
 
 
 . 29 
 
 . 
 
 33 
 
 
 S3 
 
 . 
 
 35 
 
 
 38 
 
 . 
 
 40 
 
 
 42 
 
 . 
 
 42 
 
 
 42 
 
 combin- 
 
 
 
 43 
 
 , 
 
 44 
 
 
 44 
 
 , 
 
 46 
 
 
 46 
 
 , 
 
 48 
 
 
 49 
 
 
 57 
 
 the bal- 
 
 58 
 
 GENERAL ADDRESS OF THE PATENTEE, 
 
 BY LETTER, POST PAID, 
 
 TO JAMES HERRON, 
 
 CIVIL ENGINEER, 
 
 PHILADELPHIA. 
 
 Entered according to the Act of Congress, in the year 1841, by James Herron, in the Clerk's OfBce of the District Court for the Eastern District of 
 
 Pennsylvania. 
 
 (^^ 
 
 <,1')> 
 
'mi 
 
 .;.'5j''»: 
 
 
 t'.'Al.iKj 
 
 rj-inei 
 

 ./«? 
 
 HERRON'S PATENT RAIL-WAY 
 
 *ii:. 
 
 
 STRUCTURE 
 
 .^^" t 
 
 ii' 
 
 :jT^ 
 
 .vi; 
 
 y m- 
 
 Piil;-.,. 
 
 
 s^ 
 
 .ftrM^^^ 
 
 -,■■ ,ii;V- 
 
 
 
 
 
 ^'*'>?2^^'^*iS^1V*'S'i?^:-??V /v\ 
 
 
 ^mmmm 
 
 ; .^r '■Mi'-:-^'-':^ 'HiVt:^';;J^':H:^fe^^ 
 
 
 -Si ni'ij ittrcr 
 
 IKijL 
 
 JSI 
 
 "^ ® 
 
 r3~i: 
 
 ^4 
 
 -ir 
 
^i^.i?^:'-: 
 
 
 ■^M^^i- 
 
 
 :;;\S''iv^; 
 
 
 ■'■;.'^ '■!:&■ 
 
 / 
 
 Sin 
 
-SFsmirr m, mri i« iin /"« 
 
 J'lato H- 
 
 HERRON'S PATENT RAILWAY STRUCTURE. 
 
 
 2* 
 
 /,,/ ; /'/,/// „/■ liiu /, /' 
 
 
 
 
 
 
 
 '■•^v>''k;>"S^^-'' * I' 
 
 W ,/ 
 
 rS?;«^^/?;AJvlJ' 
 
 r 1> xSiSfK^s; 
 
 Wi^: 
 
 
 ■.^¥11 
 
 
 
 
 
 
 
 
 
 i¥i^ 
 
 Fiv- S Suit lYrir 
 
 
 |i 
 
 =^E=^==^ 
 
 
 ,y ; A '" 
 
 yx" 
 
 37^w^^ 
 
 "75- Qradt 
 
 w^p.f^^^s:?^?f^^^i^m^^^^^^^^-f: 
 
EI.LIS 
 
 , ® 
 
■ f>a'!>- 10 &lli i^'an ''*'' *< *"" '«""■' '^ """^ H//:' art liiili' Ihr rnll .riir. 
 
 ?iW7»-r TRELLIS RAILWAY STRUCTUREi 
 
 Cit'il Engi'neej: 
 
 //./ /:' 
 
 
 
 ^H ' 
 
 K 
 
 "' ' 
 
 J 
 
 II 
 
 ' 
 
 /•Jii/nitri/ fy ./ Yeatjfr 
 
il 
 
Fi. 
 
 r/a/e Jr. 
 
 
 »!:»4 
 
 pi' H ' 
 
 SIRIK.®!^ 
 
 Mi 
 
 m 
 
 ^t 
 
 m 
 
 f ,1 
 
 *^F '->^ ieag&r 
 
PRACTICAL DESCRIPTION 
 
 OF 
 
 H E R R O N ^ S 
 
 PATENT TRELLIS RAILWAY STRUCTURE, 
 
 EMBRACING THE MOST APPROVED MODIFICATIONS; 
 
 ALSO, 
 
 THE PATENT WROUGHT IRON RAILWAY CHAIRS, NEW AND IMPROVED MODE OF JOINING 
 THE ENDS OF RAILWAY BARS, SCARFING TIMBERS, AND IMPROVED FASTENINGS: 
 
 ILLUSTRATED BY FOUR LARGE PLATES OF WORKING PLANS, 
 
 ACCOMPANIED BY EIGHTEEN ACCURATE ESTIMATES. 
 TOGETHER WITH 
 
 A COMPENDIOUS ACCOUNT OF THE PROCESS OF KYANIZING, IN USE ON THE ENGLISH RAILWAYS, 
 FOR PRESERVING THE TIMBER FROM DECAY: 
 
 ' AND THE 
 
 REGENT DISCOVERIES OF M. BOUCHERIE BY MEANS OF THE PYROLIGNITE OF IRON. 
 
 PRECEDED BY 
 
 PRACTICAL OBSERVATIONS ON THE DEFECTIVE NATURE OP THE RAILWAY STRUCTURES IN USE ; WITH AN 
 INVESTIGATION OF THE PRINCIPLES AND STRUCTURE ESSENTIAL TO THE STABILITY AND, 
 , PERMANENCE OF RAILWAYS, IN WHICH THE OPINIONS OF MEN EMINENT 
 
 IN SCIENCE AND ENGINEERING ARE COLLATED, 
 
 /j . BY JAMES HEREON. 
 
 ^ -f ^ '^ CIVIL ENGINEER. 
 
 " The author of a discovery has always to contend with those whose interests it may affect, with the obstinate partisans i 
 jealous and the envious. Tliese classes combined, form, we are obliged to confess, the greater part of the public. This compact mass of opponents, 
 time alone can separate and destroy; but time is not enough, they must be attacked boldly, they must be attacked without ceasing; the means of action 
 must be varied, imitating the chemist, who is taught by experience, that the entire dissolution of certain alloys, requires the successive employment of 
 several acids." — M. Arago. 
 
 PHILADELPHIA: 
 
 E. G. DORSEY, PRINTER, LIBRARY STREET. 
 1841. 
 
^■^ fir 
 
 
 ' iW 
 
 '4 
 
-V /) 
 
 <lJi 
 
 4i 
 
 h 
 
 ^^-y^' 
 
 
PEELIMINARY OBSERVATIONS ON THE USUAL MODES OF CONSTRUCTION, AND 
 MOST CONSIDERABLE DEFECTS IN RAILWAYS. 
 
 Railways, in connection with steam engine machinery, are vast mechanical enterprises 
 that have, within a few years, been extensively applied to the purpose of effecting inland 
 transportation with unprecedented rapidity, certainty, and great comparative economy. The 
 railway itself, however, is believed to be of considerable antiquity, although of a very rude 
 construction, being first used only for temporary purposes. Railways of a more permanent 
 character, it is certain, have been in use, in the mining districts of England, for at least two 
 hundred years. On these, the heavy minerals were slowly transported, either by manual 
 labour, horse-power, or by gravity; and, in some few cases, slow-motioned locomotive steam- 
 engines were in use for several years anterior to their application to commercial transportation. 
 It was the favourable results obtained with these engines on the mineral railways, together 
 with the establishment of the very important facts, that the resistance from friction did not 
 increase with an increase of velocity, and that the adhesion of the engine wheels on the 
 cast iron rails, then alone in use, was adequate to the traction, that first led to their projection 
 and successful application in the rapid transportation of persons and property. 
 
 After the introduction of the steam engine, it was perceived that animal moters were no 
 longer applicable; the feet of the horses, by throwing gravel and dirt on the rails, obstructed 
 the motion of the carriages, and greatly increased the wear of the railway; nor was their 
 speed at all commensurate with the close calculations now made of the value of time, and 
 interest on capital, so that the animals were in many cases literally driven to death. 
 
 The steam engine was, therefore, almost exclusively looked to, although it is said to have 
 possessed only one-fourteenth of the power of the engines now in use. The projection of 
 the Liverpool and Manchester railway, which was the first grand application of machinery to 
 rapid transportation, opened a new era in railways and locomotive machinery. This railway 
 was, however, actually built before it had been determined whether locomotive or fixed steam 
 engines should be used — a circumstance that may seem quite unimportant, until we consider 
 that a locomotive engine in drawing itself and load, by the adhesion of its wheels, exerts a 
 strain on the rails and their fastenings somewhat similar to the strain that would be thrown 
 on a chain, or rope, by which a fixed steam-engine would draw a load equal to the locomotive- 
 engine and its train at the same velocity. Hence we find the slender fastenings in use are 
 torn loose, and the rails "driven" endwise, particularly down the grades, and the sills, or 
 blocks, are displaced by the same cause. In addition to which, it is well known, that great 
 injury results to the railway, as at present constructed, from the great additional weight of 
 the engines over that of the cars. 
 
 The Directors of the Liverpool and Manchester railway, being unable to decide on the 
 
4 OBSERVATIONS ON THE USUAL MODES 
 
 motive power, called to their aid two of the most experienced engineers then in England, who, 
 after a thorough examination into the working of both systems, as then in existence, decided 
 in favour of the greater economy of the fixed steam-engines, while the Company's engineers 
 maintained the superiority of locomotive. The question was finally decided in favour of the 
 locomotives, in consequence of the splendid improvements made in their structure, as exhibited 
 for the first time at the celebrated Liverpool contest, held in 1829, for the prize of 500/. 
 sterling, offered by the Directors of the Company "for the most improved Locomotive engine, 
 that would draw three times its own weight at the rate of ten miles an hour," — a task which 
 was not only readily performed, but even far surpassed by, what was then considered, the 
 astonishing powers and speed of the Rocket engine; to which was awarded the prize. This 
 engine, including fuel and water, weighed only 41 tons, empty 3^ tons, yet its evaporating 
 surface was three times that of the old engines, which weighed more than 7^ tons; its greatest 
 speed in the experiments was 29| miles an hour, and average velocity 14i miles, moving a 
 gross load of 17 tons at these rates on a level railway. It was subsequently improved by Mr. 
 G. Stephenson, the inventor, by increasing the draught, and drew a load of 40 tons at 13J 
 miles an hour, the steam not exceeding 50 lbs. on the inch. 
 
 "It was soon after found, that by constructing engines of greater size, the increased 
 evaporating powers would make ample amends for the additional weight, and a strong desire 
 was accordingly manifested of having heavier engines on the Liverpool and Manchester 
 Railway; but owing to the rails not being sufficiently strong to carry them, [rolled iron edge 
 rails, 35 pounds per yai'd on 3 feet bearings,] they were found objectionable, and there was 
 accordingly a constant struggle between light and heavy engines for some time, but the line 
 being now relaid with heavier rails, [50 and 60 lbs. to the yard,] engines weighing from 9 to 
 13 tons are exclusively used." 
 
 And eventually at the present day, we find that on the Great Western Railway, extending 
 from London to Bristol, locomotive engines weighing from 14 to 18 tons are used at the rate 
 of 40, and in some cases nearly as high as 60 miles an hour. This railway being built 
 entirely of timber, which had to be imported from America. 
 
 There was evidently, in the previous instance, but little, if any, unity of design between the 
 locomotive engines and the railway on which they were to work, and as the engines were 
 increased in size, the innate weakness, and other radical defects of the iron railway resting 
 on isolated supports, was rendered more palpable. Accordingly we find that the annual 
 repairs of the Liverpool and Manchester Railway, exclusive of new rails and labour in laying 
 them, averaged per mile 422/. 13s. 6d. 
 
 Locomotive engines, though enlarged, and very much improved in all their details, 
 proportions, and power, continue to be constructed on the same general principles as the 
 "Rocket" engine. Owing to the bad and unsuitable railways on which these engines are 
 worked, however, their steam joints are soon strained, and other parts of the machine are 
 bruised, bent, and eventually broken by the endless repetition of jars and jolts to which they 
 are subjected; for the rough road re-acts on the locomotive machinery, and thus both engine 
 and road are damaged by the concussion. The engine being, however, much the more 
 complex machine of the two, is damaged in the greater degree; hence we find that the expense 
 
OF CONSTRUCTING RAILWAYS. 5 
 
 of repairs on locomotive machinery, always exceeds the cost of repairs on the railway, usually 
 from fifty to a hundred per cent. 
 
 The superstructure of other railways, in England and elsewhere, has, until recently, 
 continued to be constructed on the same erroneous principles as those of the Liverpool and 
 Manchester; so that, as will hereafter be shown, a total change in the principles of construction 
 was necessary before any considerable improvement could be effected. On this railway the 
 iron rails rest in cast iron pedestals, or "chairs," which are, except across Chat Moss, seated 
 on square stone blocks: the settling of a single block will, therefore, occasion a serious, and 
 perhaps fatal defect in the track, by which many lives may be lost; and as this cannot be 
 discovered by the eye until the rail is actually bent, or broken, which would probably be 
 coincident with an accident, men are kept constantly walking over the hne, each one with a 
 crow-bar in his hands, with which he strikes on every block he passes, "sounding" it, as it is 
 termed, to ascertain if it be hanging to the rail. But even were all the stone blocks firmly 
 and evenly seated on the road bed, still the iron rails will spring, or deflect, between the points 
 of support, and the latter form rigid summits, against which the wheels impinge, or bound over, 
 when in rapid motion. The whole line is, therefore, a series of short elastic planes, divided 
 only by the rigid points of support; this gives to the engines and carriages that ruinous 
 undulatory or bounding motion; and, as the deflexure of the rails is more considerable next 
 to the joints, a lurch is added to the bound, which thus results in an awkward wabbling 
 gait, that very soon destroys the engines, carriages, and railway, causing the enormous 
 amount of annual repairs that absorbs, in so many cases, the whole income of the Railway. 
 
 With us in America, the stone block system of construction, and the still more erroneous 
 one of laying longitudinal sills of granite, plated with flat bars of iron, have been very 
 generally abandoned, and transverse sills of timber used in their stead. These cross sills rest, 
 in many cases, on the natural earth, or are bedded on broken stone, which is laid under them 
 in little trenches, or small square pits. It is found that the sills thus bedded soon settle very 
 unequally; those under the joints of the rails being (according to a natural law) pressed into 
 the road-bed by double, or rather quadruple the weight the other sills have to sustain. Farther, 
 rain water collects in the interstices of the stone, softening the earth beneath, and, as the 
 weight and momentum of the engines, &c., press alone on the ends of the sills, a slight 
 inequality in the track, tending to throw the weight more or less on the one side or the other, 
 will cause the sills to rock on the ballasting, or earth, that is necessarily under the middle of 
 each; thus, cavities are formed beneath the ends of the sills, into which they sink on the 
 passage of a train, and by the natural elasticity of the timber, as quickly recover their former 
 position when relieved of the load, so that this great defect can scarcely be discovered, except 
 in wet weather, when the water may be seen spurting from beneath them. 
 
 To remedy this defect, mud-sills have been introduced in the construction of most tracks. 
 These are longitudinal planks, three or four inches in thickness, and from eight to twelve in 
 width, one of which is placed on each side of the track, under the cross-sills, beneath the part 
 on which the rails rest; on these, the cross-sills w^ere made rest in some cases exclusively, but 
 
Q OBSERVATIONS ON THE USUAL MODES 
 
 this had soon to be abandoned, and the track filled in, to give it stability. The mud-sills 
 necessarily have joinings, where they are found to settle very unequally, the end of one 
 
 sinking or rising above the other, no scarfing having been hitherto devised that would prevent it; 
 from this, and other causes, they soon separate from the cross-sills, notwithstanding the 
 strong oak pin which is usually driven through the end of each cross-sill into the mud-sill; 
 but passing, as it does, through the sills at right angles, or nearly so, to their surfaces, it is 
 readily drawn out. The mud-sills in use, are, however, otherwise inadequate in strength and 
 much too limited in the extent of bearing surface, to sustain the weight and momentum of 
 heavy engines, which experience has shown can alone be profitably employed. And to 
 maintain the track in tolerable order, men have to be kept constantly at work raising up the 
 mud-sills, and ramming ballasting under them; but this being attended with considerable 
 labour, is usually done in a very imperfect manner, long hollow spaces being left, into which 
 the sill deflects, or sinks again on the passage of a train. To avoid this evil, resort is had to 
 the fatal expedient of driving wedges between the sills, which soon results in a total dislocation 
 of the frail fabric, after a very limited use. The unequal settling of the mud-sills is found 
 to take place even when the track has been carefully laid on a well prepared and costly bed 
 of broken stone; — costly, as I believe, in more senses than one, for it would seem that the iron 
 and other materials of the track were, in many cases, literally crushed by the lurching of the 
 engines, and ground between the wheels and this rubble stone bed. 
 
 That there is obvious necessity for very extensive improvements in the structure of 
 railways, few who have at all considered the subject, will attempt to deny. But should the 
 uninformed require a proof more cogent than the very large amount of annual repairs, a short 
 walk on any one of our best constructed railways, that has been two or three years in use, and 
 an attentive observation of the track, during the passage of the trains, will afford them very 
 ample occular demonstration. It will be seen that the joints of the rails sink under the 
 wheels, while their opposite extremities act with a leverage to prize up and tear loose the 
 fastenings, and it is thus that the hook-headed spikes are drawn out; which indeed serve but 
 to keep the cross-sills loosely attached to the iron rails, rather than to combine the structure. 
 When an engine approaches a joint, that end of the rail on which the wheel rolls at the 
 instant will be depressed, while the next rail in continuation exerts its elastic force to regain 
 its former position, or to start upwards; and this not only occasions an injurious jolt, but, 
 should one of the cast iron chairs, with which the ends of the bars are usually secured, break, 
 the most fatal results may follow, as has already been the case on some melancholy occasions. 
 
 By standing on the track, and looking along the line, it will be seen whether the rails are 
 regular in direction, without any of those short bends, or zigzags, which so frequently cause 
 the engines and cars to run off the track, and occasion resort to be made to the awkwai-d 
 and unsightly expedient of propping the tottering structure, as we would a common ruin. 
 
 It may next be seen whether the ends of the iron rails be evenly joined, particularly on the 
 inner side, so that the flanges of the wheels may not catch on them; and that their top surfaces 
 
OF CONSTRUCTING RAILWAYS. 7 
 
 are in the same plane, otherwise a severe and injurious joU will be felt on the passage of the 
 wheels. 
 
 It may also be observed whether the rails be not generally sunken at the joints, the bars 
 having bent, or taken a permanent set, which cannot be removed without ripping them up and 
 resorting to an anvil to straighten them. 
 
 While looking on the track, it may be well to bear in mind, that these sinks in the surface, 
 and bends in the direction, are not only the pregnant cause of many accidents, but that they 
 in fact defeat, in a great measure, the object for which the railway is designed — that of forming 
 an eve7i, smooth, and a hard elastic surface for the loheels to roll on; by which great loads can be 
 hauled at very small expense. 
 
 Without a most thorough combination of the parts, we cannot have evenness of surface or 
 general elasticity. And as my reasons for considering the latter quality an essential requisite 
 in a good railway, may not be generally understood, I will explain that I mean such general 
 elasticity in the whole structure of the track, that while it may not bend in any very sensible 
 degree under the insistent weight of the engines and carriages, it will nevertheless spring 
 under their momentum. Otherwise, we have the hard and rigid surface of the granite track, 
 with the rapid wear and breakage of machinery, and also of the railway; or a feeble and 
 crippled structure, totally void of recuperative energy.* 
 
 Mr. Stephenson attempted to remedy that ruinous defect, found so troublesome and costly 
 on the Liverpool and Manchester Railway, the continued settling and lateral displacement of 
 the stone blocks, by having blocks of granite cut two feet square, and dressed to an accurate 
 thickness of one foot. Each block consequently weighed about 660 pounds. These pon- 
 derous masses he had raised by a spring pole, and let fall with a force computed to equal 
 two tons; so that each block was made to descend like the ram of a pile-engine on its destined 
 bed, formed of finely broken stone, the interstices having been previously filled with fine 
 gravel and sand, which was also thrown under the block, from time to time, during the 
 operation; and this ramming was continued until no farther impression could be produced by 
 the impact, and the block was considered firmly seated. It was computed that the force 
 employed to seat the block, was greater than could possibly be produced by any action from 
 the engines. 
 
 * When the "eternal granite" track was first proposed, 1831,1 was engaged on the surveys for the Lynchburg and 
 New River Rail-road, under the direction of C. Crozet, Esq., principal engineer of Virginia, and I doubt not he will 
 remember the homely comparison I drew, — "that its projectors might profit by the experience of the blacksmith, 
 whose anvil they would find was never placed on a stone." Why, let me ask, does he prefer the wooden block? And 
 why has it been found necessary to place the anvil block of the tilt mill on an elastic timber frame? Is it not, also, a 
 well known fact, that the vibratory motion of a fixed steam-engine will shake down the most solid masonry when it is 
 connected with the walls of a building, and that they are now universally erected on an elastic timber frame, uncon- 
 nected with the walls. If such effects as these result from the action of the steady fixed engines, is it at all surprising 
 that the rapid locomotive, with its ponderous mass and momentum, should so soon destroy itself, and the rigid track 
 on which it bounds? 
 
g OBSERVATIONS ON THE USUAL MODES 
 
 N. Wood, Esq., Civil Engineer, in the third London edition of his "Practical Treatise on 
 Rail-roads," page 712, remarks, after describing the above method of construction — 
 
 "We must, however, in this place, observe, that great care is requisite to form a firm and 
 solid seat for the block; the sand must be so spread amongst the interstices of the broken 
 coating, as to make the block bear equally upon its entire base; otherwise, it will rest on the 
 points of the broken pieces of coating; and when once seated, the block must not, in the least 
 degree, be disturbed in laying down and keying the rails. On embankments, not perfectly 
 consolidated, this plan of seating the blocks is rendered abortive by the least shrinking; 
 and, consequently, on such ground, transverse wooden sleepers and longitudinal wooden rails 
 are, generally, laid down, until the embankment becomes perfectly consolidated. When blocks 
 are properly set, and the base of the railway perfectly solid, this plan forms a very firm and 
 stable line of railway; yet, when the heavy engines pass over the blocks, we find, in practice, 
 they yield or compress the coating, to a certain extent; and we likewise find that the lurching 
 or vibration of the carriages tends to displace the blocks from their seat, and that manual 
 labour is still required, by forcing ashes or sand underneath, to keep them at their proper level. 
 We have deemed these observations on the principles of the difierent modes previously 
 practised of laying down railways, necessary, in order that we may show the defects which it 
 is the object of Mr. Brunei's plan to remedy." 
 
 Seeing the utter impossibility of maintaining a regularity of line, and uniformity of surface, 
 while the rails rested on isolated bearings, Mr. Brunei (notwithstanding the great scarcity and 
 high price of timber in England) has, in the construction of the Great Western Railway, 
 resorted to the use of large beams of timber, which are placed longitudinally under each line of 
 rails, so as to have a continuous bearing on the soil, and afford, in turn, a continuous support 
 to the railway bars. 
 
 Timber is, from its natural elasticity, better suited to bear without injury the shocks and 
 jolts of heavy machinery than rocks or other rigid materials. And, we accordingly find that 
 Mr. Brunei's example has been extensively followed, not only in laying timber on the new 
 railways, but by the removal of the stone blocks and substitution of timber on some of the 
 principal railways in Great Britain and Ireland. 
 
 A thorough knowledge of the practical results obtained from the very extensive and 
 repeated experiments made during the last ten years, both in Europe and America, in the 
 formation of the superstructure of railways, with the effects resulting from the constant action 
 of locomotive engines thereon, are the only certain means by which we can draw correct 
 conclusions to guide us in devising suitable remedies that will obviate the many existing 
 defects. For, it may with truth and propriety be remarked, that all the railway tracks hitherto 
 laid, have not only been mere experiments on a large scale, as witness the many re-construc- 
 tions in so short a period, but there is not, even of those recently laid, a single track in 
 existence that is not, in some respects, an acknowledged failure. 
 
 As Mr. Brunei's great work has proved in some respects to be the nearest approach to 
 perfection yet made, an attentive consideration of the principles upon which he formed his 
 plan, taken from his own reasoning on the subject, together with the important feature in 
 which it has failed, and which he has subsequently abandoned, (without, it would seem, 
 providing any adequate substitute,) will enable me, thus, from practical experience, to establish 
 
OF CONSTRUCTING RAILWAYS. 9 
 
 the correctness of the reasoning on which I have based my proposed improvements in railway 
 structure. I accordingly subjoin a description of his plan, with extracts taken from his 
 reports. 
 
 DESCRIPTION OF THE GREAT WESTERN RAILWAY, ENGLAND. 
 
 I. K. BRUNEL, Esa., ENGINEER. 
 
 This railway has two tracks, which are united in one frame-work by transoms, usually 
 placed fifteen feet apart; in juxtaposition with these transoms, and a little on the outside of 
 the centre of each track, beech piles, 10 inches in diameter, are driven by a pile-engine, to a 
 depth of 8 or 10 feet in the cuttings. And on embankments, the piles are driven quite through 
 the bank, and five or six feet into the natural soil, so that some of them are more than 36 feet 
 long. Every second pair of piles are united by two transoms, of American pine, 6 inches 
 wide and 7 inches deep, which, as stated above, extend across both tracks, and are let into 
 the heads of the piles, one on each side, to which they are also firmly bolted. The interme- 
 diate pair of piles are united by a single cross-tie, or transom, 6 inches wide and 9 inches 
 deep, let into the piles, and bolted as the others. The double timbers are laid 13 inches, and 
 the single timbers 9 inches, below the line of the rails. On these transoms, string-pieces of pine 
 30 feet long, 14 inches wide, and 9 inches deep, are laid on their flat side; so as to be 7 feet 2i 
 inches wide, from centre to centre, for each track; and 6 feet between the two tracks. All 
 the joints of the longitudinal timbers are placed in one line, across the road, over the double 
 transoms — the double transoms having a piece of timber inserted between them, and pinned 
 wherever the string-pieces cross them; on this a joint-piece, about 2-j feet long, is laid, on which 
 the longitudinal timbers abut, the end of each being secured by a single screw-bolt | of an 
 inch in diameter. This screw-bolt passes through the end of the string-piece, through the 
 joint-piece, and through one of the transoms. The middle transom has suitable notches cut 
 in it, 2 inches deep, to receive the longitudinal timbers, each of which is there secured by a 
 single f inch screw-bolt. So that a string piece ^0 feet long, is secured by only 3 bolts of | inch 
 iron, and a notch 2 inches deep in the middle transom. 
 
 The timber frame-work being thus formed, finely screened gravel is rammed under the 
 longitudinal timbers and cross-ties; and this is continued by men driving opposite to each 
 other until the string-pieces are sprung upwards J or f of an inch; the top surface of the 
 longitudinal timbers is then adzed, or planed down to one uniform surface. A plank of hard 
 wood, American elm, oak, or ash, about 1 1 inches thick and 8 inches broad, is well nailed on 
 the top of the string-piece, the surface having previously been well coated with tar; the heads 
 of the nails, which are in double rows, at 2^ and 5 inches apart, are well punched in, and the 
 top surface of the hard wood is then dressed to a slope inwards of 1 in 20. 
 
 A bridge section iron rail, with a very broad base, weighing 44 pounds per ya,rd, is laid on 
 the strip of hard wood, a piece of felt being interposed between the base of the rail and. the 
 2 
 
10 DESCRIPTION OF THE 
 
 wood. The rail is then fastened to the hard wood and string-piece by very- large wood 
 screws passing through the base of the rail, also through the hard wood into the string-piece. 
 The iron rails are 15 feet long, and their joints are placed in one line across the tracks 
 without the use of chairs, the base of the rail being 6 inches wide. 
 
 The whole of the timber is Kyaiiized; the quantity used in the double track per mile of the 
 part finished, exclusive of that in the piles, has been 292,072 feet, board measure, or 24,339J 
 cubic feetj put together with six tons of iron bolts, and 30,000 wood screws, also nails and 
 wooden tree-nails, &c. 
 
 Mr. Brunei calculates that he throws a pressure upwards, by the packing of the ballasting, 
 equal to about one ton to the foot lineal of the string-pieces. This calculation gives a vertical 
 pressure on each pile equal to about 30 tons, without taking into consideration the pressure 
 on the cross-ties; which estimate must have much over-rated the actual pressure, as according 
 to it there would have been a strain of 15 tons on the single bolt, f inch in diameter, that holds 
 down the middle of each string piece; a force more than sufficient to tear in two a bar of the 
 best iron of the same section as the bolt. Notwithstanding the great care thus taken with the 
 packing, the ballasting has been so compressed by the action of the engines, that the string- 
 pieces ceased to rest on it, and the piles were found to be the points of support, and have since 
 been abandoned. 
 
 The following extracts from Mr. Brunei's reports, give the reasoning which induced his 
 plan, with his subsequent opinions on the results. 
 
 "The peculiarity of the plan which has been adopted, consists principally in two points; 
 first in the use of a light flat rail, secured to timber, and supported over its entire surface, 
 instead of a deep heavy rail, supported only at intervals, and depending on its own rigidity. 
 Secondly, in the timbers which form the support of this rail being secured and held down to 
 the ground, so that the hardness and degree of resistance of the surface upon which the 
 timbers rest, may be increased by ramming to an almost unhmited extent. 
 
 "The first, namely, the simple application of rails on longitudinal timbers, is not new; 
 indeed, as mentioned in a former report, I believe it is the oldest form of railway in England; 
 but when lately revived and tried upon several different railways, it has not, I think, succeeded 
 as fully as was anticipated, and I believe this is very much owing to the want of some such 
 means as that which I have adopted for obtaining a solid and equal resistance under every 
 part of the timber, and a constant close contact between the timber and the ground. As I 
 believe this to be entirely new, and to constitute an essential part of the plan, I trust I shall 
 be excused dwelling upon it for the purpose of fully explaining it. 
 
 "In all the present systems of rail-laying, the supports, whether of stone blocks or wooden 
 sleepers, simply rest upon the ground, and, consequently, only press upon the ground with a 
 pressure due to their own weight; this is trifling, compared either to the weight which rolls 
 over them, or even to the stiflriess of the rail which is secured to them. The block or sleeper 
 must lie loosely upon the ground; if you attempt to pack under it beyond a certain degree, 
 you will only raise it, and, for the same reason, it is impossible to pack under the whole 
 surface of a block or sleeper; one corner or one end is unavoidably packed a little more than 
 another, and from that moment the block or sleeper is hollow elsewhere. If this block yield 
 as the weight rolls over, the rail itself, resting on two contiguous supports, is sufficiently stiflf 
 to raise it again, and the support becomes still more hollow; such is the operation, which may 
 frequently be observed by the eye, more or less, in the best laid railways. 
 
GREAT WESTERN RAILWAY, ENGLAND. n 
 
 "Where continuous longitudinal sleepers have been tried, they have also been laid loose 
 upon the ground; having no vs'eight in themselves, their length has rendered it impossible that 
 they should be well supported by the ground underneath, or that they should continue so, even 
 if it were practicable to lay them well in the first instance. 
 
 "It will be perceived at once that two lumps, or two hard places in the ground may leave 
 such a timber unsupported for the interval of twenty or thirty feet in length, and, under the 
 weight of an engine running rapidly over, it must, in such a case, yield and spring from the 
 ground. 
 
 "In the present plan these timbers, which are much more substantial than those 
 hitherto tried, are held down at short intervals of fifteen to eighteen feet, so that they 
 cannot be raised; gravel or sand is then rammed and beat under them, until at every point a 
 solid resistance is created, more than sufficient to bear the greatest load that can come upon 
 it; as the load rolls over, consequently, the ground cannot yield — the timber, which is held 
 tight to the ground, cannot yield, neither can it spring up as the weight leaves it; and, if the 
 rail be securely fixed every where in close contact with the timber, that also is immoveable: 
 such was the theory of the plan, and the result of the experiment has fully confirmed my 
 expectation of success." 
 
 Such were Mr. Brunei's views, as presented in his report of the 22nd of January, 1838, to 
 the Directors of the Great Western Railway. 
 
 At a subsequent meeting (Oct. 10th, 1838,) of the Great Western Railway proprietors, Mr. 
 Brunei, in his report, expresses himself as follows: — 
 
 "The mode of laying the rails is the next point which I shall consider. It may appear 
 strange that I should again, in this case, disclaim having attempted anything perfectly new, yet 
 regard to truth compels me to do so. I have recommended, in the case of the Great Western, 
 the principle of a continuous bearing of timber under the rail, instead of isolated supports, — 
 an old system recently revived, and as such I described it in my report of January, 1838. 
 The result of many hundred miles laid in this manner in America, and of some detached 
 portions of railway in England, w^ere quite sufficient to prove that the system was attended 
 with many advantages; but since we first adopted it, these proofs have been multiplied; there 
 need now be no apprehension. There are railways in full work, upon which the experiment 
 has been tried sufficiently to prove beyond doubt, to those willing to be convinced, that a per- 
 manent way in continuous bearings of wood may be constructed, in which the motions will be 
 much smoother, the noise less, and consequently — for they are effects produced by the same 
 cause — the wear and tear of the machinery much less; such a plan is certainly best adapted 
 to high speeds, and this is the system recommended by me and adopted on our road. There 
 are, no doubt, different modes of construction, and that which I have adopted as an improve- 
 ment upon others, may, on the contrary, be attended with disadvantages. For the system I will 
 strenuously contend, but I should be sorry to enter with any such determined feeling, into a discus- 
 sion of the merits of the particular mode of construction. I would refer to my last report, for the 
 reasons which influenced me, and the objects I had in view in introducing the piling; that part 
 which had been made under my own eye, answered fully all my expectations 
 
 "It may be inferred that I am disposed still to defend the system of piling. I certainly could 
 not abandon it from a conviction of its inefficiency, for I see proofs of the contrary; and I feel 
 that, under similar circumstances, I could now prevent the mischief that has occurred. Upon 
 that portion of the line where the permanent way must next be formed, piling could not be 
 resorted to, the ground being a solid hard chalk for many miles. I had intended, however, to 
 recommend the same principle, but in a different form, holding down the longitudinals by small 
 iron rods driven into the chalk; but the same objection could not exist, because the chalk 
 cannot yield under the timbers like clay or even gravel. But I should wish most anxiously to 
 
12 DESCRIPTION OF THE 
 
 avoid anything like obstinate adherence to a plan, if tlie object I believe essential can be obtained 
 by any other means; particularly when that plan, being my own, I may be somewhat prejudiced 
 in its favour. I find that the system of piling involves considerable expense in the first 
 construction, and requires, perhaps, too great perfection in the whole workj and that if the 
 whole, or a part of this cost, were expended in increased scantling of timber and weight of metal, 
 that a very solid continuous rail would be formed; for this as a principle, as for the width 
 of gauge, I am prepared to contend, and to stand or fall by it, beheving it to be a most essen- 
 tial improvement where high speeds are to be obtained. I strongly urge upon you not to 
 hesitate upon these two main points, which, combined with what may be termed the natural 
 advantages of the line, will eventually secure to you a superiority which, under other circum- 
 stances, cannot be attained. 
 
 "As regards the expense of forming the permanent way on this principle, I am quite prepared 
 to maintain what I have on a former occasion advanced, that even on the system which Ave 
 have adopted between London and Maidenhead, the total cost does not materially exceed that 
 of a well-constructed line with stone blocks. I did not make in the outset an exact estimate 
 of the total cost of either mode. I was unable to obtain the cost which has actually been 
 incurred on other lines; but a comparative estimate was made, and the result of that compa- 
 rison led me to state that the one might exceed the other by 500/. a mile. The actual cost of 
 our permanent way appears, by the detailed account which has been made out, to have been 
 above 9000/., [which, at $4 85 to the pound sterling, is $43,650 per mile of double track, or 
 $21,825 per mile of single track piled,] including expenses of under-draining and forming the 
 surface, which cannot be included in the cost given in other cases, because that drainage, 
 although, I believe, generally forming a part of the plan, is not yet constructed: this sum 
 includes the sidings at the stations, switches, points, and other contingencies, and also the 
 expenses incurred during the first month of working the line, and which, as I have before 
 stated, consisted in removing and replacing work which had been improperly executed. These 
 items will make a considerable reduction; and, besides these, large reductions may be effected 
 in parts of the work which were new; and, from the circumstances naturally attending a first 
 attempt, were not so economically conducted as they might be, or indeed, as they were towards 
 the close, of the works, when the different parts were let by contract. Taking the prices at 
 which the work was latterly actually executed, 8000/. per mile would be a liberal allowance for 
 our future proceedings, even adopting the same system: and, with a modified system, such as 
 that suggested, of simple longitudinal bearers of large scantling, and rail of 54 pounds per yard, 
 at the present high price of iron, the cost calculated upon our actual past expenditure would 
 not exceed 7400/. per mile. [= $35,890 for the double track, or it will be $17,945 for a single 
 track.] This, I am aware, is a larger sum than that which has usually been assumed as the 
 cost of the permanent way. I cannot prove that others have cost more, or even so much as 
 this, as I have nothing but the published accounts to refer to; but this I can state, and prove 
 if necessary, that rails and blocks, such as are now being adopted on the Manchester and Liverpool 
 Railway, would, upo7i our line, cost at least as much. 
 
 "The prime cost of rails and chairs, delivered on the Hne, would alone amount to half the 
 money; and nothing is pei'haps more certain, than that the experience of other lines, within the 
 last two or three years, has proved that this part of the construction of a railway is unavoid- 
 ably much more expensive than was ever calculated for, at the time our estimates were made. 
 
 "I am, gentlemen. 
 
 Your obedient servant, 
 
 I. K. BRUNEL." 
 
 We see, in the case of the Great Western Railway, that a beam of timber 30 feet long, 14 
 inches wide, and 9 inches thick, strengthened by the close union of a strip of hard wood and 
 an iron rail of 44 lbs. per yard, bolted to piles at every 15 feet, and having the ballasting 
 
GREAT WESTERN RAILWAY, ENGLAND. 13 
 
 rammed hard under it, had not sufficient extent of bearing surface on the soil; as we find that, 
 notwithstanding the great strength of the combined rail of timber and iron, the packed coating 
 of the road was every where compressed under the action of the passing loads. How then is 
 it to be expected, that a simple mud-sill, of three or four inches in thickness and eight in width, 
 not even united at the ends, but loosely, if at all, attached to the cross-ties, and simply resting 
 on the ballasting or soil, can possibly maintain an even surface on one of our frail railways? 
 
 While Mr. Brunei adhered to the use of piles, the ^om^s of the string-pieces and rails were 
 at least well supported, and all lateral deviation of the track effectually prevented; what sub- 
 stitute he intends providing for these weak points, or how he will prevent the springing of the 
 rail from the ground, as before described, does not appear. 
 
 It has been found, universally, that the joints of the iron rails, and also those of the 
 longitudinal timbers, are the weakest points in railways as hitherto constructed. This weak- 
 ness is clearly indicated by the sinking and lateral deviation of the track at these points. 
 Were it possible, therefore, to construct railways without joints, their most serious defects 
 would be thereby obviated. But, as this is naturally impossible, recourse must be had to 
 mechanical ingenuity; so that, by a proper arrangement and combination of the parts, the 
 joints may be made of equal, or, to be on the safe side, of greater strength than the other 
 parts of the structure. This, however, has been a long sought desideratum. 
 
 I have been induced to discuss, thus . fully, the defective nature of the railway structures in 
 use, — which experience has every where shown to be the cause of the enormous amount 
 annually absorbed in repairs, — in order that the object and necessity of the improvements I 
 am about to describe may be more clearly understood. 
 
 PLATE I. 
 
 REPRESENTS THE PATENT TRELLIS RAILWAY STRUCTURE, 
 
 DEVISED BY JAMES HERRON, CIVIL ENGINEER. 
 
 For the purpose of designating the principal modifications of this new method of railway 
 structure, I shall term the Plan, Fig. 1, a " Right-angled bolted Trellis." Here the trellis 
 foundation of the track is represented as combined with two of the most approved forms of 
 rolled iron rails now in use, by means of my patent wrought-iron joint-chairs, and a new 
 description of screw-bolt fastenings. 
 
 On the side B D, of the plan of track. Fig. 1, rolled iron rails of the "bridge" section are 
 represented as secured to the diagonal truss by bolts that are cut in a suitable manner for 
 screwing into wood, which, passing through the string-piece d, are thus screwed into the 
 diagonal sills a and h, as shown in the cross-sections, Figs. 3 and 4. And, on the side A C, 
 the broad base i rails are represented as secured by bolts that pass through the string-piece 
 d, and, also, through the diagonal sills a and 6, being screwed into cast-iron socket-screws 
 s, s, below the structure. Fig. 2 is a longitudinal side view of the lower side, A C, of the 
 
14 PATENT TRELLIS RAILWAY STRUCTURE. 
 
 Plan, Fig. 1. Fig. 3, a cross-section at A B; and Fig. 4, a cross-section at C D of the other 
 two figures. The same letters designate the same parts in the four figures. It will of course 
 be understood that the two descriptions of rails and fastenings are introduced on the same 
 plan for the purpose of showing that either may be used. 
 
 CONSTRUCTION. 
 
 On open soils, that permit the free percolation of water, it will be sufficient that the embank- 
 ments are properly consohdated, and the road-bed evenly graded. On the surface so formed, 
 plank, or timbers, a c a, of any suitable dimensions, say 14 feet 9 inches long, 3 inches in 
 thickness, and 8 inches in width, are to be laid obliquely on the centre line of each track, at 
 5. feet apart, from centre to centre, measured along the line. This series of oblique sills are 
 crossed by other timberSj b c b, of the same dimensions, which are simply laid upon the former 
 sills, without notching, in a direction that is also oblique to the centre hne, and nearly at right- 
 angles with the first series. These sills intersect each other on the centre line of the track at 
 c, also under the two lines of string-pieces and rails, d and r, which it will be seen are laid on 
 the diagonals of the latticed sills, the ends of which I prefer lapping on each other outside of 
 the rails at a b; one sill in the upper series, b cb, will, therefore, rest on five sills of the lower 
 series, a c a. The total width of this trellis foundation will be 10 feet 8 inches, when the 
 width of gauge between the rails is 4 feet 85 inches. The middle intersections, c c, and the 
 outside laps of the sills at a b, are each secured by two wooden pins,^:>^, driven nearly at 
 right angles to each other, or in the form of the letter V. 
 
 The string-pieces, d d,6zc., on which the iron rails have a continuous bearing, should be of one 
 uniform length for all the straight lines, and of the same length for the outside of the curves; 
 which, inclusive of an allowance on each for scarfing, should be equal to the expanded, or 
 summer, length of the iron rails. The latter, in the present design, are 20 feet long; and the 
 string-pieces, in the rough, 20 feet 8 inches long. 
 
 There was no scarfing, or method of joining the ends of timbers, hitherto known in carpentry, 
 that would answer well for this purpose. I, therefore, devised the angular scarf, shown open 
 on Plate II. Fig. 5. The end of the string-piece A, is cut down to half its depth at any 
 required angle, other than a right-angle, say 45°, which will be in the direction a b; the piece 
 is then turned over and cut down at the same angle, which, on being again turned up, will be in 
 the direction c d, and consequently crossing the direction of the former cut. The end of the 
 piece B, is cut at the same angle and in the same manner; the cuts in it will, therefore, be in 
 the direction a' b', c' d'. On bringing the two pieces together, the angular lap c a, of A, will 
 rest on the angular lap c' a', of B; and the angular lap b' d', of B, will rest on its equivalent, 
 b d, of A. So that they will be interlocked, and all deviation, either vertically or laterally, 
 is efiectually prevented, while the strength of the timber is retained. And the whole can be 
 cut, with great ease and accuracj^, by a common mitre. 
 
 The string-pieces, d d, &c., in the present design, Plate I, are 8 inches Made and 5 inches 
 
PATENT TRELLIS RAILWAY STRUCTURE. 15 
 
 thick; being scarfed as described, they are laid on the flat side, so as to rest on five intersec- 
 tions of the latticed sills; the joinings, T, of the string-pieces will, therefore, rest on an 
 intersection of the latticed sills; and those on the one side of the track I prefer placing oppo- 
 site to the middle, S, of each string-piece on the other side, as represented. 
 
 Almost any of the iron rails in general use may be advantageously laid on my timber 
 structure, but I vs^ould prefer a rail intended for a continuous bearing, either of the "bridge 
 section," or the broad base J,, as shown in the plan. 
 
 The bolt fastenings that I have devised for securing heavy iron rails to the trellis substruc- 
 ture, are made perform the double purpose of effectual fastenings for the iron rails, and 
 combining bolts to the timber truss; by which means very great strength of structure is obtained 
 with much economy in the cost. The plan of track on Plate I. shows two systems of fastenings, 
 those on the one side B D, of the track. Figs. 1, 3 and 4, are all secured from above; and 
 those on the other side A C, are all secured below. I shall first describe the latter. But, in 
 either case, I place the middle of the iron rail, being its strongest and stifFest part, over the 
 scarfing of the string-pieces. Two | inch bolt-holes are formed in the base of the rail, one 
 on each side of the middle of its length, in the present case, at about 6 inches from each 
 other, C, Fig. 2; these holes are placed close to the middle web of the rail, one on each side 
 of it; so that the upper end of the bolts may be under the top table of the rail, and thus be out 
 of the way of the flanges of the wheels, T, Fig. 4. Or they might be made to pass up into the 
 middle web of the rail. For convenience in construction, the upper end of each bolt has a 
 suitable hole to receive a key, and the latter has a notch cut in the top edge to keep it from 
 working out by the vibrations of the rail, as shown half size on Plate III. Fig. IS. Both 
 bolts pass through the base of the iron rail, one of them,- through the contiguous end of 
 each string-piece, d d, and both of them through the two diagonal sills, a and b, below which 
 they are secured either by a key or screw-nut. The screw-nut I consider more efficient than 
 the key, but as it is liable to rust in such situations, and cannot then be tightened or removed, 
 I have devised a cast iron nut, or "socket-screw," so as to cover the end of the bolt, as repre- 
 sented on a large scale, Plate III. Figs. 19, 19\ being external views; Fig. 20 the face, and 
 Fig. 21 a vertical section. There is a suitable cavity, VV, in the nut, Fig. 21, for the purpose of 
 containing a little bitumen, native turpentine, or tar; and, also, a countersunk cavity, X, in the 
 face, z, Fig. 20, of the nut for the same purpose; the face of the nut, z, being screwed up tightly 
 against the under side of the timber, will effectually exclude all moisture, and thus prevent the 
 oxidation of the screw. 
 
 The iron rail is thus permanently held at its middle point, and, together with the diagonal 
 sills, become a part of the splice of the string-pieces; so that the iron rail, two of the string- 
 pieces, and two diagonal sills are securely bolted together at this one point, usually one of the 
 weakest, but now the strongest in the structure. In addition to which, the crossed or diagonal 
 sills have more than three times the extent of bearing surface on the soil, at these points, that 
 is obtained by common transverse sills. 
 
 The above arrangements being completed at each scarfing, T T, &:c., Plate I, the ends of 
 
15 PATENT TRELLIS RAILWAY STRUCTURE. 
 
 the iron rails will of course meet on the middle of each string-piece, at S, S, &cc. And, it should 
 be remembered, when cutting the string-pieces, to allow in proportion to the temperature for 
 the expansion, in the present case of 20 feet, of the iron bars. 
 
 We have seen that the joints of the iron rails are the weakest points in a railway — this 
 fatal defect having hitherto baffled the numerous attempts of the most skilful engineers in 
 devising an adequate remedy. In the modification of my plan under consideration, the ends 
 of the iron rails, near S, rest on the middle of the string-piece, d d, over an intersection of 
 the diagonal sills, a and b; so that the general support of this point is amply provided for. 
 Should the end of the rails be let rest simply on the timber, they will very soon be pressed 
 into it by the weight of the passing loads; it is, therefore, found to be necessary to insert 
 beneath them a plate of metal of sufficient extent. It is also necessary to hold the ends of 
 the bars securely without the possibility of their escaping, or even working loose, yet by snch 
 means as will most readily permit the contractile and expansive motion of the bars. Beside 
 which there should be some simple provision by which the workmen can readily join the abut- 
 ting ends of the bars in an even manner on the flange side of the rails, and, also, on the top 
 surface; otherwise, a very severe and injurious jolt will be felt on the passage of the wheels. 
 And the whole of this should be accomplished without any large projections, against which 
 the wheels might strike, in case a car were thrown off" the rails, and thereby cause a dreadful 
 concussion, breakage of the cars, and probably loss of life. 
 
 The little part of a railway, that should perform these important functions, is known as a 
 "pedestal," or more commonly a "chair," in the present case a "joint-chair." 
 
 "The Board of Directors of the London and Birmingham Railway Company, desirous of 
 carrying on the great work in which they are engaged on the most scientific principles; and, 
 if possible, to avoid the enormous cost of repairs which has attended some large works of 
 a similar description, offered, by public advertisement, a prize of one hundred guineas 'for 
 the most improved construction of railway bars, chairs, and pedestals, and for the best manner 
 of affixing and connecting the rail, chair, and block to each other, so as to avoid the defects 
 which are felt more or less on all railways hitherto constructed;' stating, that their object was 
 to obtain, with reference to the great momentum of the masses to be moved by locomotive- 
 steam-engines on the railway, 
 
 "1. 'The strongest and most economical form of rail. 
 
 "2. 'The best construction of chair. 
 
 "3. 'The best mode of connecting the rail and chair; and also the latter to the stone 
 blocks or wooden sleepers. And that the railway bars were not to weigh less than 50 pounds 
 per single lineal yard.' 
 
 "In consequence of this advertisement, a number of plans, models, and descriptions were 
 deposited with the company within the time limited by the advertisement; and others were 
 received afterwards, which although not entitled to the prize, were still eligible to be considered 
 with reference to their adoption for trial." 
 
 Professor Barlow, from whose valuable work on the strength of Materials and Construc- 
 tion, page 310, the above extracts are taken, states that the directors appointed J. U. 
 Rastric, Esq., of Birmingham, N. Wood, Esq., of Newcastle, civil engineers, and himself, to 
 examine and report upon the same, with a view to awarding the prize, &c., which they 
 
PATENT TRELLIS RAILWAY STRUCTURE. 17 
 
 did accordingly. It seems, however, that though many of the plans offered showed great 
 ingenuity and information, none of them were of a description that they could recommend 
 to be employed. So Mr. Rastric stated before a committee of the British Parliament. 
 
 The directors next engaged Professor Barlow to make a series of experiments on the best 
 form of rails, modes of joining them, &c., and to ascertain '■'•whether permanently fixing the rail 
 to the chair, for which there were several plans, would be safe in practice,'''' In his Report on the sub- 
 ject, he thus urges the necessity of evenly joining the abutting ends of the rails: — (page 373.) 
 
 "I have, above, alluded to the gauging the ends of the rails and openings in the joint- 
 chairs, and I have also spoken, in the description of my experiments, of the advantage of 
 keeping the blocks of the two lines of rails parallel. On all these points, it is probable, I shall 
 be considered by many as entering into refinements neither called for nor practicable, in the 
 case of railways; but, I would ask, why is it found so much breakage takes place, and that so 
 many repairs are rendered necessary? There is no theoretical reason why a heavy load, 
 passing with great velocity, should cause more damage than the same load passing slowly, if 
 the road were perfect; the mischief, therefore, is in the imperfect practical execution, and the 
 disregarding of small inequalities, as we would disregard them in common cases. It has, 
 perhaps, never occurred to such persons, that the difference of level at a joint-chair, between 
 the two abutting rails of only tV of an inch will, when the carriage is moving from a higher to 
 a lower level, at its greatest speed, cause the wheel to pass the distance of a foot without 
 pressing on the rail, and consequently throwing the whole weight, which ought to be borne 
 equally by the two rails, wholly upon one; yet this is a fact, resting on a natural law, and cannot 
 be otherwise. To fall roth of an inch, by the action of gravity, requires Tith part of a second, 
 and in that time the carriage will have advanced a foot; consequently, for that space, the whole 
 weight has been borne by one rail only." 
 
 Commenting on the tensile strain produced by the contraction of the bars, he says: — 
 
 "The question is now satisfactorily answered. We have seen that, with about ten tons per 
 inch, a bar of ii'on is stretched the ToVoth part of its length, and its elasticity wholly excited 
 or surpassed. Again, admitting 76° to be the extreme range of the thermometer in this 
 country, [England,] between summer and winter, it appears, from the very accurate experi- 
 ments of Professor Daniel!,* that a bar of malleable iron will contract with this change 
 2oVoth part of its length. And hence it follows that if the rails were permanently fixed to the 
 chair in the summer, the contraction in the winter would bring a strain of five tons per inch 
 upon the bar, and a strain of twenty-five tons upon the chair, (the bar being supposed of five inch 
 section,) thereby deducting from the iron more than, or full half, its strength, and submitting 
 the chair to a strain very likely to destroy it. Every proposition, therefore, for permanently 
 attaching the rail to the chair is wholly inadmissible. 
 
 "These remarks may also be carried farther, for, if it be dangerous to attach the rail directly 
 to the chair, it must be bad in practice to affix it indirectly by wedges, cotters, or otherwise, 
 beyond what is absolutely essential to give it steadiness under the passing load. 
 
 '•'•'The 'problem, therefore, luhich engineers have to solve, is '■To find a mode affixing the rail to the 
 chair, which shall give sufficient steadiness to the former; but which, at the same time, shall produce 
 the least possible resistance to the natural expansion and contraction of the bar.'' 
 
 "The quantity of motion which thus takes place is certainly but small, viz., about rrth of 
 an inch, between summer and winter, with a fif\een-foot bar; but the force of contraction is 
 great, amounting to five tons per sectional inch for the annual extremes, and frequently to not 
 less than two and a half tons, between noon and night of our summer season, while the whole 
 power of iron, within the limits of its elasticity, does not exceed nine or ten tons. 
 
 * See Philosophical Transactions of 1831. 
 
X8 PATENT TRELLIS RAILWAY STRUCTURE. 
 
 "This is an important consideration, and, for want of attention to it, or rather in conse- 
 quence of its amount not having been ascertained, a practice of wedging or fixing the rails 
 has prevailed, which must necessarily have been the cause of great destruction to the bars, 
 
 "I would also suggest here, as a matter deserving the attention of practical men, that, as 
 the bar must necessarily contract, it will draw from that side which is least firmly fixed, and 
 hence all the shortening will most probably be exhibited at one end, however slight the hold on 
 either may be; and, when it happens that the adjacent end of two bars both yield, the space 
 between the two is rendered double that which is necessary. To avoid this evil, one of the 
 two middle chairs in each bar might be permanently attached to the rail, in which case the 
 contraction must necessarily be made from each end, and the space occasioned by the short- 
 ening of the bars would then be uniform throughout, and much unnecessary and injurious 
 concussion thus saved both to 'the rail and to the carriage." 
 
 The Directors of the Baltimore and Ohio Rail-road Company published a report of their 
 engineers, dated January 8th, 1838, on the principal rail-roads in the Northern and Middle 
 States, and upon a railway structure for a new track on the Baltimore and Ohio Rail-road, 
 &c. This report quotes from Professor Barlow so much of the foregoing as relates to the 
 tensile strain, and goes on to say: — 
 
 "In estimating the value of what is contained in the foregoing extract, in relation to this 
 highly interesting subject, we must not lose sight of the fact that the climate of England is more 
 uniform in temperature than that of the United States. The extreme range in England, as 
 indicated by Fahrenheit's thermometer, is taken, as above, at 76°; whereas at Baltimore it is 
 from 40 to 50 per cent, greater." 
 
 After stating their data, they go on to say: — 
 
 "// is with extremes, hoivever, that we have to do in this connection; and we must observe that in 
 the interval between 1825 and 1836, the mercury fell to — 10, at Baltimore, and to — 16 at 
 several other places in and about the same latitude on both sides of the Alleghanies where 
 observations were had. The highest temperature in the shade may perhaps he taken at 100°. 
 And thus we obtain 116° as the extreme range in a period of many years. 
 
 "Now the change of length in a bar of malleable iron, according to Daniell, as quoted 
 by Barlow, is the xraVooth part for 1° of heat. Consequently it would be the aoVoth part of 
 the length for 76° in England, as stated in the report above quoted; but it would be ttsVcto 
 xll6 = the TiVoth part in the United States near the latitude of 40°. 
 
 "In a bar of 18 feet, or 216 inches long, therefore, the extreme variation in length would 
 be, in England, -108, or about the ^th of an inch, and in the United States -165, or the ith of 
 an inch; the latter being fully 50 per cent, greater than the former. It follows, therefore, that 
 with us, the extreme force exerted in the contraction of the bar, from a change of atmospheric 
 temperature, will be 7^ tons to the square inch, or 37^ tons in a bar of 5 inches section, being 
 the size of good rails generally. And, it is very evident, that no railway fastenings, likely to 
 be employed, will be able to withstand the action of anything like such a force. Consequently, 
 the motives for leaving the rail free to slide in the fastenings, in its advancing and receding in 
 length, from changes of heat and cold, are much stronger here than in England. In justice to 
 what has been done in this country, it is proper to state, that, in laying down the rails upon 
 the Baltimore and Ohio, as well as all other rail-roads in America, as is believed, provision 
 was made, in the extent of opening allowed at the joining of the bars, for changes in length, 
 to be caused by variations in temperature. It is probable, however, that this precaution 
 has not been so scientifically, or strictly, attended to, as would have been best; and that the 
 force of contraction, with the mischiefs thence arising in the absence of due provision for it, 
 have not been appreciated; as is evident from the employment of keys, in imitation of the 
 
PATENT TRELLIS RAILWAY STRUCTURE. 
 
 19 
 
 same in the English structures, and of which we have already spoken in detailing the plan of 
 
 construction upon the several rail-roads visited." 
 
 "The permanent attachment in the middle of the bar, rather than at one end, has this 
 further advantage, that whilst in the latter case the loose end of the bar would move the i^th 
 of an inch between the extremes of heat and cold in this climate, in the former case the 
 distance moved at the end (both ends participating alike in the movement) would be only the 
 i of this, or the tV of an inch. This at once lessens, by one-half, all danger of violence to the 
 fixings." 
 
 These extracts show how fully the engineers of the Baltimore and Ohio Rail-road Company 
 concur in the scientific deductions of Professor Barlow. Yet there is one point on which I 
 must differ from the Report; this, however, simply relates to the temperature that should be 
 taken for the purpose of ascertaining the expansion of railway bars. I have itahcised the 
 extracts where, it seems to me, the conclusions arrived at are not in accordance with the 
 premises laid down. 
 
 The report goes on to remark, "// is with extremes that we have to do in this connection.'''' Yet 
 the highest temperature in the shade is taken, whereas, we know that railway bars are exposed 
 to the direct rays of the summer's sun; acting, moreover, on the surface of the road, and 
 reflected, perhaps, from the sides of a rock-cut. Should not the thermometer, therefore, be 
 exposed to the sun in a similar situation? Or, more correctly, should not the bulb of the ther- 
 mometer be placed in juxtaposition with the railway bar, when under such circumstances, that 
 it may acquire the same degree of heat — as it is evident, that iron, when exposed to the sun, 
 and particularly while in connection with the earth, in rocky and other dry. situations, where 
 there is but little evaporation, acquires a much higher degree of heat than the surrounding 
 atmosphere? Now, it is the heat which the iron bar acquires, when thus exposed, that we have 
 to ascertain; and we use the thermometer to measure that heat. Or, still more simply, our 
 object being to ascertain how much an iron bar expands in length when exposed to the sum- 
 mer's sun, in a railway, and it being rather inconvenient to measure it directly, we place the 
 bulb of a thermometer in close contact with the bar, that it may acquire the same degree of 
 heat, knowing, by the experiments of Daniell, that the metal mercury in the tube of the 
 thermometer will expand one division, or degree, of the graduated scale, with the same addition 
 of heat that expands an iron bar the xyzWoth part of its length. Taking, therefore, the extreme 
 range of the thermometer, from the lowest depression by cold, to its greatest elevation in the 
 above circumstances, and multiplying the whole number of degrees by the rs^Wth part, we 
 have the difference in length of the iron bar, expressed in parts of its length. 
 
 What degree of heat railway bars acquire, in the above circumstances, I have not at present 
 the means of ascertaining, but have reason to believe that it will be found much above what 
 has been assigned. The thermometer, in our summer's sun, frequently rises to 130°. Mr. J. 
 H. Alexander, Topographical Engineer to the State of Maryland, informed me that he once 
 placed a thermometer in the sun which rose to 130°, after which it burst, being graduated no 
 higher; and that he also applied a thermometer, at about two o'clock in the day, to a mass 
 of iron ore, which had been exposed to the sun during the summer; this ore, he states, con- 
 tained about forty per cent, of metal, and notwithstanding the sun was carefully shaded from 
 
20 PATENT TRELLIS RAILWAY STRUCTURE. 
 
 the mass and thermometer, during the experiment, yet the latter rose to 145°. Pure iron 
 would, undoubtedly, raise it still higher. Taking this experiment, however, 145 + 16 := 161°, 
 as the extreme range; this will occasion a difference in length equal to the -gTTth part, which, 
 in a 20 foot bar, will be a little more than i of an inch. This contraction of the bar, if both 
 ends were made fast in the summer, would produce a tensile strain, at the extreme cold, equal 
 to 10-^ tons on the sectional inch; being 52J tons on a bar of 5 inches section; which is a 
 strain exceeding the elastic strength of iron.* 
 
 * For the purpose of ascertaining whether the opinion I had expressed last autumn, and reasoned on above, in 
 regard to the heat which I believed iron would acquire when exposed to the sun, was correct in point of fact, I nnade 
 the following experiments, at the Franklin Institute, in the city of Philadelphia, on the 11th, and again on the 24th 
 of June, 1841; the intermediate days being generally unfavourable. 
 
 The result, it will be seen, fully sustains my views, and, I hope, may induce some more competent person, who has 
 the facilities for doing so, to experiment, on a large scale, in connection with the direct expansion of long railway 
 bars, and to continue them regularly, through the summer and winter, for one or two years. 
 
 Not having at hand a piece of wrought iron with a suitable hole in it, I used an old cast-iron heater, of a russet- 
 brown colour, and cylindrical form, about 6^ inches long, and 21 in diameter, having af inch hole in the axis, about 
 4 inches deep. I also used a small cast-iron vase, about 7 inches high, turned tolerably smooth, and nearly black in 
 colour; this had a hole passing through the bottom, which I stopped with a cork, turning it bottom upwards. 
 
 I left both these on the elevated platform above the roof of the Institute on the 10th June. On the 11th, mercury 
 to the depth of an inch was poured into the hole in each iron, and let stand about half an hour to recover its tem- 
 perature. 
 
 There was a strong, cool breeze blowing at the time, which caused some fluctuations in the temperature. 
 
 The thermometer, when first placed in the shade of a chimney, sheltered in a great measure from the wind, and 
 exposed to the reflected heat from the roof, rose to 90°; being plunged into the mercury in the hole of the iron vase, 
 it rose immediately to 98i°; and in the cast-iron heater to 102°. On being hung up in the sun it fell immediately to 
 98°; removed to the shade of the chimney it fell to 92°; the temperature in the rooms below being 88°. At half- 
 past two o'clock, both the iron vase and heater indicated, on several trials with different instruments, a heat of 110°; 
 and, on the thermometer being placed in the sun, against a post much sheltered from the wind, (as I was apprehensive 
 of its being broken,) and exposed to reflected heat, fell immediately to 101°; at which it always continued until near 
 3 P. M., when it rose, in a lull of the breeze, to 105J-°, but the irons indicated no change of temperature. 
 
 I then removed the iron vase and heater to the window on Seventh street, over the principal entrance. This 
 window has a western exposure, and being open, together with the doors and windows of the library, there was a 
 strong current of air passing through it, but not so strong as on the roof. 
 
 The thermometer I selected to continue the experiments with, fell, in the room, to 88°. At 3 P. M. the iron vase 
 indicated 110°, and the heater iron 110|. 
 
 On the 24th June, at 2 P. M., I found the thermometer, standing in the little front room, at 82°, at which it con- 
 tinued after being shook, the doors and windows being open. On plunging the bulb into the mercury in the hole of 
 the iron heater it rose to 104°, and fell, along side of it, in the sun, to 101°. 
 
 The mercury in the iron vase having been emptied out by some person, was replaced by cold mercury, and let 
 stand until 3 P. M. At this time the thermometer in the library, marked standard, was at 841°. The instrument I 
 was using fell in the shade to 86°; plunged in the mercury of the vase it rapidly rose to 111°; being placed in the 
 sun, with the vase, it fell to 100°. At half-past three, the heater iron indicated 110°, the vase 113°; ten minutes 
 after it had fallen in the sun to 106°. At 10 minutes past 4, the heater and vase both indicated 111°; and the sun 
 100°. At 6 P. M. the same thermometer, in the shade, had fallen to 86°; the heater iron indicated 110i°, and the 
 vase only 102°. The thermometer was several times passed from the one to the other, with precisely the same results. 
 
PATENT TRELLIS RAILWAY STRUCTURE. 
 
 21 
 
 We see how very important it is to the perfection and stabihty of a railway, that the above 
 described functions should be fully provided for in the joint-chair. I give the following prac- 
 tical solution of Professor Barlow's problem. 
 
 PATENT WROUGHT-IRON JOINT-CHAIR. 
 
 Plate III. Fig. 12, represents a flat plate of good rolled, or hammered, iron, which may be 
 i inch, I, or i, &c., in thickness, as the case may require. This plate has four cuts, i i i i, 
 &;c., made in it; and also two holes, c c, to receive the bolts for securing the chair on the 
 trellis, or sill. Being heated in the usual manner of working wrought iron, the parts I K and 
 J H, are bent up and formed so as to embrace the base of the rail, with a part extending 
 upwards against the middle web of the rail, to afford it support, in case the j, rail. Fig. 13, be 
 used; but if the "bridge section" rail. Fig. 15, be used, the latter part is omitted. Should 
 there be any discrepancy in the size of the railway bars, by making the space in the chair rather 
 wider than the base of the rails, leaving small openings at e, e, the workman, when putting 
 the track together, will be enabled to join the abutting ends of the rails evenly together, as 
 follows: he heats the chair red-hot, and, running in the ends of the two rails, he hammers 
 the parts I K, J H, to the right or left, until the bars join evenly, and the parts of the chair fit 
 closely to the rails; and, should one rail be higher than the other, he hammers on the higher 
 rail until the top surfaces ai-e even. This may be done for sharp curves, while the rails are in 
 their proper position on the string-pieces, by inserting a flat plate of iron, or even a common 
 crow-bar, to keep the chair from burning the wood; but on straight lines, and indeed on the 
 generality of curves, it will be better to have a suitable piece of timber for the purpose. The 
 chair, on cooling, contracts on the ends of the two rails, and holds them tightly, yet it is by an 
 elastic, or spring pressure, which will thus admit of the linear contraction and expansion of 
 the bars. And, as the chairs are made of the same metal as the rails, their rate of expansion, 
 and, consequently, their pressure on the rails, will continue uniformly the same at different 
 degrees of temperature; unaffected, also, by wet or dry weather. Being, moreover, made of 
 tough wrought iron, they will of course be much less liable to break from sudden jars than 
 cast-iron chairs. 
 
 . The base of the chair is let into the middle (S, Plate I.) of the string-piece d d, so that the 
 rails may have a continuous bearing on the timber; the middle part of the chair c c, (Plate 
 III.) extending quite across the string-piece; the whole is then secured by two | inch screw- 
 bolts, which pass through the holes c c into the string-piece, and through the two brace sills, 
 a and b, Plate I., below which they are, in the present case, secured by cast-iron screw-nuts, 
 s s, as before described, for the middle bolts, and shown on a large scale. Figs. 19, 19^, 20, 
 and 21. 
 
 The chairs I had made for the Baltimore and Susquehanna Railway were of h inch iron, 5 
 inches wide, and weighed less than 6 lbs. each (exactly 5.82 lbs.); the middle part, c c, for 
 
22 PATENT TRELLIS RAILWAY STRUCTURE. 
 
 bolt holes, Plate III. Figs. 12 and 13, was made 2 inches wide, and 8 inches long; and the 
 wings, J, H, I and K, that hold the rails, 1^ inches wide each. These dimensions, when the 
 rails rest on a continuous bearing, will generally be found ample, if we may judge by compari- 
 son, since, on many of our rail-roads, the ends of the bars have no better security than the 
 adhesion of a couple of spikes, driven into the timber. 
 
 These chairs will be much better, and more rapidly made, under a fly-press, than they can 
 possibly be by hand; and, as it makes the cuts i i, &c., it should, at the same time, bend the 
 limbs I, K, J and H about 45°, so that a second press will instantly complete the chair at one 
 heat. When made for j, T rails, instead of the cuts i i, &c., it will be better to make two cir- 
 cular cuts, as shown by the dotted lines, leaving the plate full out to the straight dotted lines, 
 so that the limbs, I J and K H, on each side of the chair, will be united at their upper part by 
 a bridge of metal that will extend across the joint of the rails, and thus strengthen the parts. 
 
 Such is the simple practical solution I prefer giving to Professor Barlow's problem. 
 
 I say prefer; for I first devised a cast-iron chair in which the ends of the rails are secured 
 by the pressure of a spring-headed screw-bolt, which thus allows of the contractile and expan- 
 sive motion of the bars; a model of which may be seen in the Patent Office. And, as this 
 chair admits of the rails being removed with rather more facility than the foregoing, it would 
 probably be preferred by those who, having formed their ideas from the imperfect railways in 
 use, look more to the ease with which the parts of the racked and half-worn structure may 
 be removed, than they do to the very great importance of deferring the necessity of such removal 
 by contrivances that will give more strength, durability, and fitness to the structure at the same 
 expense. Yet I do not by this mean to admit that there is the least practical difficulty in the 
 removal and replacing of my wrought-iron chair, or any other part of my railway; but it 
 may, and doubtless will, whenever it should be required, take more time to remove the parts, 
 than it would the parts of a railway which shakes to pieces in ordinary use. 
 
 Thus, any rail in a track secured by my wrought-iron chairs, can be removed with great 
 ease, by simply driving the chair end-wise on to one or other of the rails, the bolts that secure 
 the chair to the structure having been drawn out, and the chair raised from its seat, after which 
 there will of course be no difficulty in removing the rail. 
 
 I have shown that the middle of each iron rail is firmly bolted to the trellis, and in fact made 
 to form part of the scarfing of the string-pieces, at the points T, Plate I.; this being therefore a 
 fixed point in the bar, the contraction and expansion of the iron must necessarily take place from 
 the middle to the ends, and thus the motion of the ends be the least possible; these ends meeting 
 in the wrought-iron chairs at S, on the middle of the timber string-pieces d d. Thus the middle 
 of each iron rail, and the ends of two string-pieces, and the middle of each string-piece, and the 
 ends of two rails are alternately bolted together, and these hitherto weak points effectually 
 secured. The fewer joints there are the more perfect the railway would be, were it not for the 
 contraction of the bars; and the fewer chairs would be required. There is besides, however, 
 a practical limit to the length of the bar, from the difficulty of manufacturing it without increase 
 of expense; and also to the length at which the string-pieces can be conveniently obtained. 
 
PATENT TRELLIS RAILWAY STRUCTURE. 23 
 
 Rails are now generally made from 16 to 18 feet long. I have assumed 20 feet as a more 
 suitable length; and it is possible that they may be manufactured 25, or even 30 feet long, 
 vrithout materially increasing the expense. The contraction of a rail 30 feet long, vphen the 
 thermometer stands at — 16°, would only be | of an inch; and its weight, at 50 lbs. to the yard, 
 500 lbs.* They are now usually laid on the track with the top surface inclined inwards, to suit 
 the conical form of the wheels. There would be some considerable difficulty in most places 
 of obtaining a sufficient supply of string-pieces of the latter length, so that, all things con- 
 sidered, 20 feet would seem to be the most suitable length to calculate on at present. 
 
 With rails 20 feet long the middle and end fastenings will of course be 10 feet apart, so that 
 an intermediate fastening will be required at the points Q Q, &c., which will place the fasten- 
 ings at distances of 5 feet apart, from centre to centre. These intermediate fastenings may 
 be made with narrow chairs, which are formed in a similar manner to the joint-chairs. But, 
 as the system of fastenings we are at present considering is secured below the structure, we can 
 effect this object more simply, and at less expense, by means of two hook-headed screw-bolts, 
 (Fig. 17, Plate III.,) say | of an inch in diameter. These bolts, it will be seen, have a double 
 hook-head, the long one at s is made to bear on the base of the rail; while the short one, /, 
 extends farther downwards on the stem of the bolt, so that it may rest on and be pressed into 
 the string-piece (as shown by the dotted lines u li) before the long one, s, comes in contact 
 with the base of the rail. It thus adds greatly to the strength of the bolt, forming a buttress, 
 as it were, while it prevents an undue pressure from being placed on the rails. These bolts, 
 at Q, &c., Plate I., after passing through the string-piece, pass also through an intersection of 
 the two diagonal sills a and b; and, being secured by socket-screws below, like the other bolts, 
 serve likewise to combine or bolt the structure together. 
 
 The system of fastenings I have just described, is such as was used on the trellis track laid 
 by the Baltimore and Susquehanna Rail-road Company, beginning a little north of their Bel- 
 videre Depot. It will be seen that the iron rails are secured alone by screw-bolts of the above 
 description, without a single spike being used in the structure. The rails are such as could be 
 had at the time, and are only 16 feet long, instead of 18 or 20 feet; the intersections of the 
 latticed sills were, in consequence, placed at 4 feet from centre to centre, and the string-pieces 
 are reduced in size to 4x8 inches; they are of white pine, and were sprung edge-wise, without 
 difficulty, to conform to an 8° and also to a 12° curve. The chairs were made by a negro 
 blacksmith in the city, and, for want of a travelling forge, were driven on the rails cold. The 
 whole is laid in a thin bed of soft sand, resting on a springy loarii; and was constructed, under 
 many disadvantages, by a party of hands who are retained on the road for making repairs. 
 
 * The JQ rails now being imported for the Columbia Railway, are in bars 18 feet long, weighing 60 lbs. to the yard, 
 or 360 lbs. to the bar; and will cost, delivered in Philadelphia, a little over $50 a ton, say $55. A rail 21 feet long, 
 of 50 lbs. to the yard, which is ample, will weigh 333A lbs. A 21 foot bar 350 lbs. There is, in the first bars, 20 lbs. 
 of iron per yard of single track, or 15.71 tons per mile more than would be required to make a much stronger Irack on 
 my plan; which, at $55 per ton, amounts to $864' per mile; so that on a double track, of the length of this railway, 
 (82 miles,) exclusive of sidelings, &c., there would, at this rate, be an unnecessary expenditure of $141,696! 
 
24 PATENT TRELLIS RAILWAY STRUCTURE. 
 
 These were very rough workmen and labourers, accustomed to the old log-laying systems of 
 construction, without experience in carpentry, or suitable tools, such as I could point out, to 
 facilitate and perfect the execution of the work. 
 
 PROPER ELASTICITY OF TRACK. 
 
 The general elasticity of surface, without dislocating the joints, of which I have spoken 
 before, may be observed in the above track. This quality I estimate highly, on account of 
 the easy motion it gives to the carriages, and the very great saving it will occasion, by pre- 
 venting the breakage of the locomotive machineiy. The celebrated M'Adam, in his system 
 of road-making, requires precisely the same thing. He states that the small cubes of broken 
 stone bind themselves together by their angles so as to resemble a plank. (I use the plank 
 itself.) Farther, that he preferred the shaking surface of a bog, to make a durable road on, 
 to the solid surface of a rock; and that he always had the rock covered with earth before 
 laying on the "metal" surface. Experience has shown, however, that the cubes of stone do 
 not bind themselves together, but are pressed separately into the earth. And this has been 
 experienced on rail-roads as well as turnpikes. The degree of elasticity in my track will of 
 course depend very much on the nature of the soil, or road bed, and, also, on the dimensions 
 of the timbers and rails, compared with the passing loads. The greater the elasticit}^ the less 
 liability there will be to breakage of machinery, and the easier the motion of the carriages. 
 How far this should be carried is a nice question to determine between the loss of power it 
 might occasion, and the value of the machinery it might save. In the absence of experiments 
 to determine the question, I would recommend my track to be packed, and loaded with the 
 ballasting, so that it may not bend under the weight of an engine at rest, but will slightly so 
 when in motion. The reaction of the track, in this case, will prevent any loss of power 
 whatever, and there may still be sufficient elasticity to save the machinery. 
 
 FASTENINGS REMOVABLE ABOVE. 
 
 It has been objected that the system of fastenings, which I have just described, are inconve- 
 nient to get at when it may become necessary to tighten or remove them. This I conceive 
 to be an advantage, particularly near cities and villages, as it would prevent evil-disposed 
 persons from being able to remove a rail, and they will rarely require tightening until the track 
 is renewed. But it may be readily obviated, by making the bolts removable from above, for 
 which I have designed several systems of fastenings. 
 
 First system. Let the cast-iron nuts, or "socket-screws," (Plate III. Fig. 20,) be made square 
 at the collar, z, as shown by the dotted lines y y y y, with four points to take hold of the 
 wood and prevent its turning, when screwing in the bolt. All that is then required is that 
 the shank of the bolt be round and its head square, or octagon; so that a wrench, or turn- 
 
PATENT TRELLIS RAILWAY STRUCTURE. 35 
 
 screw may be used on it, to turn it in and out. The intermediate fastenings, Q Q, &c., may- 
 be made with two similar bolts, one side of the octagon head of each pressing on the base of 
 the rail, or by the use of narrow chairs. Figs. 24 or 25. 
 
 The middle fastenings of the bars, when there are socket-screws below, may also be made 
 with a narrow chair. Fig. 16. The space between the horns, m m, is less than the width of 
 the base of the rail; and they stand in the first instance upright, as represented; the dotted 
 lines, n n, Fig. 13, on the section of the rail, represent a notch cut on each side of the base 
 of the rail, into which the horns m m, Fig. 16, fit. The chair being heated red-hot, and the 
 rail inserted between the horns, m m, they are hammered down on the base of the rail. So 
 that this chair is permanently attached to the middle of the bar. The two-holes, c c, Fig. 24, 
 are for the bolts to pass through that secure the chair to the frame-work. When these 
 middle-chairs are used, I give the scarf of the string-pieces the form of a vertical half lap, as 
 shown on Plate IV., and also on the upper side of Fig. 1; the end of each lap is, however, 
 cut down at a small angle, so that the laps of the scarfing will be longer on the upper side 
 than they are below. Instead of cutting the ends of the laps down at an angle, the scarfing 
 will be more effectual for some purposes, if the difference in length of the top and bottom of 
 'the laps be effected by a square shoulder at about one-third of the depth of the piece from the 
 bottom, as represented on the drawings annexed to my patent. The middle-chair will thus 
 rest on, and be let into, a lap of each string-piece, and the chair-bolts will pass through the 
 end of each, also through the two oblique trellis-sills, a and b, into the socket-screws, s s. 
 
 Similar chairs may be used for the intermediate fastenings at Q; in which case the horns, 
 m m, are made wide enough apart to admit of the whole base of the rail passing freely between 
 them; being made of the proper form in the first instance, they are readily driven on the rail 
 from one end, which may be done with greater facility by moderately heating the chair to 
 expand it. 
 
 These chairs may be manufactured most readily by punching up the horns, m m, Fig. 16, from 
 the middle of a flat plate of iron. Fig. 24, Plate IV. 
 
 TIMBER SCREW-FASTENINGS. 
 
 The fastenings that I have hitherto described, are screwed into metal-nuts, or socket-screws; 
 but a much cheaper system of fastenings may be made by having five or six inches from the 
 point of each bolt, cut with a sharp deep thread, so as to be suitable for screwing into timber. 
 These timber screw-bolts will be two or three inches shorter than the former, which will be a 
 considerable saving, and the expense of the socket-screws will also be saved. The round part 
 of the bolt will pass through the string-piece, and the screw will be worked into the diagonal 
 sills, a and b. Previous to which the whole should be tightly clamped together. 
 
 These timber-screws may be used to secure the chairs, as at T, Fig. 3. Or they may be 
 made, at the middle of the rail, to pass obliquely through the base, by holes similar to the first 
 4 
 
26 PATENT TRELLIS RAILWAY STRUCTURE. 
 
 system of fastenings. The oblique direction is given to the bolt, so that it may readily be 
 screwed in from above, as the bolt at o in the cross-section Fig. 11, Plate III. 
 
 The above screws are nearly a foot long, and serve the double purpose of fastenings to the 
 iron rails and combining bolts to the structure. Similar screws, of only five or six inches in 
 length, may be used for securing the rails to the string-pieces, by having the latter secured to 
 the trellis sills, by means either of separate bolts or of wooden pins, jop, &c.. Fig. 6, Plate 
 II., which will be fully described hereafter. 
 
 The intermediate points, Q, Fig. 1, may be securely held, as before described, by one side 
 of the octagon head of the screw-bolt pressing on the base of the rail, for which purpose the 
 screws should, also, be entered obliquely. 
 
 PACKING AND BALLASTING. 
 
 The track being completed, with either description of fastenings, fine gravel, or coarse sand 
 ballasting, with a little clay on top to bond it, when these materials can be readily obtained, 
 should be well rammed and beat under the upper series of diagonal sills bob, and under the 
 string-pieces d d; filling in the track with the same materials as high as the base of the 
 iron rails. The ballasting will thus rest on the diagonal sills of the treUis work, and load the 
 track down in a regular manner, so that it may not spring from the earth, or rise in front of 
 the engines. Were a track in this climate to be bolted down to piles, like the Great Western 
 Railway, the frost in winter would heave it in the spaces between the piles, so that they would 
 be low points, if it did not, indeed, burst it loose from its fastenings; and in the thaw of spring 
 these spaces would subside below the "level," so that the piles would be high points; and I am 
 inclined to think that this has been more or less the case even in the mild chmate of England. 
 
 My object has been to make the I'ailway structure a perfect unit throughout, giving it a 
 uniform depth, and weaving it with the surface, or ballasting, in which it is imbedded, that they 
 may be identical, so that they may rise and subside together; otherwise, we must rest the 
 foundation of every thing within the limits of the track below the limits of frost, probably a 
 depth of three feet from the surface. 
 
 On soft clay lands, quicksands, bogs, and other bad foundations, where gravel or other good 
 ballasting cannot be conveniently obtained, the dimensions of the timbers may be increased in 
 proportion to the softness of the soil, which I am inclined to think will be found in every case 
 an ample compensation for the absence of open ballasting. Thus the string-pieces may be 
 made 7 by 9 inches, or even 8 by 10 inches section, and the diagonal sills 3 by 10 inches, or 
 3 by 12 inches section; or longitudinal mud-sills may be added, as shown by the dotted lines 
 e e, which will stiffen the structure in a vertical direction. And, on soft bogs, the structure 
 should be made rest on hurdles, like the track crossing Chat Moss. 
 
 When clay is used as packing, it is known to tend greatly to the preservation of the timber. 
 
ESTIMATES OF THE SCREW-BOLTED STRUCTURE. 27 
 
 and the rain water that falls between the rails, can be readily got rid of by a few cross-drains 
 formed under the track. 
 
 From the foregoing description, it will be perceived, that, when the railway is finished, no 
 part of the structure will be seen above the surface of the earth except the two lines of rails; 
 so that, were a carriage to run off the rails, there would be nothing for the wheels to strike 
 against, and they will continue rolling on the ballasting until stopped by the increased resist- 
 ance it offers; the rails serving as "guards" to keep the carriage from running off the embank- 
 ment, or road-way. Farther, all the cross-timbers being sunk below the surface are secured 
 against fire and the action of the atmosphere. And should there be a necessity for using 
 horses at any time there will be nothing to interfere with their feet. 
 
 ESTIMATES OF THE TRELLIS RAILWAY STRUCTURE.— PLATE I.* 
 
 Eslimate A. — Plate I. 
 
 Estimate of the quantity of materials, and probable cost of constructing one mile of the right-angled trellis 
 structure, the rails and timbers being united by screw-bolts having cast-iron socket-screws. The track being 
 laid with rolled iron edge rails, either of the "bridge" section, Fig. 15, or the broad base. Fig. 13, in bars 20 feet long, 
 weighing 50 lbs. to the yard. This track is 4 feet SJinches between the rails, and 10 feet 9 inches from out to out. 
 
 528 string-pieces, ((/ cZ,t) 20 ft. 8 in. long, 5x8 in. section, = . . 36,379 ft. b. m. 
 2112 trellis sills, [a c a and b c b,) 14 ft. 9 in. long, 3x8 in. section, == 62,304 " 
 
 Total quantity of timber per mile, 98,683 ft. at f 10$ m. $986 83 
 
 6336 white oak pins, (;>7),) 9 in. long, 11 in dia. at 5 ct. 31 68 
 
 528 wrought-iron joint-chairs at (S,) 6 lbs. ..... 3,168 lbs. 
 
 1056 chair bolts, I in. dia. at (S,) 1-6 lbs. = 1,689-6 lbs. 
 
 1056 scarf, or middle bolts, I in. diam. at (T,) 1-8 lbs. = . . 1,900-8 lbs. 
 
 2112 intermediate bolts at (Q,) 1-2 lbs. = . . . . . 2,534-4 lbs. 
 
 Wrought iron fastenings per mile, 9,292-8 lbs. at 9 cts.§ 836 35 
 
 2112 cast-iron socket-screws (s s s,) for the joint and scarf-bolts at 
 
 (S and T,) U lbs. = 2,640 lbs. 
 
 2112 cast-iron socket-screws (s at Q,) 1 lb. == . . . . 2,112 lbs. 
 
 Cast-iron socket-screw fastenings per mile, = ....... 4,752 lbs. at 6 cts.|| 285 12 
 
 Workmanship laying 1760 yards of track at 45 cts.ll 792 00 
 
 Cost of trellis frame and iron fastenings per mile = . . . . . . . . . . $2,931 98 
 
 785 tons of rolled iron rails (r r,) in bars 20 feet long, weighing 50 lbs. to the yard, at 50** a ton, . $3,925 00 
 
 Cost per mile with unprepared timber, ............ $6,856 98 
 
 Mineralising 8287 cubic feet of timber and treenails, with corrosive sublimate (Kyanizing,) at 4i cts.f f 372 92 
 
 Cost of one mile complete, A — I, . . . $7,229 90 
 
 JVotes to Estimate A. — Plate I. 
 
 * This plan is designated, in the Report of the Committee of Science of the Franklin Institute, as No. 4. 
 
 t These letters refer to the drawings, to designate the parts named. 
 
 :): Persons residing in large cities, accustomed to the high prices which the finer qualities of heart pine timber 
 used in building now commands, owing to the rapid consumption of it on railways, may consider this price too low; 
 it will, however, be found not only ample, but even too high a price to pay for the coarser description of perishable 
 indigenous timber, which, it should be remembered, the process of mineralising, provided for in the estimates, will 
 render, even to the sap wood, more durable than locust, or any timber in a natural state. 
 
 § This price is so ample that I should not have thought of making a comment, was I not aware that it may be 
 
28 ESTIMATES OF THE SCREW-BOLTED STRUCTURE. 
 
 represented to the Directors of railways, (through inadvertence, or the want of due consideration no doubt,) that 
 because the little chair bolts in use, weighing, including the nut, probably not more than half a pound, costs, manu- 
 factured, 125 or 15 cts. per pound, so should bolts of three or four times the weight. It must be obvious, however, 
 that the iron has a value separate from the workmanship; and, that were we to form two or three bolts out of the iron 
 used in making one, the labour on each bolt being nearly the same, the cost per pound of the small bolts should be 
 increased in proportion to the additional labour. 
 
 My chairs, manufactured under a fly-press, should not cost more than 7 or 8 cents per pound. The cotter-bolt 
 fastenings of the Baltimore and Susquehanna Railway cost less than 6i cents. 
 
 II The cast-iron chairs on the Philadelphia and Reading Rail-road, cost only Sj cts. per lb., and a proposal was 
 made to furnish them on the Columbia railway as low as $35 a ton. 
 
 IT The laying of the track on the Philadelphia and Reading Railway, including all the mechanical work and labour, 
 cutting and curving bars, dressing and boring sills, digging trenches, and bedding sills on the broken stone, was done 
 complete for $676 per mile, or 40 cts. per sill, being 1690 sills to the mile, which is a little over 38 cts. per yard. 
 The Lancaster and Harrisburg was laid the same year, 1840, at 33 cts. per yard on mud-sills. 
 ** Railway iron rose, within the last five y^ars, as high as $70 a ton; the extraordinary demand for it having now 
 abated, it has fallen to nearly its former price, and may now be imported, delivered in Philadelphia, at $50 a ton, cash. 
 This is what the j, rails on the Baltimore and Washington Rail-road cost in 1835. The T rails on the state works 
 of Pennsylvania, imported a little anterior, cost, I understand, only $48 50. The Messrs. Ralstons, of this city, 
 inform me, that the heavy j, rails, which they are now importing for the State of Pennsylvania, contracted for last 
 year, will cost, delivered, only a little over 50 a ton. These rails are in bars 18 feet long, weighing 60 lbs. to the 
 yard, or 360 lbs. to the bar. Rails, such as I have estimated for, 20 feet long, 50 lbs. to the yard, will weigh only 333 j 
 lbs. to the bar. The English factories being now provided with slings find no difficulty in rolling long heavy bars. 
 
 3Ir. R. Ralston has, also, shown me a manufacturer's proposal to furnish certain modifications of Evans' patent rail 
 at 12s. over the cost of bar iron, the latter is quoted at from ll. to 8/. per ton. One of these rails approaches in 
 shape my improved "bridge'' section rail. Fig. 15, Plate III., which can be made with equal facility. 
 
 f f It may be seen from the Reports of Engineer officers to the War Department, communicated to Congress, that 
 the Government pays Mr. Kyan's agents $1 35 per pound, in New York funds, for corrosive sublimate, and 50 cents 
 on each pound for the use of the patent, together equal to $1 85 a lb.; and at this rate, including transportation and 
 labour, but exclusive of the cost of tanks, the cost of saturating timber was found to average six cents the cubic 
 foot, when the process was applied in a suitable season of the year. That is, the corrosive sublimate and patent fee 
 costs $5 55 per 100 cubic feet of timber, leaving only 45 cents for the labour in loading and unloading tanks, &c. 
 
 When corrosive sublimate is wanted in such large quantities as would be required on railways, I have shown under 
 the head of preservation of timber, that it can readily be supplied at $1 25 or less per pound. My estimate for 100 
 cubic feet is: — 
 
 3 lbs. of corrosive sublimate at $1 25, . . .... $3 75 
 
 Labour, including proportional cost of tanks, . . . . . . 75 
 
 $4 50 
 Mr. Kyan's agents inform me that they have reduced their patent charge in the case of railways to $25 a mile, 
 instead of a charge per pound. I must leave it, however, to those using the process to add it or not to the estimates, 
 according to the views they may take of the validity of the patent. The specifications, accompanied by remarks of 
 the editor, have been published in the 23d vol. of the Journal of the Franklin Institute, page 396. From the opinions 
 I have heard expressed, by persons of competent judgment, it seems to require a judicial decision to establish it. 
 
ESTIMATES OF THE SCREW-BOLTED STRUCTURE. 29 
 
 Estimate B. — Plate I. 
 
 Quantity of materials and probable cost of constructing one mile of right-angled trellis railway; the rails and 
 timber trellis being united by "wood" screw-bolts, other parts the same as in A, — I. 
 
 98683 feet board measure in trellis, per A, — I, est. at $10, $986 83 
 
 6336 white oak pins, (jo,) 9 in. long, I5 in dia. at 5 ct 31 68 
 
 528 wrought-iron joint-chairs, (S,) at 6 lbs. := ..... 3,168 lbs. 
 
 528 middle chairs, (T,) at 21 lbs. = 1,320 " 
 
 2112 chair bolts, at (S and T,) 1-4 lbs. = ...... 2,956-8 " 
 
 2112 intermediate bolts at, (Q,) -9 lbs. = ~ . 1,900-8 " 
 
 Wrought iron fastenings per mile, .......... 9,345'6 at 9 cts. $841 11 
 
 AVorknianship laying 1760 yards of track at 40 cts. . 704 00 
 
 Cost of the trellis frame and iron fastenings, ........... $2,563 62 
 
 IBs tons of rolled iron rails, (r r,) in bars 20 feet long, 5 in. section, and 50 lbs. to the yard, at $50 a ton, $3,925 00 
 
 Cost per mile with unprepared timber, $6,488 62 
 
 Mineralising 8,287 cubic feet of timber and pins with corrosive sublimate at 4^ cents, . . . 372 92 
 
 Cost of one mile complete B, — I, $6,861 54 
 
 Estimate C. — Plate I. 
 
 Estimate of the cost of a trellis the same as B, — I, when laid with rolled iron rails of 45 lbs. per yard. 
 
 Cost of the trellis and fastenings, &c., B, — I $2,563 62 
 
 70'715 tons of edge rails, 45 lbs. to the yard, at $50 a ton, . 3,535 75 
 
 Mineralising timber, per B, — I, estimate, ........... 372 92 
 
 Cost of one mile complete C, — I, .- . . . . , . . $6,472 29 
 
 It will be perceived that by diminishing the weight of the rails only five pounds per yard, the difference in cost 
 will more than cover the cost of mineralising the timber, and thus preserve it from decay; and as the mineral united 
 with the timber will increase its durability and stiffness, the probability is that a prepared timber structure laid with 
 light rails will, even in the first instance, exceed in strength an unprepared timljer structure laid with heavy rails, 
 which, there can be no doubt, would be the case at the end of three years; so that, were I limited by considerations 
 requiring the strictest economy, I should thus proportion the outlay. 
 
 Estimate D. — Plate I. 
 
 A trellis track laid with rolled iron edge rails of the same weight as those in use on the Baltimore and Washington 
 railway, 40 lbs. to the yard, is estimated to cost as follows: — 
 
 Cost of trellis and fastenings, per B, — I, . . . . ... . . . . . $2,563 62 
 
 63 tons of edge rails 40 lbs. to the yard, at $50, 3,15000 
 
 Mineralising timber per estimate B, — I, . . . .... .... 372 92 
 
 Cost of one mile complete, D, — I, $6,086 54 
 
 PLATE II. 
 
 TIMBER PINNED STRUCTURE. 
 
 Fig. 5 represents a timber-pinned right-angled trellis railway structure. 
 
 In the plan of a trellis track, Plate I., the same bolts that attach the iron rails to the string- 
 pieces, serve also to secure the string-pieces and trellis sills together, and are thus combining 
 bolts to the structure. In the present case I purpose showing how the trellis foundation of my 
 railway structure may be made entirely of timber, without the use of iron bolts; or with bolts 
 independently of those used as fastenings for the iron rails. 
 
30 PATENT TRELLIS RAILWAY 
 
 By this means the light rolled iron plate rail may be used for a time on the trellis structure; 
 and, when worn out, be replaced by heavy rolled iron rails, secured either by my improved 
 fastenings, or by the fastenings in common use. 
 
 Farther, I conceive that by means of this strong trellis structure a railway may be laid with 
 rails of our native cast iron, and prove not only eq^ual, but even greatly superior to the best 
 railway now in use on this side of the Atlantic* 
 
 The rolled iron rails superseded the cast iron in England, from a belief, first, that having 
 fewer joints a rnore even surface could be maintained, which has certainly been, in a measure, 
 the case, but by no means to the extent that was expected. Secondly, both being in the form 
 of edge rails, resting on detached points of support, there was much less risk of breakage 
 with the rolled iron, and they could, consequently, be made much lighter than the cast iron, 
 the difference in weight amply repaying in England the greater additional labour bestowed on 
 the rolled rails. In adhesion of the wheels and durability it does not appear that the rolled 
 iron were at all superior to the cast. 
 
 With regard to the first, we have seen that the joints of the rails cannot be maintained 
 without a substructure; and hence the necessity for the trellis, which, from its nature, is calcu- 
 lated to maintain many joints, properly placed, equally as well as a much less number. Seeing 
 this important advantage, I turned my attention to the subject of devising a more perfect and 
 simple joint for the rails than those hitherto in use. The joint I have succeeded in devising, 
 dispenses with the use of chairs, cotters, screws, and wedges, and while it admits of great 
 latitude in motion, it is believed that it will maintain an evener surface, and be more secure 
 from dislocation than any joint in use, as will be demonstrated in the sequel. 
 
 Secondly, we have seen, also, that isolated points of support for the rails have been aban- 
 doned, and that the most approved mode, pointed out by experience, is to rest the rails on a 
 strong continuous bearing of timber. Now this timber, it is obvious, would guard effectually 
 against the breakage of short cast iron rails placed on it, and it will only be necessary to join 
 them in an effectual and even manner at the ends, which is the object of my improved joint. 
 
 CONSTRUCTION OF THE TIMBER-PINNED TRELLIS. 
 
 Fig. 5 is a plan of the track; the diagonal sills, a c a and h ch, may be of the same length 
 and dimensions as those described in plan first; and they are to be laid on the graded surface 
 of the road in precisely the same manner. On the diagonals of this lattice-work, string-pieces 
 of about 6 by 7 inches section, or larger, may be laid, as in the first case, with the joints on 
 the one side opposite the middle of each string-piece on the other side of the track; in which 
 case the string-pieces should be about 30 feet long, as I consider it necessary in the present 
 construction to introduce transom sills, for reasons that will be stated hereafter. 
 
 It will be seen, by an inspection of Fig. 5, that the scarfing of the string-pieces, rests on a 
 
 * Professor Barlow states that in 1817 malleable iron for the purposes of railways was unknown. 
 
TIMBER PINNED STRUCTURE. 3X 
 
 transom sill, h h, instead of being placed on an intersection of the trellis sills, as in the former 
 case. Were the scarfings placed as before, the driving of large pins, or treenails, so near to 
 the ends of the string-pieces would probably split them, and, even did they not split in the first 
 mstance, the strain that would be thrown on them in use would soon produce such a result; 
 as it is a well established fact, that if the spaces between the points of support, on which the 
 rails, &c. rest, be equidistant, the rails will spring, or be deflected, in a much greater degree in 
 the spaces next to the joint than elsewhere; hence, for the purpose of equalizing the deflections, 
 these spaces are usually made less, which may be effected on my trellis structure by the intro- 
 duction, in the space on each side of the joint, of a short transverse sill, half the length of 
 h h. But, as the quantity of timber in those two short sills Avill be equal to that in a sill, h h, 
 sufficient to extend across the whole structure, it will, evidently, be better to rest the scarfing 
 of the string-pieces on the transom sills, as represented; which, it will be seen, reduces the 
 unsupported space on each side of the joint to less than one-half the width of the other 
 spaces. For this purpose alone, however, it is not necessary that the transom should extend 
 across the structure, but as it will thus form an effectual tie, and be more symmetrical, I think 
 it preferable. Should the scarfings on the one side of the track be placed opposite to the 
 middle of the string-pieces on the other side, as premised, the string-pieces should be 30 feet 
 long, which will place the transom sills at distances of 15 feet. The plan Fig, 5, shows a 
 more economical mode; the scarfings being placed opposite each other, half the number of 
 transom sills will answer, as they are indispensable only at the joints. In the present case the 
 transom sills are placed at distances of 20 feet, the string-pieces being supposed of that length. 
 These transom sills are 10 feet long, and may be about 6 by 8 inches section; they are laid on 
 the flat side, and notched on to the diagonal or trellis sills, leaving 2 inches of the transoms 
 above the trellis, as shown in the cross-section. Fig. 8. There are also two suitable notches 
 cut in the top of the transoms, 2 inches deep, for the string-pieces, d d, to rest in, and thus 
 relieve the wooden pins, p p, &c., of undue lateral strain, which they might otherwise be 
 subjected to near the joints. I consider the angular scarfing, A B, essential to the stability of 
 the present track, and, resting it in the notch of the transom, together with its peculiar struc- 
 ture, will effectually prevent any tendency to splitting there might be in the ends of the string- 
 pieces. 
 
 The ends of the string-pieces d d, &c., by resting in the indents of the transoms h h, are 
 effectually secured from depression and lateral deviation. The next object is to secure the 
 string-pieces, in an effectual manner, on the intersections of the trelhs sills a ca and b c b; this 
 might of course be done by iron bolts, similar to the first track, or by having the top series of 
 obhque sills of sufficient thickness to notch the string-pieces into them; but I desired to avoid 
 the use and expense of iron bolts, and at the same time to hold the string-pieces down in their 
 places in a much more secure and effectual manner than could possibly be done by notches in 
 the sills, which would, also, require a large addition of timber. It may seem strange that the 
 very simple and effectual contrivance which I subsequently devised, of driving two timber 
 pins in directions oblique to the surfaces to be united, and nearly at right-angles to each other, 
 
32 PATENT TRELLIS RAILWAY 
 
 as shown by p p, Fig. 6, should have required the expenditure of more intense thought than, 
 I beUeve, any of the other improvements I have devised. 
 
 Timber pins are one of the oldest means in use for securing the string-pieces, or timber 
 rails, to the sills; but, I believe, they were always driven vertically, or without any object if 
 otherwise, in consequence of which, when used to secure long string-pieces, the weight coming 
 on one end of the string-piece, some one of the intermediate sills would act as a fulcrum, and 
 thus the string-piece became a lever to draw out the pins in the other end; when the pins are 
 drawn in the least, so that the timbers may separate, they are easily crushed by the lateral 
 strains, which would have done no injury had the timbers remained close together. 
 
 Two pins driven obliquely, as shown by the dotted lines, &c., p p, Fig. 6, cannot be drawn 
 by any force applied to the timbers in which they are driven, although each pin might be very 
 readily drawn by a small force applied to itself. 
 
 It is absolutely essential to the perfect success of this plan, that the timbers to be united by 
 oblique pins, should, before the holes are bored, be squeezed together as tightly as possible by 
 means of a screw-clamp, or other similar contrivance. 
 
 The two diagonal sills and the string-piece should be pressed tightly together at each inter- 
 section by such a screw-clamp; and the pressure should be continued not only while the holes 
 are being bored, but until the pins are driven in. 
 
 The pins, which may be about \\ inches in diameter, will be best made of tough white oak, 
 and after being Kyanized and well dried, should fit the holes tightly. The driving of the pins 
 will be facilitated by dipping their points in tar, which will, also, be serviceable in filling up the 
 interstices; or it will be better to run a little tarred mop into the holes. 
 
 The other intersections of the trellis, a, 6 and c, and also the middle and ends of the transom 
 sills h h, are each secured by two pins, p p, driven obliquely, as described and shown in the 
 cross-section, Fig. 7. 
 
 On the timber trellis, just described, it is perfectly obvious that any description of iron rails 
 in use may be laid, and although secured with common fastenings, will, nevertheless, constitute 
 a very superior description of railway. Still better if the rails be made and secured as those 
 in use on the Great Western Railway of England. Or the rails may be secured with my 
 wrought-iron joint-chairs, and short timber screws, (estimate A, II.) 
 
 The light rolled iron plate rail will be much better sustained on this strong timber trellis than 
 on the frail tracks in use, consequently the rails will last much longer than at present, and the 
 liability to accident much lessened. The plate rails may be laid as follows: — 
 
 On the string-pieces, d d, a. strip of hard wood, x x, about 5 inches wide and 2J inches thick, 
 should be spiked or treenailed. On the inner edge of this strip, and over the middle of the 
 string-piece d, Figs, 5, 7, and 8, the fiat bar, or plate rail r", should be laid, and spiked down 
 in the usual manner. When this strip of hardwood becomes bruised and worn, it may, at any 
 time, be ripped up and replaced by another, or by heavy iron rails; means having been taken, 
 in the first instance, to preserve the timber trellis foundation from decay, as by Kyanizing, &c. 
 
 One of the greatest objections that can be urged against the flat bar rail, is the great liability 
 
TIMBER-PINNED STRUCTURE. 33 
 
 it occasions to accidents, and the appalling nature of those caused by the starting of the ends 
 of the bars when they happen to pass up through a passenger car. To obviate this, I have 
 devised a simple and, I believe, an efficient means for uniting the ends of the bars, as foUovrs:- 
 The ends of the plate rails, A and B, Fig. 29, Plate IV., are split equally for two or three 
 inches, as represented; the parts V of A, and a! of B, are then pressed downwards equal to 
 the general thickness of the bars, as shown in the side view of same, Fig. 31, and the parts 
 a of A, and b of B, are cut off to about-one third the length of the others, which completes 
 the joint. The bars A and B, when brought together, will appear as shown at Fig. 30, and 
 in the edge view. Fig. 31, the part a of A, it will be seen, rests on a' of B; and the part b of 
 B rests on V of A. Thus the parts d and II serve instead of the usual splicing-plate, and, it 
 is perfectly obvious, from the manner in which the bars interlock, that all deviation is entirely 
 prevented, and that the bars cannot possibly separate while on the track, unless one should 
 be broken, in which case there will be no sharp point to the bars, so formidable in those used 
 at present. 
 
 IMPROVED CAST-IRON RAILS. 
 
 The form of section I should generally prefer giving cast-iron rails for a continuous bearing 
 on timber is shown at Fig. 26, Plate IV., which is the same as the "bridge" section rails with 
 the centre part cast solid. As it will be expedient, for many reasons, to have these rails cast 
 in short lengths, not exceeding five feet, to insure their regularity in line, and relieve the fast- 
 enings from lateral strain, as well as for the purpose of retaining the parts in place, in case a 
 rail were broken, a rib, r', may be cast along each edge of the base of the rail, so as to fit 
 over the part of the string-piece, v v, Figs. 5, 6, 7 and 8, Plate II. Or the base of the rail 
 may be made throughout of the width n n, Fig. 28, so as to afford room for holes similar to 
 those at the ends, c c, to be made along it at short distances, which would enable the rails 
 to be well secured to the string-pieces by means of screws, spikes, or treenails. 
 
 The most important feature in these cast-iron rails, however, is the new method I have 
 devised for uniting the ends of the rails, so as to allow the joint to open considerably without 
 forming a transverse depression for the wheels to sink in, or there being any dislocation or 
 unevenness of surface, while at the same time there will be such mobihty in the joint as will 
 enable the rails to conform to any necessary undulations, or changes of grade and direction, 
 and, also, to allow of the free elastic action of the trellis. We are thus enabled to unite the 
 hard and durable surface of the cast-iron, which offers so little resistance to the rolling motion 
 of the wheels, with the elastic action of the trellis structure, which will impart an ease of 
 motion to the carriages, and life-like action to the locomotive engines. 
 
 Fig. 27 is a side view of the joint formed by the two rails C and D; and Fig. 28 is a plan 
 of the same, the letters C D, in the latter case, being on the top surface upon which the wheels 
 roll. It will be seen that the top limbs, a and b, of the joint rest in suitable sockets formed 
 5 
 
34 PATENT TRELLIS RAILWAY 
 
 alternately in the ends of the bars, one of which is shown by the dotted lines e, Fig. 27; the 
 limb a of C rests by the rounded surface e on the part a! of D; and the limb 5 of D rests in 
 precisely the same manner on the base b of C, as shown by the dotted lines Fig. 28. All the 
 angles of the joint are rounded to give strength to the parts and allow of mobiHty of action. 
 It will be seen, by reference to Plate II., Figs. 5 and 6, that each end of these bars is 
 formed in precisely the same manner, and that these articulated ends fit into each other when 
 placed together, so that it matters not which end of the rail is turned, it will fit equally well in 
 its place, from which it cannot possibly deviate. Thus "chairs," with their troublesome appen- 
 dages of screws, cotters, or wedges, are entirely dispensed with by this simple contrivance. 
 The facility with which these rails may be turned end for end, is very advantageous in an 
 economical point of view; as the side of the rail next to the flanges of the wheels being first 
 worn out, we are enabled to turn the unworn side of the rail to the flanges, and thus the rails 
 are rendered serviceable for double the period of time they would have otherwise lasted. 
 
 It may also be remarked of cast iron rails that, when worn out, they are of equal value 
 with "pig" iron, or the raw material, and can be recast into rafls at very little expense. 
 
 The cast-iron rails, shown in the drawing, are but five feet long, exclusive of the joint, which 
 insures a more perfect bearing, throughout their length, on the timber string-piece d d, on 
 which they rest. Thus each joint of the rails is placed over an intersection of the trellis 
 where the triple timbers, crossing each other, form a mass beneath the joint twelve inches in 
 thickness; these timbers, diverging thence, have each its separate bearing on the soil, by which 
 arrangement the timbers rest on a much larger surface than it is possible to make them bear 
 upon by any other method of construction; and, while each of the under sills is one-fourth the 
 thickness of the mass, it is loaded by only one-sixth part of the weight resting on the inter- 
 section. This would be the proportionate sustaining power of each cross of the lattice 
 taken separately, but as these crosses are united to each other at their extremities, a b,a b, 
 and on the centre of the track, c c, and as there is, moreover, a strong string-piece, d J, 
 stretching diagonally over several intersections of the treUis, to each of which it is strongly 
 united, any weight placed on one will be distributed so as to rest in a measure on three or 
 more. Thus the joints of the rails are amply sustained. 
 
 The cast-iron rails I propose using, being only one-fourth the length of the rolled-iron rails, 
 will require but one-fourth the opening at each joint. These numerous joints will also enable 
 the rails to conform to the curvature of the road without alteration, and admit of the elastic 
 action of the trellis. In addition to which the following reasons may be assigned for casting 
 the rails in short lengths: First, they will be less liable to break; secondly, they will be less 
 liable to be cast crooked or twisting; thirdly, they will be much more convenient to handle, 
 and are more readily cast vertical, which will insure their solidity and freedom from defects. 
 The top surface of the rails should be chilled, and on roads intended for rapid passenger 
 transportation a strip of railway felt should be interposed between the iron rail and timber 
 string-piece, for the purpose of equalizing the bearing and easing shocks. 
 
 It is not my intention to enter into discussion of the relative merits of cast and wrought 
 
TIMBER-PINNED STRUCTURE. 35 
 
 iron rails, which would require more space than I can devote to the subject at present; and, 
 in fact, experience is wanting to determine the question of comparative durability between 
 rolled rails and the chilled cast rails, when both kinds are laid in the now approved method of 
 a bearing on timber throughout their whole length. 
 
 That cast-iron rails can be furnished from our own mines at a much cheaper rate in the 
 United States than foreign rolled-iron rails, seems probable from the improved modes of 
 smelting iron with anthracite coal. That they will fully answer the purpose, when constructed 
 and laid as I have described, there are but few engineers of practical experience in mechanics, 
 and who are familiar with the extensive use made of cast-iron edge-rails throughout England 
 and Wales on their mineral railways, will deny. Should there be some, however, who are 
 disposed to advance adverse views, it matters not from what motives, let me ask them why is 
 it that the chilled cast-iron wheels of our rail-road cars should have so completely superseded 
 the imported wrought-tired wheels, if it be not their greater durability and cheapness that have 
 recommended them? 
 
 Within the last ten years some millions of dollars have been sent out of the country for the 
 purchase of railway iron, which has been, for the most part, admitted duty free, and, if we 
 may form an opinion from the rapid deterioration by crushing, exfoliation, and splitting of the 
 heavy rolled-iron rails but recently laid on some of our roads, many millions more must con- 
 tinue to follow them to furnish a supply for renewals and repairs, so that it will form an insa- 
 tiable drain on the currency of the country. 
 
 A material that enters so largely into, and forms so costly a part of our great pubhc works 
 should not, therefore, be imported from a foreign country, if we can by any possible device ren- 
 der our native materials available. That the large sums of money which would thus be retained 
 in the country would greatly benefit the mining, manufacturing, and farming interests, and in 
 fact all others, even the works themselves by an increase of business, must be evident to all. 
 
 ESTIMATES OF THE TIMBER-PINNED TRELLIS RAILWAY STRUCTURE. 
 
 Estimate A. — Plate II. 
 
 Quantity of materials and probable cost of constructing one mile of trellis railway, the parts being united by 
 wooden pins or treenails, p p. Fig. 6; laid with rolled-iron edge rails of 50 lbs. to the yard, in bars 15 feet long, 
 secured by independent fastenings, and the joints placed in one line across the track. 
 352 string-pieces, (d d,) 30 feet 8 in. long, 5x8 inches section, = 36,375 ft. b. m. 
 2112 trellis sills, (a c a and b c b,) 14 ft. 9 in. long, 3x8 in. sec- 
 tion, = . . '. 62,304 ft. 
 
 176 transom sills, {h h,) 10 ft. long, 6x8 in. section, = . . 7,040 ft. 
 
 Total quantity of timber per mile, 105,719 ft. at $10, $1,057 19 
 
 4224 white oak pins, {p p, Fig. 6,) 18 in. long, Ij in. dia. at 1 ct 42 24 
 
 6336 " " " (p 7), Fig. 5,)- 9 in. long, li in. dia. at d ct 3168 
 
 704 wrought iron joint-chairs at 6 lbs. = 4,224 lbs. at 9 cts. . . . 380 16 
 
 1408 screws for fastening chairs, | lbs. = 1056 " 10 " . . 105 60 
 
 704 middle bolts, f lbs. = 528 « 10 "... 52 80 
 
 6336 hook-headed spikes, lib. = 2112 " 8" . . • 168 96 
 
 Weight of chairs and other fastenings, 7,920 " 
 
 Workmanship laying 1760 yards of track at 40 cts ....■• 704 00 
 
 Cost of timber trellis and iron fastenings, $2,542 63 
 
36 ESTIMATES OF THE TIMBER-PINNED STRUCTURE. 
 
 Brought over, .... $2,542 63 
 
 78s tons of rolled-iron rails, in bars 15 feet long, weighing 50 lbs. to the yard, at $50 a ton, . . 3,925 00 
 
 Cost of track with raw timber, $6,467 63 
 
 Mineralising 8,957 cubic feet of timber and pins, at 4i cts. ......... 403 07 
 
 Cost of one mile complete, A, — II, $6,870 70 
 
 Estimate B. — Plate II. 
 
 Quantity of material and probable cost of constructing one mile of right-angled timber-pinned trellis railway 
 structure, laid with rolled-iron rails 50 lbs. to the yard, in bars 20 feet long. 
 
 528 string-pieces, (rfrf,) 20 ft. 8 in. long, 5x8 in. section, = . . 36,379 ft. 
 2112 trellis sills, (a c a and 6 c b,) 14 ft. 9 in. long, 3x8 in. section, = 62,304 ft. 
 264 transom sills, (A h,) 10 ft. long, 6x8 in. section, . . . 10,560 ft. 
 
 Quantity of timber per mile, 109,243 ft. b. m. at $10, $1,092 43 
 
 4224 white oak pins, 18 in. long, 1? dia. at 1 ct. . • . . . . . . . . . 42 24 
 
 6336 " . " " 9 " 1| dia. at i ct. . 31 68 
 
 528 wrought iron joint-chairs, 6 lbs. = . . . ' . . 3,168 lbs. at 9 cts. . . . 285 12 
 
 1056 chair-bolts, 6 in. long, J in. dia. at | lb 872 " 10 " . . . . 87 80 
 
 1056 middle bolts, 6 in. long, i in. dia. at | lb. = . . . . 872 " 10 " . . . 87 20 
 
 6336 hook-headed spikes, } lb. = 2,112 " 8 " . . . 168 96 
 
 Weight of chairs and other iron fastenings, ..... 7,024 
 
 Workmanship laying 1760 yards of track at 40 cts. .......... 704 00 
 
 Cost of the trellis and iron fastenings per mile, ......... $2,498 83 
 
 78i tons of rolled-iron rails in bars 20 feet long, weighing 50 lbs. to the yard, at $50 a ton, . . 3,925 00 
 
 Cost of the track with raw timber, . $6,423 83 
 
 Mineralising 9211 cubic feet of timber, at 4| cts. . . . . . . . . . . 414 50 
 
 Cost of one mile complete B, — II, * $6,838 33 
 
 Estimate C. — Plate II. 
 
 Estimate of the cost of track B, — II, laid with rolled-iron edge-rails of 40 lbs. to the yard, being 5 lbs. per yard 
 heavier than the rails first laid on the Liverpool and Manchester railway, and having moreover a continuous bearing 
 on the timber string-piece. 
 
 Raw timber trellis and iron fastenings, . ........... $2,498 83 
 
 63 tons of rolled-iron rails 40 lbs. to the yard, at $50 a ton, 3 ,150 00 
 
 Mineralising 9211 cubic feet of timber at 45 cts. . . _ . 414 50 
 
 Cost of one mile complete C,— II, . . . $6,063 33 
 
 Estimate D. — Plate II. 
 
 Estimate of the quantity of materials and cost of constructing one mile of the right angled timber-fastened trellis 
 railway structure, when laid with an improved description of Native Cast-Iron Rails. 
 
 528 string-pieces, {d d,) 20 feet 7 inches long, (or 40 ft. 7 in. if to be 
 
 had,) 6x7 in. section, = 38,038 ft. 
 
 2112 trellis sills, {a c a and b c b,) 14 feet 9 inches long, 3x8 in. sec- 
 tion, = 62,304 ft. 
 
 264 transverse sills 10 feet long, 6x8 in. section, = . . . . 10,560 ft. 
 
 Total quantity of timber per mile = 110,902 ft. at $10, $1,109 02 
 
ESTIMATES OF THE TIMBER-PINNED STRUCTURE. 37 
 
 Brought over, $1,109 02 
 
 4824 white oak pins 18 inches long, I5 in. sec. at 1 ct . . 42 24 
 
 6336 white oak pins 9 inches long, 1 J in. sec. at ^ ct. . . . . . . . . . . 31 68 
 
 Workmanship laying 1760 yards of track at 40 cts. ......... 704 00 
 
 4224 engine-turned treenails, 6 in. long and I in. in diameter, for securing the ends of the rails, at 1 ct. . 42 24 
 
 Railway felt, {t t,) per mile of track, 30 00 
 
 Cost of track exclusive of rails, $1,959 18 
 
 110 tons of cast-iron rails, in bars 5 feet long, (»"' r'. Fig. 5,) having a section of 7^ inches nearly,and weigh- 
 ing 70 lbs. to the lineal yard, exclusive of an allowance at the ends for the improved joints, in place of 
 chairs, as under; base of the rail 5i inches wide, including lateral ribs, (r' ?•', Fig. 26,) to have a con- 
 tinuous bearing on a timber string-piece. Cast from the smelting furnaces, as at present, at $30 a ton, 
 which it is believed the Anthracite furnaces will be able to furnish at $20. Taking the present price 
 
 at $30, 3,300 GO 
 
 7-07 tons additional, being an allowance of 3| lbs. on each end of the bars, to increase the width of bear- 
 ing surface, and otherwise strengthen the improved joints, in place of chairs, &c. &c. ,at$30 a ton, 212 10 
 
 Cost of track with unprepared timber, ............ $5,471 28 
 
 Mineralising 9432 cubic feet of timber at 4^ cts 424 44 
 
 Cost of one mile complete, D, — II, $5,895 72 
 
 Estimate E. — Plate II. 
 
 Estimate of the cost of the trellis track D, — II, when laid with Cast-Ieon Rails, weighing 60 lbs. to the lineal 
 yard, exclusive of joints. 
 
 Cost of the trellis structure exclusive of rails and mineralising, as D, — II, estimate, . . . $1,959 18 
 
 96-07 tons cast-iron rails, (r' r',) 60 lbs. to the yard, at $30 a ton, 2,882 10 
 
 7-07 tons additional, in lieu of chairs, at $30, 212 10 
 
 Cost of one mile unprepared timber, • . . 5,053 38 
 
 Mineralising 9432 cubic feet of timber at 41 cts. . . ' . . 424 44 
 
 Cost of one mile complete E,— II, $5,477 82 
 
 Estimate F. — Plate 11. 
 
 Estimate of the quantity of materials, and cost of constructing one mile of trellis railway, the same as D, — II, 
 but laid with rolled-iron flat bars, or plate-rails. 
 
 Timber in the trellis structure, per D, — II, est. .... 110,902 ft. 
 
 A strip of hard wood, {x x,) 5 inches wide and 2 in. thick, spiked on 
 
 the string-pieces, to bear the flat bar rails, ()'",) . . . 8,800 ft. 
 
 Total quantity of timber per mile, 119,702 ft. at $10, $1,197 02 
 
 4224 white-oak pins, (p JO, Fig. 6,) 18 inches long, I5 in. dia. at 1 ct 42 24 
 
 6336 white-oak pins, {p p,) 9 inches long, Ij in. dia. at 5 ct. . . . 31 68 
 
 Workmanship laying 1760 yards of track, at 40 cts 704 00 
 
 Cost of the timber frame work, , $1,974 94 
 
 7040 spikes for securing plate-rails, = 1760 lbs., at 8 cts 140 80 
 
 22-55 tons of flat bar, or rolled-iron plate-rails, 2j inches wide and f inch thick, with an allowance at the 
 
 ends for the improved joint, (Figs. 29, 30, 31,) at 50 a ton, 1,127 50 
 
 Cost of one mile with unprepared timber, $3,243 24 
 
 Mineralising 10,000 cubic feet of timber at 4i cts 450 00 
 
 Cost of one mile complete, F, — II, $3,693 24 
 
38 ' PATENT TRELLIS RAILWAY 
 
 PLATE III. 
 
 OBLIQUE-ANGLED TRELLIS STRUCTURE. 
 
 Fig. 9 represents a simple modification of my Trellis Railway Structure, designed chiefly with 
 a view to the following objects: Firstly, for the purpose of diminishing, as far as possible, con- 
 sistently with strength, the quantity of timber in the track, in accordance with the views still 
 Entertained by the engineers who rest the iron rails directly on transom sills, or isolated points 
 of support, while these transom sills, in some cases, do not even rest on mud-sills, as this addi- 
 tion would at once double the quantity of timber in the track. The narrow transom-sills have, 
 therefore, no other connection than that formed by the iron rails, the flections of which rock 
 them from side to side as the load rolls over them. The hopelessness of the expectation that 
 by constantly ramming broken stone beneath these sills a dense foundation and "level" surface 
 will be the ultimate result, is clearly shown by the long experience of the most noted English 
 railways, where the ponderous stone blocks, with their broad bearing surface, placed imme- 
 diately under each line of rails, so as to have as much bearing outside as within the rails, and 
 rammed on the ballasting with the force of a pile engine, failed, nevertheless, to effect this 
 much desired object. (See page 7.) 
 
 Secondly, to reduce the extreme width of bearing surface of the treUis to eight feet; that 
 being the usual width of the tracks in use, and thus make it suit the narrow graded surface 
 that many of our railways have adopted, from motives of economy in the first cost. 
 
 The lower series of trellis sills, a, c, a, are to be well bedded on the ballasting by repeated 
 blows of a heavy maul or rammer, until the top surface of the series is in one plane, coinci- 
 dent with, or parallel to, the grade line. These sills are laid more obliquely to the rails than 
 in the former plans; so that a sill under the rails at the first point of support on one side of 
 the track will extend under the third point of support on the other side; or from 1 to 3', the 
 next from 2 to 4', and so on from 3 to 5', &c. &c. And the trellis sills in the top series, b c b, 
 being laid on the other, so as to incline at the same angle in an opposite direction, wifl extend 
 from the intersection under the rail at 1' to the intersection of the sills under the opposite rail 
 at 3; and the next sill from 2' to 4; so on from 3' to 5, from 4' to 6, &c. &c. One sill in the 
 top series will, therefore, intersect and rest on seve?i sills of the lower series, as represented. 
 
 It is evident that this system of laying one series of sills obliquely on the othei-, and uniting 
 them together, will have the effect of perpetuating the bearing surface in as unbroken and 
 regular a manner as if it were one broad plank, wide as the trellis, and long as the railway. 
 
 Diagonally on this trellis-work, rolled, or cast-iron edge rails of any desired weight and 
 section, either semi-elliptical or parallel, may be laid. Those represented are the parallel broad 
 base x» in bars 18 feet long, laid so as to break joint with those on the opposite side of the track, 
 as represented, r r being the rails. The trellis sills are three feet apart from centre to centre, 
 measured along the rails, but, owing to their oblique position, there is but one foot eight inches 
 of the rail clear of the bearings in each space, when of the proportions shown in the draw- 
 
OBLiaUE-ANGLED STRUCTURE. 39 
 
 ing. The sills in the present design are all I25 feet long and 3 inches thick, which gives a 
 thickness of 6 inches under each bearing point; the sills on which the joints of the rails rest, 
 marked a' a' and 6' 5', stretch alternately from joint to joint on the opposite sides of the track; 
 there are four at each joint of 3 by 8 inch section, and eight intermediate sills, a a and b b, of 
 3 by 6 inch section under each rail. This trellis is united by wooden pins, p p, driven 
 obliquely, as before described. 
 
 As the deflexure of the rails would be much greater in the spaces on each side of the joints 
 of the rails than elsewhere, compensation pieces, x x, are introduced at one-third of the 
 space from the joint to support the rails. 
 
 The joints of the rails are secured by the wrought-iron joint-chair, before described, which 
 is fastened to the trellis by two timber-screws, i i, Figs. 9 and 11. The middle of each rail 
 is permanently held from "driving," or moving end-wise, by a timber screw-bolt, 0, and spike, 
 t. The intermediate points may be secured by hook-headed spikes, 1 1, driven as shown in 
 the cross-section. Fig. 10. 
 
 The ease and rapidity with which this trellis may be laid, and its excellence when done, will 
 go far to recommend it even for temporary railways. Yet, on close examination, it will be 
 found to possess some remarkable features of what would seem to be intricate combination. 
 Take any one rail in the track for instance, it will be found to rest on seven intersections, 
 formed by fourteen sills within the length of eighteen feet; these sills divide and diverge so far 
 as to spread out under nearly the whole length of the two rails on the opposite side of the 
 track; or to a distance of 33 feet. The extended ends interlace with other sills that spread to the 
 length of three rails on the first side, and so on alternately. Notwithstanding this seeming 
 intricacy any sill may be removed and replaced in a very short time; for this purpose it will 
 only be necessary to draw the spikes, 1 1, &c., which secure the rails, a saw being inserted 
 between the sills will cut the pins,jo p, or they may be driven out; the sill being of a regular 
 thickness throughout is very readily driven out endwise, and its place as easily filled by another 
 sill of the same thickness. 
 
 In wet cuts, and other bad foundations, the above described trellis may be greatly stiffened 
 by the addition of longitudinal timbers, shown by the dotted lines e e, placed as mud-sills, and 
 secured to the trellis by the pins, p p. Or these timbers may be laid under the rails as string- 
 pieces, d! d, &c., which will enable us to cover the trellis with the ballasting. But both string- 
 pieces and mud-sills of suitable dimensions, as in the estimates, may be added to the trellis, 
 and there will still be less timber in this strong trellis track than there is in the Baltimore and 
 Washington, Baltimore and Port Deposit, and other noted lines of railway, laid without the 
 use of broken stone. 
 
40 ESTIMATES OF THE OBLIQ,UE-ANGLED STRUCTURE. 
 
 ESTIMATES OF THE TRELLIS RAILWAY STRUCTURE.— PLATE III. 
 
 Eslimale A. — Plate III. 
 Quantity of materials and cost of constructing one mile of the above described trellis railway track, when laid 
 with Rolled-Iron Edge Rails, r r, in bars 18 feet long, weighing 50 lbs. to the lineal yard. The trellis being eight 
 feel wide, laid without mud-sill or string-piece. 
 
 880 joint-sills, (a' and b\) 12i feet long, 3x8 in. section, = . . 22,000 ft. b. m. 
 
 2640 intermediate sills, (a and 6,) 125 ft. long, 3x6 in. section, ^ 49,500 " 
 
 1174 joint-pieces, (a;,) 2 ft. long, 3x4 in. section, = . . . 2,348 " 
 
 Total quantity of timber per mile, 73,848 ft. at $10, $738 48 
 
 15840 white oak pins, {p p,) 9 in. long, I5 in dia. at 5 ct. . . . . . . . . . 79 20 
 
 586f joint-chairs at (S,) 6 lbs. = 3520 lbs. at 9 cts. . . 316 80 
 
 1760 screws for chairs, &c., I lbs. = 1320 lbs. at 10 cts. . . 132 00 
 
 5280 hook-headed spikes at i lb. = 1760 lbs. at 8 cts. . . 140 80 
 
 Weight of all the fastenings and chairs = ..... 5600 lbs. per mile. 
 
 AVorkmanship laying 1760 yards of track at 30 cts. per yard, ........ 528 00 
 
 Cost of the trellis frame and iron fastenings, &c. .......... 1,935 28 
 
 78g tons of rails, 50 lbs. to the yard, at $50 per ton, 3,925 00 
 
 Cost of track with unprepared timber, ......... . . ■. $5,860 28 
 
 Mineralising 6181 cubic feet of timber at 42 cts. . . . . . . - . . . . 278 15 
 
 Cost of one mile complete A, — III, ............. $6,138 43 
 
 Estimate B.— Plate III. 
 
 Estimate of the above track with an addition either of a mud-sill, (e c,) or a string-piece, {d' d',) of 24 inches 
 section. 
 
 Timber in the trellis work as above ...... 73,848 ft. 
 
 10560 lineal feet of mud-sills of 3x8 in. sec, or string-pieces of 4x6 
 
 in. sec. = 21,120 ft. 
 
 Total quantity of timber per mile 94,968 ft. at. $10, $949 68 
 
 15840 white-oak pins, (j!) jo,) 9 in. long, 1| in dia. at 5 ct 79 20 
 
 Iron fastenings as in estimate A, — III, ............. 589 60 
 
 Workmanship laying 1760 yards of track at 35 cts. . ......... 616 00 
 
 Cost of trellis frame, iron fastenings, &c. ........... $2,234 48 
 
 78s tons of rolled-iron rails of 50 lbs. to the yard, at $50 a ton, 3,925 00 
 
 Cost per mile with unprepared timber, ............ $6,159 48 
 
 Mineralising 7941 cubic feet of timber at 45 cts. . . . . . . ' . . . . . 357 34 
 
 Cost of one mile with mud-sills or string pieces, B, — III, ........ ^6,516 82 
 
 Estimate C— Plate III. 
 
 Estimate of the quantity of materials and cost of constructing one mile of the foregoing trellis track with the 
 addition both of mud sills (e e,) and string-pieces, {d' d'.) 
 
 3520 trellis sills, (« and b,) 12§ feet long, 3x6 in. sec. = . 66,000 ft. b. m. 
 
 586| string-pieces, (rf' d\) 18 ft. 6 in. long, 4x6 in. sec. = . . 21,707 " 
 586| mud-sills, (e fi,) 18 feet long 3x8 in. section = . . 21,120 " 
 
 Total quantity of timber per mile, = 108,827 b. m. at $10, = $1,088 27 
 
 5600 lbs. of chairs and other iron fastenings per mile, as in estimate A, — III, ..... 589 60 
 
 19360 white oak pins, (jf ;J,) 9 in. long 1| in. dia. at 5 ct. . . . . . . . . . 96 80 
 
 Workmanship laying 1760 yards of track at 40 cts. ......... 704 00 
 
 Cost of the trellis frame and iron fastenings, &c. ......... $2,478 67 
 
ESTIMATES OF THE OBLIQ,UE-ANGLED STRUCTURE. 4X 
 
 Brought over, $2,478 67 
 
 Mineralising 9204 cubic feet of timber at 4j cts. .......... 414 18 
 
 If this strong timber trellis be laid with rolled-iron edge-rails of the same section and weight as those 
 now in use on the Baltimore and Washington, and several other noted lines of railway, being 5 lbs. 
 heavier, per yard of bar, than the rails first laid from Liverpool to Manchester on detached bearings 
 3 feet apart, it would seem to be evident, according to the well-known principles of constructive car- 
 pentry, that the result would be a structure much superior in strength, in combination, in evenness of 
 surface, in correctness of direction, and in the extent of bearing it would have on the soil, to any tran- 
 som or block-formed structure composed of the same quantity of materials; consequently the iron rails 
 and timber structure would endure the action of wear for a much longer period, and as the limits Of 
 durability of mineralised timber, from the evidences we have had, must greatly exceed the wear, we 
 may conclude that the actual wear of attrition on the iron rails will alone limit the duration of the 
 track. Now as the wear on the surface of the rails is scarcely an appreciable quantity, apart from 
 the oxydation of the bars, which can readily be prevented, and as the strength of the structure would 
 save the bars from the crushing and exfoliation caused, in the old mode, by the rapid flexures of the bars 
 separating the laminae of the iron, we may reasonably estimate the duration of a well made trellis track, 
 formed of mineralised timber, at 50 years; by which time the iron rails, protected from oxydation, 
 would not have lost more than one-tenth of their original weight, let the trade be what it might*. 
 63tons of rails, 40 lbs. to the yard, at $50 a ton, $3,150 00 
 
 Cost of one mile complete C,— III, . .' $6,042 85 
 
 The foregoing tracks are of ample strength, under suitable circumstances, to endure the 
 constant action of locomotive engines, weighing from 9 to 12 tons, at the rate of 20, 30, or 
 even 40 miles an hour, without injury to the road. And 16 ton engines at the rate of 10 or 
 12 miles an hour. 
 
 Estimate D. — Plate III. 
 
 By laying the timber trellis C, — III, with iron rails of 50 lbs. to the yard, a track will be formed of ample 
 strength to bear the action of 16 or 18 ton engines at high velocities. 
 
 Timber trellis, iron fastenings, workmanship, and mineralising, as per estimate C, — III, . . . $2,892 85 
 781 tons of rolled-iron rails, 50 lbs. to the yard, at $50 a ton, ....... 3,925 00 
 
 Cost of one mile complete, D,— III, $6,817 85 
 
 Estimate E. — Plate III. 
 
 Estimate of the quantity of materials and cost of constructing one mile of trellis railway formed nearly as A, — 
 III, but having all the trellis sills, {a c a and b c b,) of one size, and without the compensation pieces, {x x,) the track 
 being laid with "fish-bellied," or semi-elliptical Cast-Iron Rails, 3 feet long, weighing 65 lbs., including an allow- 
 ance at the ends for a wide seat and improved joint, in place of chairs. 
 3520 trellis sills, (a c « and 6 c b,) 12i feet long, and 3 by 6 inches section, = 66,000 fl.b. m. at $10, = $660 00 
 
 15840 white-oak pins, (p JO,) 9 inches long, I5 in. dia. at 5 ct. 79 20 
 
 Workmanship laying 1760 yards of track at 20 cts. per yard,t ........ 352 00 
 
 7040 engine-turned tree-nails, for securing the ends of rails, 6 inches long, | diameter, 1 ct. . . . 70 40 
 
 102-14 tons cast-iron rails, 65 lbs. to the yard, at $30 a ton, , 3,064 20 
 
 Cost of one mile without mineralising, ............ 4,225 80 
 
 Mineralising 5658 cubic feet of timber at 4i cts 254 61 
 
 Cost of one mile complete E,— III, ..."........ $4,480 41 
 
 *I have seen it stated that a 15 foot rail laid loosely on the ground along side of the track, lost 8 oz. in a year, 
 which was precisely the same as the loss in weight of a similar rail in the track over which a large amount of tonnage 
 had passed. Professor Barlow, p. 340, states that among the papers submitted to them, they found it estimated at 
 l^th of a pound per yard per annum; but that he had since seen an account of experiments made by Mr. Dixon in 
 which the whole loss was found to be only ^Lth of a lb. If the sides and bottom of the rail were protected from 
 oxydation, I am inclined to believe that the loss from the top surface would not exceed one ounce per yard per annum 
 with a trade of 350,000 tons, except on steep grades were the brake was constantly used. 
 "I" This, although the lowest price per yard, is nevertheless the best of any in the estimates. 
 6 
 
42 PRIMITIVE TRELLIS RAILWAY STRUCTURE. 
 
 The last track estimated for is not only very far superior to any of those tracks laid with 
 rolled-iron plate-rails, but it will be found superior in strength and durability to many of the 
 tracks now in the United States laid with heavy iron. Nor is there any more reason to 
 apprehend the breakage of these rails than there is the breakage of the cast-iron wheels of 
 our railway carriages; yet I have designed it with a view chiefly to the mineral and branch 
 railways, particularly those through the streets of cities, on which horses or slow-motioned 
 engines would be used, for which purpose it will be found greatly superior to the granite plate 
 track. 
 
 PLATE IV. 
 
 THE PRIMITIVE, THE DIAMOND, AND THE IRON TRELLIS. 
 The Primitive Trellis Railway. 
 
 Fig, 22 is intended to illustrate three distinct modifications of the trellis railway structure. 
 The shaded lines, a' a and hi b', represent an iron trellis, which we will suppose to be entirely 
 removed while we are considering the timber trellis. There are two modifications of the 
 timber trellis; the first I shall describe is identical with the track named in the Report of the 
 Franklin Institute as No. 1, which was in fact the original trellis, or the first I devised. 
 
 T, T, &c. are the scarfings which unite the ends of strong string-pieces of timber, over 
 each of which is placed the middle of a heavy iron edge rail, and the joinings of the latter 
 rest on the middle of each string-piece at S S, as before described. The scarfings T, and the 
 joints S, are of course the weakest points in this continuous rail of timber and iron, and would 
 be the first to give way under the action of any force. The chief force to which it would be 
 exposed in a railway would be nearly in a vertical direction from the action of the weight and 
 momentum, tending to press the whole rail into the earth; and if it were thus laid down, the 
 scarfings T, and joints S, would undoubtedly be the first points to sink, for want of stiffness. 
 The next force in order is in a lateral direction from the action of the flanges and conical 
 form of the wheels, tending to spread the rails apart. On a very well-constructed track this 
 force may be reduced to tV or to of the former force, but, on the best tracks in the United 
 States, it is fully one-tenth, or more; and, if the track be at all zigzagged, it will be so greatly 
 increased as frequently to equal, or even exceed, the vertical weight, which is the case when 
 an engine or car runs off by the mounting of a flange on the rails; the weight on the wheel 
 being evidently lifted, and of course exceeded, by the intensity of the lateral force. 
 
 The very great importance of fully sustaining the scarfings and joints of the combined rails 
 of timber and iron seemed to my mind perfectly obvious, and led me to devise the trellis 
 bracing. Extending a suitable piece of timber, a c a, in one direction under the scarfing T, 
 T, so proportioned that it may have the greatest possible extent of bearing surface on the 
 earth, consistent with strength; and, crossing it by another, b c b, laid on the former and 
 
PRIMITIVE TRELLIS RAILWAY. , 43 
 
 extending under the scarfings in another direction, we shall have the thickness of both pieces 
 under the scarfing T, and as the ends of both pieces project as far outside the rails as the 
 usual cross-ties, but, being obliquely placed, it is obvious that these two have more than double 
 the bearing surface of the cross-tie, both being the same width. These sills tend to strengthen 
 the scarfs, not only by their extent of bearing surface on the soil, but, also, by their oblique 
 position extending so far in a longitudinal direction under the ends of both string-pieces. The 
 middle of the iron rail, and the ends of the two string-pieces being securely bolted to the two 
 oblique sub-sills at each scarfing T, as represented, the scarfings on each side of the track 
 are alternately the apices of isosceles triangles, the string-piece on the opposite side of the 
 track being the base. At the middle point of this base, S, is the other weak point of the 
 continuous rail, caused by the joint of the railway bars. From under these joints on the one 
 side of the track to those on the other side, S, S., &c., oblique sills, the same in every respect 
 as the former, are laid down and secured by the chair-bolts, thus forming a counter triangle, 
 having the iron rail r r for its base, which completes my first devised trellis structure. 
 The distances from T to S, on the same side of the track, were 8 feet, and the string-pieces 
 were 6 inches thick and 10 inches wide; to which the rail was secured by spikes in- the above 
 distance. The oblique sills were let project about eighteen inches outside of the iron rails, as 
 in the present design, and cut square off, so that the structure might be within the limits of 
 the tracks in use. 
 
 If the continuous rail of timber and iron 10 inches wide, which forms one side of this track, 
 has sufficient bearing surface on the soil to sustain the passing loads, the only weak points in 
 it, T and S, are evidently very amply sustained by the cross of the oblique sills on which each 
 of these points rest. The oblique sills, in addition to the enlarged bearing on the soil which 
 they afford, serve also as direct ties to keep the track from spreading; braces, to keep it from 
 lateral deviation, which, we have seen, so materially increases the flange friction and risk of 
 accident; longitudinal ties to strengthen it, and, being loaded with the ballasting, they serve to 
 hold the track down and keep it from rising in front and rear of the trains. 
 
 I have previously stated that the parts of my railway structure were united so effectually 
 that the whole form but one; and, as I have expressed it in my specifications, "I thus by the 
 union in one, to an indefinite extent, of such materials as those that usually compose railway 
 tracks, obtain by a united framing a more extensive and uniform bearing on the soil than the 
 individual parts would have; all other railways having to depend upon the uniformity of soil, 
 or artificial road-beds, for their evenness of surface. Whereas my railway track is inde- 
 pendent in its formation of the soil on which it rests," as a ship is of the ocean on which she 
 floats. Farther, it is an established fact, according to the best writers on the strength of 
 materials, that a beam of timber held tightly at the two ends will sustain double the weight that 
 would have broken the same beam were it to rest loosely on the points of support. Hence, it will 
 be perceived, that the mud-sills and other parts of a track resting loosely on the soil, can at 
 most sustain but half the weight of similar parts in the trellis structure, the ends of which are 
 
44 . DIAMOND TRELLIS RAILWAY. 
 
 tightly bolted to the iron bars; and for the same reason I prefer uniting the ends of the 
 oblique sills. 
 
 The Diamond Trellis Railway. 
 
 In the present design the distances S T, T S, &:c., Fig. 22, Plate IV., are 10 feet long, and 
 the main trellis sills running under the scarfings and joints, are 18 feet long, 3 inches thick, 
 and 8 inches wide. Instead of using a large continuous beam of timber for the string-piece 
 d d,l conceive that with much less timber a more extensive bearing surface on the soil may 
 be given to the track, together with greater vertical and lateral stiffness, by using a string- 
 piece about 5 inches thick and 8 inches wide, supported at Q on the intersection of two short 
 intermediate or sub-sills,yy, each 3 inches thick and 6 inches wide, which are lapped with and 
 pinned to the main trellis-sills, as represented. The iron rails at the intermediate points 
 Q, might be secured by long hook-headed spikes driven obliquely into the sub-sills, or by some 
 of the other fastenings before described, but it is represented as secured by a wrought-iron 
 clamp, or intermediate chair, Fig. 25, which is driven on the rail from one end. This chair 
 being driven to the right point is secured to the trellis by screw-bolts, passing through the 
 holes c c, and having nuts above, so as to afford greater facility in the removal of a rail. 
 These bolts, and also those that secure the chairs S and T, may require to be put in a little 
 obliquely, which can readily be done by bending downwards the parts of the chairs through 
 which the bolts pass, and chamfering the string-pieces to suit. It is obvious that timber- 
 screws, as before described, may be used to secure those chairs; or that plain long spikes 
 driven obliquely through the string-piece into the trellis would make a very effectual and cheap 
 fastening to the whole, but would not afford the same facility in a removal of the rails, or other 
 parts, as the screw-bolts or timber-screws. 
 
 The chairs at T are also of wrought-iron, formed as before described. Figs. 16 and 24; 
 the base of the rail has a notch cut on both sides, and the chair being heated red hot, the rail 
 is inserted between the parts m m; which are then hammered down; so that the rail will be 
 permanently held at the middle by this chair; the chair being let in across the two laps of 
 the string-pieces and bolted. 
 
 The joints of the iron rails S, are half lap-joints, secured by the wrought-iron joint-chairs, 
 which have been fully described before. 
 
 The Iron Trellis laid in Asphaltum Cement. 
 
 I have hitherto described the trellis. Fig. 22, Plate IV., as composed of timber, as I con- 
 ceive that when formed of this material, properly preserved from decay, it will, from its 
 cheapness and other qualities, be of more extensive practical utility than when formed in the 
 imperishable but somewhat costly manner I am about to describe. 
 
 The shaded lines, a' a' and hi b', are, as before stated, an iron trellis made in precisely the 
 
IRON TRELLIS RAILWAY. ^ 45 
 
 same manner as the timber one just described, the bars being riveted to each other at their 
 intersections. The main trellis bars, a! a! and b' 6', extend under the joints and scarfings of the 
 rails and string-pieces S and T; these bars are 16 feet 10 inches long, and may be about 2 
 inches wide by i inch in thickness; the intermediate bars extending under the points Q are 5 
 feet 9 inches long, about 1^- inches wide, and f inch thick. 
 
 If this iron trellis were to be laid on the ballasting it is evident that it would very soon be 
 buried and distorted for want of sufficient bearing surface; and the bars would be rapidly 
 destroyed by oxydation. 
 
 By means of any bituminous substance, such as asphaltum cement, we can form a tough 
 conglomeration round the bars that will not only afford the requisite extent of bearing surface 
 on the ballasting, but, being impermeable to water, will likewise hermetically seal the iron bars 
 from oxydation or decay, and thus by the combination of these minerals we will be able to 
 form an imperishable trellis foundation for our railway. 
 
 Where a suitable bituminous substance can be cheaply obtained, the whole of the ballast- 
 ing may be united by the cement, and thus too a timber trellis might be preserved; but where 
 it is an object to economize the cement, we should proceed thus: Iron moulds should be had 
 to suit the shape of the different parts which may be in sections as shown by the dotted lines 
 i i i i, i k, i k, k k k k, &c., which show the points of juncture. These moulds are to be filled 
 with the particles of finely broken stone, and the hot asphaltum cement poured into the inter- 
 stices, so as to unite the whole in one mass. On cooling, the cement blocks are turned out 
 of the moulds and laid together on the compact ballasting in the above form, shown by the 
 outlines which before served to represent the timber-sills. The cement blocks having been 
 cast with a suitable groove in the upper surface to receive the iron trellis bars, the latter is 
 put together and hot bituminous cement poured over the iron in the groove, which will flow 
 between the blocks and unite their ends in one continuous mass; the hot cement is likewise 
 poured over the top surface of the blocks, and thin cakes of the compound cement laid on 
 them, as u u, Fig. 23, which completely covers the iron work from moisture or the atmos- 
 phere, and completes the iron trellis foundation. Under the joints S and T the blocks should 
 be enlarged to increase the bearing surface, as shown by the dotted lines k k; and in Fig. 23. 
 
 This trellis may be formed so that continuous asphaltum sills would support the rails, but I 
 would prefer laying a continuous string-piece of timber, d d, on the trelhs, on which the iron 
 rails would be laid and secured as before described; an iron under-bar, shown in the cross- 
 section, being riveted to the intersection of the trellis-bars, and through this the chair-bolts 
 pass, holes being left in the asphaltum blocks to remove them when necessary; the head of 
 the bolt being underneath, as shown in the cross-section N, O. 
 
 The iron rails shown in the section are an improvement on the common "bridge" section 
 rail, as shown half size. Fig. 15, Plate III., the base extending inwards to give more equal 
 bearing on the string-piece; and I would also fill the cavity with hot asphaltum cement, which 
 would, in effect, extend the bearing surface to the whole width of the rail, and also preserve 
 the interior from loss by oxydation. 
 
46 IRON TRELLIS RAILWAY. 
 
 It would evidently be improper to lay such a structure as I have last described, on freshly 
 formed embankments or the loosely graded surface of a road; on the contrary the embank- 
 ments should be well consolidated, and the whole graded surface rendered as compact as possible; 
 which would probably be done in a most uniform and effectual manner by means of a very 
 heavy roller. I would, also, form the graded surface high under the middle of each track, as 
 at A in the cross-section. Fig. 23, taken at N, O, of the plan, and sloping off on each side, as 
 at B and C. This sloping surface, particularly in stiff clay lands, should be rendered as com- 
 pact and smooth as possible, that the rain water, as it percolates through the ballasting, may 
 flow off on each side, as we see it do in natural formations. To facilitate the egress of the 
 water, open ballasting of large gravel, or broken stone, should be first laid on at B and C, and 
 carried gradually up towards A; on this a good bed of fine gravel should be laid, the top sur- 
 face being formed horizontal-transversely; the whole should be rendered compact and even by 
 the use of a very heavy roller. On this well-prepared bed the asphaltum blocks to receive 
 the iron trellis should be laid, and the whole being finished, as before described, the trellis is 
 covered by similar ballasting of fine gravel, which motives of economy would limit to the lines 
 B y y C; but the full security of the cars and engines, so as to keep them from ever running 
 off a bank, would require the ballasting to be carried out a distance from the rail exceeding 
 the width of gauge, and to be raised at the outer limits, as shown by the outline w. 
 
 Such is my plan of construction for a Finished Permanent Railway; and I think it one par- 
 ticularly well adapted to France and Belgium, and perhaps to Austria and England; also to 
 railways in the streets and vicinity of our large cities. With regard to the effect which the 
 expansion and contraction of the iron bars composing the trellis would have, I apprehend that 
 it would be found to exhibit itself only in a very slight increase and diminution of the width 
 of gauge, there being no transverse ties to prevent motion in this direction, and all the bars 
 being proportional, or forming common sides to similar triangles, united by a single cylindrical 
 rivet at each intersection. 
 
 ESTIMATES OF THE PRIMITIVE, THE DIAMOND, AND THE IRON TRELLIS RAILWAY 
 
 STRUCTURES. 
 
 Primitive Trellis. — Plate IV. 
 
 Estimate of the quantity of materials and cost of constructing one mile of trellis railway track, designated in the 
 Report of the Franklin Institute as No. 1, — when laid with rolted-iron edge-rails weighing 50 lbs. to the yard, in 
 lengths of 16 feet; the rails having a continuous bearing on a strong string-piece of timber, which is united to trellis 
 sills at every 8 feet; these trellis sills have but three intersections, or one in the top series rests on three of the 
 lower. 
 
 660 string-pieces, (d d,) 16 ft. 10 in. long, 6x10 in. section, = . . 55,550 b. m. 
 1320 trellis sills, (a and b,) 16 ft. long, 3x8 in. section, = . . 42,240 " 
 
 Total quantity of timber per mile, 97,790 at $10, = $977 90 
 
 1320 white oak pins, (/I/),) 9 in. long, 14 dia. at 5 ct. • 6 60 
 
 660 wrought-iron joint-chairs, (at S,) 6 lbs. each =: . . . 3,960 lbs. 
 
 1320 chair-bolts, washers and keys, 2,112 " 
 
 1320 scarf-bolts, washers and keys, at (T,) . . . . . 2,284 " 
 
 8,356 lbs. at 9 cts. . . . $752 04 
 
 $1,736 54 
 
ESTIMATES OF THE PRIMITIVE, DIAMOND, AND IRON TRELLIS. 47 
 
 Brought over, $1,736 54 
 
 5280 hook-headed spikes, i lb. each, = 1,760 lbs. at 8 cts. . . 140 80 
 
 Workmanship laying 1760 yards of track, at 35 cts. . . ; 616 GO 
 
 $2,493 34 
 
 785 tons of rails, weighing 50 lbs. to the yard, at $50 a ton, 3,925 00 
 
 Mineralising 8150 cubic feet of timber at 41 cts. ........... 366 75 
 
 Cost of one mile complete. A, — IV, . $6,785 09 
 
 Diamond Trellis. — Plate IV. 
 
 Estimate of the cost of constructing one mile of the Diamond Trellis Railway, — or that formed by the intro- 
 duction of intermediate trellis-sills under the points Q., so that the outer limits of the track may not exceed 8 feet, 
 with a 4 feet 85 inches gauge; the track being laid with rolled-iron edge-rails, 50 lbs. to the yard, in bars 20 feet long, 
 secured throughout by wrought-iron chairs and timber-screws. 
 
 528 string-pieces, (d d,) 20 feet 8 in. long, 5x8 in. section, = . 36,379 ft. b. m. 
 
 1056 joint-sills, (a and b,) Hi feet long, 3x6 in. section, = . 36,960 " " 
 
 2112 intermediate sills, (/,) 6i ft. long, 3x6 in. section, = . . 20,592 « " 
 
 Total quantity of timber per mile of track, 93,931 ft. b. m. at $10,= $939 31 
 
 5280 white-oak pins, (pjo, &c.) 9 inches long, 1^ in. dia. at 5 ct 26 40 
 
 528 joint-chairs, at (S,) 6 lbs. each, 3,168 lbs. 
 
 528 middle-chairs, at (T,) 2i lbs 1,320 « 
 
 1056 intermediate chairs, at (Q,,) 1-75 lbs 1,848 " 
 
 2112 joint and scarf chair-bolts for wood, at (S and T,) I5 lbs. . . 2,640 " 
 2112 intermediate chair-screws, at (Q,) I lb. ..... 1,584 " 
 
 Total quantity of iron fastenings per mile, ....... 10,560 lbs. at 9 cts. $950 40 
 
 Workmanship laying 1760 yards of track, at 40 cts. per yard, 704 00 
 
 $2,620 11 
 Mineralising 7880 cubic feet of timber at 45 cts. 354 60 
 
 78| tons of rolled-iron rails, 50 lbs. to the yard at $50 a ton, . . . . '. . . . 3,925 00 
 
 Cost of one mile complete B,— IV, $6,899 71 
 
 Iron Trellis.— Plate IV. 
 
 Estimate of the quantity of materials and cost of constructing one mile of the wrought-iron trellis railway struc- 
 ture laid in asphaltic mastic, or cement of Seyssel, as represented. 
 
 IRON TRELLIS WORK. 
 
 1056 joint and scarf trellis tie-bars, («' and b',) 16 feet 10 in. long, 2 in. wide, 
 
 and 2 in. in thickness, = ......... 22-2 tons. 
 
 2112 intermediate trellis bars, (/,) 5 feet 9 inches long, I5 inches wide and 
 
 I in. thickness, = 10.25 " 
 
 Weight of the trellis-bars per mile, 32-45 tons at $50, = $1,622 50 
 
 1584 joint, &c. trellis-rivets, I in. diameter, = 342 lbs. 
 
 5280 sub-trellis rivets, f in. diameter, = 605 " 
 
 Weight of rivets for trellis, 947 lbs. at 7 cts. = 66 29 
 
 1056 under-bars at joints and scarfings, (S and T,) at 2\ lbs. each, . . 2,376 " 
 1056 intermediate under-bars at (Q,,) Ij lbs. ...... 1,584 " 
 
 528 joint-chairs at (S,) 6 lbs. each, 3,168 " 
 
 528 middle, or scarf-chairs, at (T,) 2^ lbs. 1,320 « 
 
 1056 intermediate chairs, at (Q,) 1| lbs . . 1,848 " 
 
 Weight of chairs and under bars per mile, 10,296 " at 9 cts. = 926 64 
 
 2,615 43 
 
48 
 
 ESTIMATES OF THE PRIMITIVE, DIAMOND, AND IRON TRELLIS. 
 
 Brought over, . . 
 
 2112 joint, and middle, chair-bolts and nuts 65 inches long, I in. diameter, 1 lb. 
 
 each, = 2,112 lbs. 
 
 2112 intermediate bolts and nuts, 65 in. long, and f in. diameter, I lb. = . 1,584 " 
 
 Weight of screw-bolts per mile, 3,696 " at 10 cts. 
 
 2,615 43 
 
 369 60 
 
 Cost of all the iron work per mile, exclusive of rails, ......... $2,985 03 
 
 17,600 superficial feet of asphaltic cement sills, 4 inches thick, laid and cemented on the iron trellis, at 
 
 30 cts. a foot, 5,280 00 
 
 528 timber string-pieces 20 feet 8 inches long, 5 by 8 in., equal 36,379 ft. b. m. at f 10, . . . 363 79 
 
 78i tons of rolled-iron rails, 50 lbs. to the yard, $50 a ton, 3,925 00 
 
 Workmanship laying track at 40 cts. a yard, 704 00 
 
 Mineralising 3032 cubic feet of timber, at 4i cts 136 44 
 
 $13,394 26 
 
 Cost of one mile complete, C, — IV, ........... 
 
 The last track described would seem to have fair claims to a character for permanence; and 
 as regards cost, it does not equal those on the English railways, nor much exceed the gene- 
 rality of good tracks in this country; it will be seen, moreover, that more than one-third of 
 the estimate is placed to the account of Asphaltum cement, which, being a foreign mineral in 
 but very limited use here at present, bears the high price named in the estimate when laid in 
 street pavements. There can be no doubt, however, that this cement, or an equally good 
 native substitute, could be furnished at less than half the price in the estimate, probably 
 one-third, when wanted in such large quantities as would be required for railways. There 
 are places in America where bitumen may be readily obtained in large quantities conve- 
 nient to shipping, and, by the addition of lime, it can be rendered of suitable consistence. 
 Or a cement formed of lime, (perhaps hydraulic would be best,) and coal tar, would probably 
 answer quite as well. 
 
 Water cements, from their rigid and brittle nature, will not at all answer the purpose. 
 
 The natural asphaltic rock extracted from the mines of Pyrmont, near Seyssel, in the 
 department of I'Ain, and other parts of the Jura mountains, contains, it appears, in addition 
 to a small portion of aqueous matter, about 90 parts of the carbonate of lime, and 10 of 
 bitumen; this is reduced by heat to a mastic cement, and united with mineral bitumen, or 
 common pitch, is then used to cement sand, gravel, or broken stone in the form of slabs for 
 paving; these slabs, when laid in place, are united at their edges by some of the same hot 
 mastic cement. 
 
 Summary of the Estimated Cost of the Trellis Tracks. 
 
 Estimate A. — Plate I. , = 
 
 B. 
 
 
 I., 
 
 = 
 
 C. 
 
 
 I., 
 
 z= 
 
 D. 
 
 
 I., 
 
 = 
 
 A. 
 
 
 
 = 
 
 B. 
 
 
 
 = 
 
 C. 
 
 
 
 = 
 
 D. 
 
 
 
 = 
 
 E. 
 
 
 
 = 
 
 7,229 90, page 27. Estimate F.- 
 
 -Plate II., = 
 
 
 $3,693 24, 
 
 page 37 
 
 6,861 54, " 
 
 29. 
 
 ' A. 
 
 " III., = 
 
 
 6,138 43, 
 
 " 40. 
 
 6,472 29, " 
 
 29. 
 
 ' B. 
 
 " III., = 
 
 
 6,516 82, 
 
 " 40. 
 
 6,086 54, ' 
 
 29. 
 
 ' C. 
 
 " 111.,== 
 
 
 6,042 85, 
 
 " 41 
 
 6,870 70, ' 
 
 36. 
 
 ' D. 
 
 " III.,= 
 
 
 6,817 85, 
 
 " 41. 
 
 6,838 33, ' 
 
 36. 
 
 ' E. 
 
 " IIL, = 
 
 
 4,480 41, 
 
 " 41 
 
 6,063 33, • 
 
 36. 
 
 ' Primitive Trellis.— 
 
 -IV. 
 
 = 6,785 09, 
 
 " 47. 
 
 5,895 72, ' 
 
 ' 37. 
 
 ' Diamond Trellis. — 
 
 -IV. 
 
 = 6,899 71, 
 
 " 47 
 
 5,477 82, ' 
 
 ' 37. 
 
 ' Iron 
 
 Trellis.— IV. 
 
 
 = 13,394 26, 
 
 " 48 
 
ON THE PRESERVATION OP TIMBER. 49 
 
 ON THE PRESERVATION OF RAILWAY TIMBER FROM DECAY. 
 
 ' The rapid decay of railv/ay timber in our climate, exposed as it is to the sun, rain, air, damp earth and vegetation, 
 has caused much solicitude to be felt on the subject of its preservation; but as yet little has been done in a regular 
 systematic manner, by means of the certain process, so fully established in Europe, of mineralising it by means of 
 corrosive sublimate; or, as it is there commonly called "Kyanising," after the name of him who by his address and 
 perseverance succeeded in bringing the neglected process into general use. Wherever experiments have been 
 made on the railways in the Northern, Southern, or Middle States, they are officially reported to have fully con- 
 firmed the acounts we have from Europe of the astonishing preservative powers of the Bichloride of Mercury, one 
 of the chemical combinations of common salt and mercury, well known, under the name of corrosive sublimate, as 
 a deadly poison which arrests both animal and vegetable life, forming with their substances a new chemical combi- 
 nation that strongly resists all future change at natural temperatures. 
 
 On other railways desultory efforts to preserve the timber have been made at some considerable expense, but 
 strange as it may seem, with the recorded experience of nearly a century against it, and without a single fact in its 
 favour that I am aware of, lime has been the substance chiefly employed; lime — the powerful alkaline solvent of 
 both vegetable and animal substances, the active fertilizer of the soil, which rapidly decomposes the dead vegeta- 
 ble while it quickens the growth of the living, not unlike in its action to the gastric juice in the stomachs of ani- 
 mals, and entering itself as a component part of the reproduced vegetables, — employed with the view of stopping 
 decomposition and reproduction, under circumstances and in a proportion most likely to produce both results, accord- 
 ing to our observation of natural laws. 
 
 There have been hot and cold solutions of lime used, also solutions mixed with salt, and even quick lime. 
 
 Feeling desirous that some cheap and effectual process might be discovered, I have endeavoured to ascertain on 
 what grounds the use of lime had been recommended, but have been unable to trace it to anything better than a 
 vague popular notion in its favour. Every fact that has come to my knowledge has been decidedly against its use, 
 which accords with inductive reasoning from cause to effect.* 
 
 Not so with some of the metallic salts, and another substance found in nature, creosote, which chemistry has 
 separated from its combinations. To the latter, which is known to exist in the peat, is attributed the perfect pre- 
 servation of the timber so frequently dug from the bogs of Ireland and other countries, bearing evidences of having 
 lain there for ages. And it is now well known that creosote was the substance employed by the ancient Egyptians, 
 more than two thousand years ago, to preserve the bodies of their dead, which, with the wooden cases containing 
 them, have remained perfectly sound, as we see, to the present day. 
 
 The modern process for preserving timber, which has obtained such extensive use under the name of "Kyanising," 
 it is conceded, was first proposed for the purpose by the celebrated chemist Sir Humphrey Davy. 
 
 In a treatise on the preservation of timber, published by William Chapman in 1817", he states that Sir H.Davy 
 had recommended the use of corrosive sublimate, and from his own experiments he gave it a decided preference 
 over every other salt; stating that a less quantity than an ounce to the gallon of water would not answer the pur- 
 
 * Dr. Eirkbeck states, that "about the year 1769, Mr. Jackson proposed a very complicated lixivium, in which vegetable bodies were 
 to be immersed to protect them against decay. With total disregard of all chemical principles, he composed a lixivium of the Muriate 
 of Soda, [common Salt,] Epsom Salts, Lime, Potash,, Saltwater, &c. He had an opportunity of trying it on the wood of several frigates 
 and other vessels in the navy; but the result was that those vessels built with wood prepared according to his method were less durable 
 than those which had been ordinarily constructed. 
 
 "Shortly afterwards a person named Lewis attempted to accomplish the preservation of timber from decay by means of lime. The 
 Amethyst Frigate was assigned for his experiment; but decay was found to attack the vessel more rapidly than in ordinary cases. 
 
 "All are aware [says Dr. B.] that when the dead among human beings are to be rapidly dissolved or disorganized, quick lime is thrown 
 into the pit in which they are deposited, not for the purpose of protecting them from decay, but for the very reverse. Yet this is the 
 substance which, upon various occasions, and perhaps more extensively than any other, has had its preservative powers boasted of by 
 different writers." 
 
 7 
 
50 PRESERVATION OF TIMBER FROM DECAY 
 
 pose. This is equal to one pound of sublimate to 16 gallons of water, and the solution now in general use is 1 
 pound to 15 gallons. 
 
 Chapman also experimented with tlie sulphates of copper and iron; and Tredgold says recommended boiling the 
 timber in a solution of the sulphate of iron, (green copperas.)* 
 
 We have also Doct. Ure's authority that Sir H. Davy had, several years anterior to 1821, used and recommended to 
 the Admiralty and Navy Board the use of corrosive sublimate; but either from doubts of its efficacy, or more proba- 
 bly from a vague apprehension which seems to have been entertained, that it might prove deleterious to those using 
 the process, or that the prepared timber would exhale a dangerous atmosphere, its use remained circumscribed. 
 
 These groundless fears, as they have proved to be, had the effect of retarding its use, except amongst anatomists, 
 to whom it was a well known preservative of the most delicate organic tissues and parts, even those most liable to 
 putresence, such as the brainj which this metallic solution hardens in a remarkable manner, and saves from natural 
 decay.t 
 
 Although the process was known and had been published in several respectable works, which were widely circu- 
 lated, ten years or more had elasped without much, if any considerable use having been made of it for the purpose 
 of preserving timber; and, in fact, it seems to have been in a great measure forgotten, or, more probably, was deemed 
 unworthy of trial by those who controlled the means of testing its utility, when in 1828 Mr. John H. Kyan pro- 
 posed to the Admiralty to prepare timber in such a way as should resist dry rot or other decay. He was directed 
 to prepare a 12 inch cube of English oak, which he accordingly did, leaving the sap wood on the four corner?! 
 This block was deposited in the fungus pit at Woolwich Dock-yard, where it remained three years, and on re-open- 
 ing the pit, in July 1831, it was found to be perfectly sound. This astonished all those who had witnessed the pre- 
 vious action of the pit, as it seems that no preparation hitherto tried had been able to preserve timber for a similar 
 period; and, so confident were the officers in charge, of the destructive powers of the pit, that they deemed it 
 unnecessary to place an unprepared specimen of the same timber with it; particularly, as Sir T. B. Martin states, 
 there were other specimens placed in the pit prepared in various ways with which it could be compared. All these 
 pieces were destroyed by decay, and some of them, that apppeared to have been coated with lime, were covered 
 with fungi, one of them having a mushroom as large as a hat-crown growing out of it. The man who had charge 
 of the pit produced a register of the experiments that had been made, in corroboration of his statement that he had 
 never before seen timber taken out of the pit sound. 
 
 A piece of prepared Canada oak with an unprepared duplicate were now placed in the pit, and similar pieces put 
 in the lining of a dock at Woolwich infected with dry rot; but the cube of English oak remained in a loft of the 
 dock-yard for 15 months, when it was returned to the pit. In Feb. 1833, an inspection was made in presence of 
 Mr. Faraday, Mr. Lockhart, and Mr. Ferrell, an architect from Dublin, who took specimens; the unprepared pieces 
 showed signs of decay, whilst the prepared were in the best possible state. 
 
 The pit was again opened on the "19th July, 1833, in the presence of Commodore Warren, Mr. Jephson, Mr. 
 Benson, and several other gentlemen. The cube came out of the pit perfectly sound, it was sawn through the 
 
 * Chapman's work will be found in the Library of Congress, and probably in many other places, as it was on sale in Baltimore about 
 10 years ago. Tredgold's Carpentry was first published in 1820 or '21, and a second edition in '28; page 199 and 200 of the latter treats 
 of the above, and on the use of corrosive sublimate, as recommended by Davy. He also reasons on the use of lime, tar and other sub- 
 stances in his clear analytical style. 
 
 t "A solution of corrosive subliaiate has been long employed for the preservation of soft anatomical preparations. By this means 
 the corpse of Colonel Moreland was embalmed, in order to be brought from the seat of war to Paris. His features remained unaltered, 
 only his skin was brown, and his body was so hard as to sound like a piece of wood when struck with a hammer. 
 
 "In the valuable work upon the dry rot, published by Mr. Knowles, Secretary of the Committee of Inspectors of the Navy, in 1821, 
 corrosive sublimate is enumerated among the chemical substances which had been prescribed for preventing the dry rot in timber; audit 
 is well known that Sir H. Davy had, several years before that date, used and recommended to the Admiralty and Navy Board corro- 
 sive sublimate as an anti-dry-rot application. It has since been extensively employed by a joint-stock company for the same purpose, 
 under the title of Kyan's patent." — Dr. Ure's Dictionary of Arts, Manufactures, and Mines. Article Mercury, Bichloride of, page 811, 
 
MINERALISED BY CORROSIVE SUBLIMATE. 
 
 51 
 
 middle and split, and proved to be in a perfectly sound statej a certificate was affixed to the block by all the gentle- 
 men present." 
 
 The Fungus Pit at Woolwich had been in use since 1815; and the block of oak in question, it was proved, "had 
 been five years surrounded by decaying matter — by the decaying property of the pit — by the heat generated by that 
 decay — and by the quantity of carbonic acid which always existed in the pit, and escaped in great quantities when- 
 ever the doors were opened." 
 
 It seems, according to a report of Sir R. Stepping, Surveyor of the Navy, who reported the block sound, tiiat up 
 to the date of his report, (Dec. 20th, 1831,) Mr. Kyan had not explained the means he had made use of to pre- 
 serve the wood. On the 31st of March, '32 he patented the process; the veil of mystery was withdrawn, and it 
 was then seen that the effects which created such astonishment were produced by the neglected process of Davy. 
 
 Mr. Kyan's own statement is that he first applied corrosive sublimate to timber in 1825, and in March 1828 he 
 brought the subject to the consideration of the admiralty. 
 
 The success attending the process in the hands of individuals, together with the uniformly favourable results 
 which continued to be obtained, in the systematic experiments carried on by the Engineer Officers at the Royal 
 Arsenal, Woolwich, had drawn public attention to the subject, and tanks for the saturation of timber, &c. were 
 erected in many places. 
 
 In April 1835, the Admiralty appointed a Board of Commissioners to inquire into Mr. Kyan's process for- pre- 
 serving timber; their report, together with the minutes of evidence, were printed by order of the House of Com- 
 mons, 9th July, '35; from which, and other authentic sources, I select a few passages on the preservation of timber 
 in similar situations to that in railways, which may be interesting to those who have not seen the more full accounts 
 of the result of the experiments. 
 
 Sir Robert Smirke was one of the first architects to use the process; he had a couple of posts put up under a 
 dropping eave, and both were exposed to the same actions. After a certain time one decayed; the other still stands. 
 In 1825 he put some English oak paling to a house on Stanmore Hill, which was completely gone in four or five 
 years; he replaced it in the autumn of '32 with p)-epared unseasoned yellow pine, which remains quite sound; some 
 yellow pine paling put up the year before, unprepared, began to fail in a year, and is now quite gone. 
 
 He thinks it will not supersede the usual time for seasoning timber for joiners' work, but timber of large scant- 
 lings may be used the sooner for it. Timber, in his opinion, is not reduced in strength by the process; but cannot say 
 to what extent the mercury penetrates the timber; he thinks, however, that it does penetrate, and stated that of the 
 tanked wood he had used at Stanmore, in many instances a large piece of the prepared wood of a foot or more had 
 been cut off from the ends; in other instances it had been morticed quite through, and in some instances a piece had 
 been cut out longitudinally. In each case the interior wood, to the depth of at least three inches from the surface, 
 had been exposed. 
 
 Sir Robert Smirk, when examined before the Committee on Timber Duties, stated: — '-I have applied Kyan's pro- 
 cess to Yellow Canadian Pine, about three years ago, and exposed that wood to the severest tests I could apply, 
 and it remains uninjured, when any other, oak or Baltic wood, would certainly have decayed if exposed to the same 
 trial and not prepared in the same manner. 
 
 "I took pieces cut from the same log of Yellow Pine, from Poplar, and from Scotch Fir; these I placed first in-a 
 cess-pool, into which the waters of the common-sewers discharged themselves: they remained there six months: 
 removed from thence and placed in a hot-bed of compost, under a garden frame, they remained there a second six 
 months: they were afterwards put into a flower-border, placed half out of the ground, and I gave my gardener direc- 
 tions to water them whenever he watered the flowers; they remained there a similar period of six months. I put 
 them afterwards into a cellar where there was some dampness, and the air completely excluded; they remained 
 there a fourth period of six months, and were afterwards put into a very wet cellar. Those pieces of wood which 
 underwent Kyan's process are in the same state as when I first had them; and all the others, to which the process 
 had not been applied, are more or less rotten, and the poplar is wholly destroyed." 
 
 "As another example of the eifects of the process, I may mention that about two years ago, in the basement story 
 
52 PRESERVATION OF TIMBER FROM DECAY 
 
 of some chambers in the Temple, the wood flooring and wood lining of the walls were entirely decayed, from the 
 dampness of the ground and walls, and to repair it under such circumstances was useless. As I found it extremely 
 difficult to prevent the dampness, I recommended lining the walls and floor with this prepared wood, which was 
 done; and about six weeeks ago I took down part of it to examine whether any of the wood was injured, but it was 
 found in as good a state as when first put up. 
 
 "This preparation of Mr. Kyan's resists all rot — 'I cannot rot it,' added Sir Robert Smirke." 
 
 The following report of Mr. S. Beazley, Architect, will show the effectual manner in which the prepared posts 
 and palings of the Regent's Park, London, have been preserved by corrosive sublimate: — 
 
 "At the commencement of the year 1836, I surveyed and accurately examined the posts and palings in the 
 Regent's Park, for the purpose of ascertaining the comparative states of those timbers which had been prepared by 
 Kyan's Patent, and those which had not been submitted to the process of solution. In my report of that period I 
 stated that indications of decay were already perceptible in most of the unprepared timbers, both at the bottom of 
 the posts, and in those arris edges and ends of the paling which were placed in or had come at all in contact with 
 the earth, while those timbers which were marked as having passed through the solution were quite free from any 
 such symptoms. I now beg leave to state that I have this day, [March 24, 1838,] after a lapse of two years and 
 a quarter from my previous survey, again accurately examined several of the same posts and paling, digging away 
 the earth from the foundations for that purpose; and find that the symptoms of decay, mentioned in my preceding 
 report as having commenced in the unprepared timber, have so considerably increased as to have rendered the bot- 
 tom of the posts completely rotten to a depth of from one to two inches, and that in several instances fungi have 
 been the consequences of the decay; while I find the prepared timbers which are in the earth sound and in the same 
 state, with the exception of mere discoloration upon the surface, probably arising from the damp state of the earth 
 at the time of its removal. As a farther proof of the difference existing between the unprepared and the prepared 
 timber, we could cut with the greatest ease large pieces from the former with the spade, without using any force, 
 while it required great exertion to chip off very small pieces from the latter." 
 
 In the 1st vol. of the Professional Papers, published by the British Royal Engineers, a very interesting account 
 is given of the experiments made at the Royal Arsenal, Woolwich, on the effect of Kyan's process in preserving 
 various specimens of seasoned and freshly cut green timber, cordage and canvass from rotting. I select the follow- 
 ing from page 144. 
 
 "2nd. Of the pieces of wood partly driven into the ground under the eaves of a building, and exposed to the 
 united action of sun, rain, and damp earth, we all agreed, that all the five pieces of oak, ash, elm, Memel fir, 
 and American fir, 'prepared' with the patent, are quite sound; whilst of the duplicate pieces 'unprepared,' the elm 
 and ash were rotten, and the progress of decay had commenced in the three others, the oak being the least affected. 
 
 The woods used in the foregoing trials had, previously to being put down, been seasoned two years. 
 
 "We then proceeded to examine what you, at Mr. Terry's request, had placed under the same test as the last 
 described woods; it appeared to be the most severe trial, viz: — 
 
 "A piece of oak, five feet long, three inches diameter. 
 
 "A piece of ash, two feet five inches and a half long, six inches and three-quarters diameter. 
 
 "A piece of elm, five feet one inch long, three inches and one-eighth diameter. 
 
 "All of which came here quite in a green state, and with the bark and some leaves on them, and after being split 
 down the middle and marked, half of each specimen of wood was returned to be saturated with the patent, and when 
 sent back again the whole were put down 31st March, 1835. 
 
 "They were taken up a few days ago to dry, and we find at the end of the year and a half, that the 'prepared' 
 pieces, even to the preservation of the bark and sap, are perfectly sound, and the 'unprepared' quite rotten." 
 
 I should state that the foregoing inspection and report are dated 28th Sept. 1836; and that the seasoned specimens 
 had then been down three years. 
 
 The process has been found equally efiicacious in the preservation of ship timber, even where it is exposed to the 
 constant saturation and wash of the bilge-water, and external wash of the sea. It has the remarkable and highly 
 
MINERALISED BY CORROSIVE SUBLIMATE. 
 
 53 
 
 beneficial eflfect of keeping tiie ship and bilge-water free from the usual prutrescent effluvia sooflensive in new ships 
 and probably injurious to the health of those on board. 
 
 The first ship built wholly of mineralised timber was the "Samuel Enderby," of 420 tons, built at Cowes in the 
 Isle of Wight, of seasoned oak, with the exception of some of her upper deck beams, which were felled for the 
 purpose and saturated green, having sap on the corners, which was found to be as hard and sound as the other wood 
 on her return from the South Seas. 
 
 This ship was launched in August 1834, and sailed the following October for the South Sea Fisheries, from which 
 she returned to England in March, 1837. 
 
 During her absence she was a great part of the time sailing under the line. The crew, 32 in number, continued 
 unusually healthy; and the ship did not require caulking to the usual extent. The masts, yards, cordage and can- 
 vass were also mineralised; but the cordage and canvass of this ship did not realize the expectations that were 
 entertained of it: the captain supposes the solution was too strong, as it seems to have succeeded in several other 
 cases. The proofs of the efficacy of the process as regards timber have been so multiplied as to place it beyond a 
 doubt; and my intended limits, which I have already far exceeded, compel me to pass over much that is interest- 
 ing in the Reports of Lloyd's Surveyors of shipping, and from other sources. I must confine myself to a few more 
 brief extracts; the first is from Capt. Lisle's Report: — 
 
 '^Timbers. — As regards the timbers, the ship is perfectly sound in every respect, and shows no symptoms or 
 indications of decay in any part throughout the whole ship; and it is my belief that the timber and plank have 
 shrunk less than any new ship I have witnessed. Being at Cowes, inspecting the ship from the time her keel was 
 laid until she was launched, and the greater part of her deck-beams being made from green timber just felled and 
 afterwards submitted to the process, they are now perfectly sound, and firm as if cut from the most seasoned tim- 
 ber. These facts I consider to be strong evidence in favour of the process; and I am so satisfied myself of its 
 good effect on timber, that I should recommend new ships to be built of timber prepared when quite green, and to 
 let it be quite dry before placing it in the ship." 
 
 The owners of the ship state, — "we feel so well satisfied of the beneficial effects of the preparation on the tim- 
 ber, that in any cases where we may have occasion either to build or repair our ships, we shall continue to use it, 
 and ill like manner for masts; in which particular we think that we have derived considerable benefit from its 
 application." 
 
 On the return of this ship a bottle of the bilge-water was sent to Professor Faraday, who states that he found it 
 turbid and saline, but quite sweet in smell, and though carefully examined for mercury could find none in it. 
 
 "The ship 'John Palmer' left London Dec. 13, '33, on a South Sea whale voyage, and returned on 22nd of April, 
 1837, being absent 3 years and 4 months. Previous to her sailing underwent thorough repair; great part of the 
 timber, masts, bowsprits, sails and cordage were saturated with Kyan's Patent. The timber, from what I had 
 observed during the voyage, and at the present time, is in the highest state of preservation. We have been most of 
 our time exposed to a tropical sun, and the planks in the sides have not shrunk in the least. The masts are in a 
 high state of preservation, &c." 
 
 "I am of opinion that the patent has done everything for the canvass that was expected; it has prevented mildew 
 and rot, though not wear and tear." 
 
 The above extracts are taken from Capt. Laurence's statement. It should be remarked that the whole of the 
 ship's ceiling and sleeping berths of the men consisted of this timber. See the reports of Lloyd's surveyors and 
 other documents relating to her. 
 
 The process is stated to have the very important effect of shrinking timber as much in a few hours as a seasoning 
 of years would have done; and of subsequently preventing its warping when exposed to the sun and rain; and, what 
 will be found still more beneficial on railways, of preventing the splitting of the timber, which causes great de- 
 struction at present. 
 
 Doct. Birkbeck, who was one of the Commissioners appointed by the Admiralty to inquire into Kyan's Patent, 
 in a lecture delivered before the Society of Arts, Adelphi, exhibited a piece of green larch, such as was to be used 
 for sleepers in the Southampton Railway. "When it was put into the solution it had cracked in various radial 
 
g^ PRESERVATION OF TIMBER FROM DECAY 
 
 directions — some of tlie openings being large enough to admit a penny-piece. Tiie wood was now rendered perfectly 
 solid, and a slight alteration in the level showed where the fissures had been." 
 
 The rationale of the process is as follows: — It has been ascertained that the disorganization, or decay of timber, 
 commences with the putrefactive fermentation of the albuminous and gummy fluids, which soon extends to the 
 starch and saccharine matter, lodged with the former in greater or less quantities in-the sap vessels and pores of the 
 timber; and as the alburnum, or sap-wood, contains a much larger proportion of these matters, it is the first part to 
 decay. When once decomposition has commenced, the most solid part of the heart timber — the woody fibre, or 
 lignin — offers but little resistance to the universal decomposition which rapidly follows when aided by moisture and 
 a slight increase of temperature, of from 70° to 80°. This will not create surprise when it is understood that the 
 lignin itself, — the most solid and insoluble part of wood, — is composed of 50 parts water and 50 of carbonj or, 
 what amounts to the same thing, of oxygen and hydrogen in the same proportions which form water; and that 
 the other component parts contain a still larger proportion of water, so that there is no deficiency of the materials 
 of fermentation in any part of the wood. 
 
 Corrosive sublimate in addition to its anti-putrescent qualities, has the property of forming with albumen a com- 
 pound which is insoluble in water, consisting of calomel and albumen. 
 
 When wood is steeped in a solution of corrosive sublimate, this insoluble compound of mercury and albumen 
 forms in the pores and vessels of the wood, retaining with it a considerable portion of free sublimate, which Profes- 
 sor Faraday found to equal three-fourths of the amount; but that it required the most thorough disintegration of 
 the wood to remove it; and that it was his opinion, it would gradually continue, by a play of affinities, to penetrate 
 the wood while it continued in a moist state; he, therefore, regarded the excess of the sublimate, in the parts near 
 the surface of timber, as a most important condition. 
 
 Penetration. — Professor Faraday stated before tiie Commissioners, "That he had lately examined a plank cut 
 from the middle of a balk of timber 20 inches square, to see how far the corrosive sublimate had gone in. It was 
 easily found at one inch beneath the surface; and by a very careful examination at four inches beneath the surface; 
 and the process was to digest part of the wood at the spot in very diluted nitric acid, to evaporate the solution con- 
 siderably, and then examine it by the voltaic battery; mercury appeared at the negative pole. He could not assure 
 himself of its presence in the middle of the block by any means whatever. This was a piece of pine wood sent 
 him by Mr. Brunell." 
 
 The test usually employed is the hydro-sulphuret of ammonia, which turns the prepared wood black. 
 
 Dr. Birkbeck states, ''The analysis of the result performed by Fourcroy, and subsequently by Berzelius and 
 others, is, that the bichloride of Mercury has been converted into a protochloride. In that form it combined with the 
 albumen, which being no longer soluble, descends in a visible form with the protochloride. In the change the bichlo- 
 ride loses one proportion of its chlorine. Bichloride of Mercury consists of 200 parts, or one proportion of mer- 
 cury, and 72,, or two proportions of chlorine. One proportional is separated in this process, leaving the protochlo- 
 ride 200 parts of mercury and 36 of chlorine; that is, one proportional of mercury and one of chlorine; and the 
 albumen, being separated along with the protochloride or calomel, descends." 
 
 M. Lassaigne, as quoted by Dr. Dickson in his lecture on the preservation of timber, delivered before the Royal 
 Institute of British Architects the 26th March, 1838, "calculates the composition of the albuminous precipitate to 
 be 6-67 corrosive sublimate and 93.33 albumen, or 1 atom of sublimate and 10 atoms of albumen." 
 
 Dr. Dickson remarks: — "It is impossible for bichloride of Mercury to come in contact with albumen without coagu- 
 lating it. Bichloride of mercury is thus the established test of the presence of albumen, and so delicate is it that the 
 addition of corrosive sublimate to any solution of albumen will indicate the ^oVb' part, by causing a flaky appearance 
 of the fluid. On the opposite hand albumen is the established antidote to poisoning by corrosive sublimate; and, 
 though white of egg is generally preferred, milk, or flour diff"used through water, or anything containing albumen, 
 may be employed with success." 
 
 It would seem from the foregoing that a very minute quantity of corrosive sublimate would be sufficient in ordi- 
 nary cases to preserve timber from decay alone; but in cases where timber has to be protected against insects, or is 
 
MINERALISED BY CORROSIVE SUBLIMATE. 55 
 
 half buried in the damp earth, as that in the structure of railways, exposed more or less to the splitting action of 
 the sun, and to the decomposing action of vegetation, I regard the excess of the corrosive sublimate as highly 
 important. I propose, however, that timber for the interior of buildings, much of that used in shipping, and for 
 many other purposes, instead of being steeped for several days in the comparatively strong cold solution of corro- 
 sive sublimate, should be boiled for a few hours in a very diluted solution, to which pressure might be added in 
 preparing large pieces. 
 
 I should not omit to add, what seems obvious to me, that, to render the process effectual, the Umber should be peV' 
 fectly sound at the time it is steeped, otherwise the nature of the albumen will be changed, and the chemical com- 
 bination cannot take place between the corrosive sublimate and albumen. I am farther inclined to think, that the solu- 
 tion will be found to penetrate the wood much more deeply, and to be otherwise more effectual, if the timber be felled 
 in the spring about the time the sap begins to rise, and steeped as soon as possible thereafter, while the fluids are in 
 motion; and I have no doubt but that in this case also a much weaker solution than any now in use would be found 
 efifectual. 
 
 When timber has been long cut, or is affected in the slightest degree, strong solutions should be used. On some 
 of the English railways a much more rapid and effectual mode of causing the fluid to penetrate the timber, by 
 means of hydraulic pressure, instead of simple saturation, has lately been introduced. It is thus described as in 
 use on the Manchester and Birmingham railway. A cylindrical iron vessel, 30 feet long, and 7 feet in diameter, 
 formed of wrought-iron plates f of an inch in thickness, double riveted together, so as to be able to resist a pressure 
 of 250 lbs. on the inch, is filled with the wooden sleepers, as closely packed as it is possible; the solution of corrosive 
 sublimate is then forced in by one of Bramah's hydraulic pumps, worked by 6 men to a pressure of 170 lbs. to the 
 inch. By this means the timber is more completely saturated in 10 hours than it would have been in some months 
 on the old system. 
 
 Jis to the strength of the solution, with a view to the expense as in use. The solution used at Somerset House 
 consisted of 224 lbs. of corrosive sublimate to 1 ,062 gallons of water, being rather more than 1 lb. of sublimate to 
 5 gallons of water; which latter are the proportions stated in Mr. Kyan's patent; he has subsequently stipulated in 
 his licenses that the solution shall not not be used of less strength than 1 lb. to 15 gallons; which latter are the pro- 
 portions stated to be in use at the Royal Arsenal at Woolwich,* and that the average quantity of corrosive sublimate 
 used was lA lbs. to 50 cubic feet of timber. The corrosive sublimate used at the Arsenal cost 4s. per lb., that at 
 Somerset House 3s. 7d., exclusively of the patent fee. The Admiralty agreed to pay the builder of the 'Linnet' 
 packet, which was the first government vessel built of prepared timber, 30s. extra per 100 cubic feet, which is equal 
 to ST' 9,7^; this was the cost with a solution of the strength of 1 lb. of corrosive sublimate to 5 gallons of water, 
 and includes the patent fee, and all extra labour in handling, &c. 
 
 I have previously stated that Mr. Kyan obtained a patent in England, bearing date the 31st March, 1832; some 
 years afterwards application for a patent was made in the United States, which was refused, I believe, on the ground 
 chiefly of want of originality. An act of Congress was, however, obtained, dated 31st of May, 1838, authorizing 
 the Commissioner of Patents to issue a patent to Angier March Perkins and John Howard Kyan, which was accord- 
 ingly issued the 23d June, '38. This act removes the limitation of time within which the patent should have been 
 applied for after the date of the foreign patent, and leaves it to the judiciary to settle the questioji of originality of 
 invention. 
 
 In presenting this highly important subject to the consideration of the American public, and advocating its use, 
 I have felt it incumbent on me to state such historical facts connected with the subject as were within my knowledge, 
 that Mr. Kyan's claims might be fairly understood. 
 
 It is thought by some persons that the increased consumption of mercury would greatly enhance its cost, and that 
 an adequate supply could not be obtained to mineralise the timber of railways. On looking into the subject I am 
 satisfied that after a very little time the reverse would be the case, although there are some unfavourable circum- 
 stances at present. Doct. Ure in his new Dictionary of Arts, Manufactures, and Mines, informs us that the mer- 
 
 * See the 1st vol. of Professional Papers, published by the Royal Engineers, page 131. 
 
56 PRESERVATION OF TIMBER FROM DECAY 
 
 curj mine of Idria in Friuli, might easilj be made to yield 600 tons British per annum; but in order to uphold the 
 price of the metal the Austrian government has restricted the product to 150 tons. 
 
 The rich old mines of Aimaden in Spain, since 1827 have produced 110 tons per annum of mercury; and Doct. 
 Ure says that there is nearly as much more let escape into the atmosphere and lost, to the great injury of health in 
 the workmen and inhabitants, by the old and barbarous practice of aludeis, which have experienced no improve- 
 ment since the time of the Moors; his words are, "I am confident that their product might be nearly doubled, with 
 vast economy of fuel, labour, and human life." 
 
 Pliny has recorded two interesting facts: 1, that the Greeks imported red cinnabar from Aimaden 700 years 
 before the Christian era; and 2, that Rome in his time annually imported 700,000 lbs. from the same mines. 
 
 The mercury, as in most other cases, is lodged in a bed of bituminous shale, which is from 14 to 16 yards in 
 thickness, extending from the town of Chillon to Almadenejos; and contains nearly 20 per cent, of mercury. Near 
 Aimaden are the celebrated mines of Las Cuebas Almadenejos, the product of which was formerly appropiated 
 exclusively to working the gold mines; the present yield is not given. 
 
 Those of the Bavarian Rhine provinces yield from 40 to 55 tons per annum. 
 
 The Hungarian and Bohemian mines, &c. have averaged from 30 to 40 tons per annum for many years. 
 
 The present yield of the above European mines, is, therefore, about 372 tons per annum of metallic mercury; 
 which, according to Doct. Ure, who devised an improved distillatory apparatus for some of the German mines, might 
 by this means alone be nearly doubled. 
 
 The mines of Guaneavelica in Peru, explored in 1570, yielded to the year 1800, = 53,700 tons. About the 
 beginning of the century the annual product was from 170 to 180 tons. This with the product of other South 
 American mines was employed in working the gold mines. There are also red cinnabar mines in Yunnan China, 
 from which mercury was at one time obtained to work the South American gold mines. 
 
 It is stated here by the dealers that the Messrs. Rothschilds have now the sole monopoly of the Spanish and 
 Austrian mines, and that they have, in the last two or three years, advanced the price from 42 or 50 cents per pound 
 to SI 05, its present price. Manufacturing chemists charge 20 cents per pound for converting the metal mercury 
 into corrosive sublimate, furnishing the salt, sulphuric acid and fuel, and returning pound for pound. It may not 
 be generally known, however, that 100 lbs. of metal mercury will make 136 lbs. of corrosive sublimate; so that 
 instead of a charge of S20 per 100 lbs., they actually receive $57 80. Whether this charge is exorbitant or reason- 
 able I cannot say. 
 
 If we suppose 200 miles of railway track to be laid every year in the United States, and that, including turnouts 
 and other timber to equal 10,000 cubic feet per mile, which will require 300 lbs. of corrosive sublimate, the increased 
 demand for mercury will only amount to 20 tons per annum. 
 
 It is believed that there are other metallic salts which will have the effect of coagulating the albumen, and thus 
 preserve the timber. Doct. Earle, of this city, has a patent for boiling timber in a solution of the sulphates of cop- 
 per and iron, which he states, in his publications, to which I beg leave to refer the reader, will have this effect, and 
 only cost from IJ to 2 cents the cubic foot. 
 
 I have previously stated that to the presence of creosote was attributed the perfect preservation of the timber 
 found in the bogs; it is also in use, in connection with coal tar, for the preservation of timber on some of the Eng- 
 lish railways; and the following account of some recent experiments made in France with this substance, in conec- 
 tion with a metallic salt, give promise of results that seem likely to prove highly beneficial to railways, as the mate- 
 rials employed are very cheap and may be prepared at any place in our own woods. Indeed, the pyroligneous acid 
 may be made from the tops, limbs, and waste wood, and the charcoal will nearly pay the whole expense. 
 
MINERALISED BY PYROLIGNATE OF IRON. 57 
 
 M. BOUCHERIE'S PROCESS FOR PRESERVING TIMBER BY MEANS OF THE 
 
 PYROLIGNATE OF IRON. 
 
 I am indebted to Professor Frazer for the following sketch of M. Boucherie's plan for the preservation of timber, 
 as detailed in the leading article of the Annales de Chimie for June, 1840; a full translation of which article will 
 be published in the August and succeeding numbers of the Journal of the Franklin Institute. 
 
 The chief aims of M. Boucherie's process are: — 
 
 1st. To protect wood against wet and dry rot. 
 
 2nd. To increase its hardness. 
 
 3d. To preserve and develop its flexibility and elasticity. 
 
 4th. To prevent the play which it experiences, and the resulting separation, when after being worked up it is 
 exposed to atmospheric changes. 
 
 5th. To diminish very much its inflammability and combustibility. 
 
 6th. To give to it varied and permanent colours and odours. 
 
 The first result which M. Boucherie obtained from his experiments was the establishment of the fact, that "all 
 the alterations to which wood is subjected arise from the soluble matters which it contains." Having then ascertained 
 the impossibility of separating these soluble matters by mere washing — he turned his attention to such chemical re- 
 agents as would render them insoluble and therefore inert, and found that this change could be effected by all salts 
 which have an insoluble metallic base. Seeking then, among this class of substances for that which should combine 
 most advantageously the strong preservative action with economy — he found these two important conditions best 
 fulfilled by the use of the impure Pyrolignate of iron, made by the action of the acid procured from the distillation 
 of wood upon iron filings or any small pieces of that metal. 
 
 The advantages of this salt are — 1st. It may be procured at a cheap rate. 2nd. The oxide of iron forms stable , 
 combinations with almost all organic substances. 3d. Its acid, (acetic,) has no corrosive properties and is volatile. 
 4th. It contains the greatest proportion of creosote which any aqueous liquor can dissolve, and there is now no doubt 
 but that this substance protects all organic matters powerfully against every species of decay. 
 
 But the principal difference between M. Boucherie's process, and those of other experimenters upon this subject, 
 consists in the manner in which his preservative liquor is diff"used throughout the substance of the wood. After a 
 variety of experiments in various ways, Mr. B. was led to investigate the natural absorption of growing wood, and 
 came to the following interesting and important result. "If a tree of great height be cut, and its foot plunged, 
 within a reasonable time, into a saline solution, whether weak or concentrated, a strong aspiration is exercised by the 
 tree upon the liquid, which thus penetrates its tissue, and finally reaches the extreme height of the trunk and even 
 the terminal leaves — if we are careful to furnish a sufficient supply of the liquid. 
 
 Thus in six days, in the month of September, a poplar tree, 92 feet in height and 16 inches in diameter, the foot 
 of which was plunged about 8 inches into pyrolignate of iron, was entirely penetrated by the liquid, of which it 
 absorbed the enormous quantity of 10 cubic feet, (about one-tenth of its cubic capacity.) A variety of interesting 
 facts and experiments follow, the most important of which is that in all woods there is a central portion which does 
 not absorb any liquid by this means of penetration, and which will probably resist all other methods of impregnation. 
 
 In reference to other qualities M. Boucherie finds that the hardness of wood prepared in this way is more than 
 doubled, so that the workmen complained much of the difficulty of working it. 
 
 In regard to the elasticity of timber M. Boucherie deduces from his experiments, that 
 
 1. The flexibility and elasticity of wood are generally proportionate to the moisture which it contains — that these 
 qualities remain only while this moisture lasts — and that the existence of these qualities is evidence of the presence 
 of moisture even in the driest and oldest woods. 
 8 
 
58 PRESERVATION OF TIMBER FROM DECAY, &c. 
 
 2. That in numerous exceptions it appears to depend upon the constitution of the wood itself, and is probably 
 frequently owing to the presence of alkaline salts in the wood. 
 
 Following up these determinations, by causing the wood to absorb a deliquescent salt, he succeeded not only in 
 maintaining its original elasticity, but in developing a degree of pliability which it does not possess when first cut. 
 Thus pieces of pine wood, 2 feet in length and Jg-th of an inch in thickness, could be so twisted in the direction 
 of their length as to form a complete helix, or could be bent into three concentric circles without breaking. M. 
 Boucherie at first used chloride of calcium, (muriate of lime,) for this purpose, but now proposes, on account of its 
 greater cheapness, the mother waters of salt marshes, which are chiefly composed of deliquescent chlorides. Wood 
 prepared in this way had not had its elastic properties impaired after being kept 18 months. 
 
 In regard to the shrinking and warping of wood, M. Boucherie's experiments led him to the conclusion that 
 these effects, (or such of them as were due only to the drying of the wood,) began to show themselves only at 
 an advanced stage of the drying and when the wood was about to lose the last third of its moisture. Continuing 
 his experiments he found that these effects were entirely prevented by preserving to the wood this last portion of 
 its moisture, by causing it to absorb a deliquescent salt — as in the process for preserving the elasticity. For the 
 purpose of testing his results he caused to be prepared tables of considerable size and slight thickness, and found 
 that after a year, whilst those made of the prepared wood remained unchanged, those made of the timber in its 
 natural state were warped in an extraordinary degree. There are many other observations of great interest both 
 on account of their direct practical value, and from the light which they throw upon vegetable physiology. But 
 these, in reference to the preservation of timber from rotting — the maintenance of its elasticity and the prevention 
 of shrinking and warping, I judged would be of the most importance for your purpose. 
 
 I remain yours, &c. 
 
 JOHN F. FRAZER. 
 
 Ji statement showing the number of superficial feet of bearing surface per mile of single track, by which the 
 folloiving railway structures rest upon the ballasting or road-bed. 
 
 Liverpool and Manchester, England, stone blocks 2 feet square, 3 feet from centre to centre, = . 14,080 feet. 
 
 Great Western Railway from London to Bristol, = . . . . ..... 14,344 " 
 
 Baltimore and Susquehanna, on mud-sills 3x11 in. = 9680 feet, + (2112 transom sills 7i ft. long, 
 
 6x8 in. =) 7975 feet, 17,655 " 
 
 Baltimore and Washington, on mud-sills 6 in. wide = 5280 ft. + 6160 feet of round transoms, . 11,440 " 
 Baltimore and Port Deposit, mud-sills 7040 feet, round transom ties, not intended as bearers, = 4868 
 
 feet,= 11,908 « 
 
 Baltimore and Ohio, proposed track No. 1 of 1838, mud-sills 7040 feet, -j- (transoms 7 feet long 5 in. 
 
 wide,) 5,535 = . . . 12,575 " 
 
 Baltimore and Ohio, No. 4 of '38, believed to be that laid in the rubble bed near Bait., locust mud- 
 sills, 7 feet long and 6 in. wide, &c., = 6307 ft., 1172 transoms = 2800, = .... 9,107 " 
 
 Philadelphia and Reading, 1690 transom sills, 7 feet long 6x8 inches, = 7,887 " 
 
 Boston and Worcester, first plan reported to have failed, 1760 transoms 7 ft. long, 5 in. square, = 5,134 " 
 
 " " Millbury branch, on mud-sills, = ........ 11,196 " 
 
 Trellis Tracks. 
 
 In the proportions represented on Plates I. and II., = ........ 24,000 " 
 
 The same with the addition of mud-sills, as represented by the dotted lines, = .... 31,680 " 
 
 Trellis track, Plate III,— A 26,987 " 
 
 " " with mud-sills and string-pieces, = ..... 34,027 " 
 
 Diamond Trellis, in the proportions of Plate IV., = 22,451 " 
 
Opinion of Wm. Strickland, Esq., of Philadelphia, Architect and Engineer, on the New Construction of Railways 
 invented by Mr. James Herron, of Virginia, Civil Engineer. 
 
 Among the various methods now used for the superstructure of Railways in this country and in Europe, I know 
 of hone to compare with Mr. Herron's patent horizontal truss, or diagonal braced floor. It has the great advantages 
 of surface-bearing lateral connection, and longitudinal combination of strength and evenness of level. 
 
 "It is calculated to rest alike secure in all the various characters of soil, which so materially injures the cross level 
 of the present railway beds; whether from this circumstance or from the action of frost, Mr. Herron's invention 
 will be found to resist with the utmost degree of permanency all the vicissif;udes of the caved and washed embank- 
 ments which so frequently undermine the present mud-sills and cross-ties of the road-beds now in use. 
 
 "The simplicity of the combination of plank, the ease with which a carpenter, with no other tools than a saw 
 auger, and axe, ma.j frame and pin the laps of the bracings, goes still farther to recommend the adoption of Mr. 
 Herron's plan; and upon the score of cost of materials and workmanship, it will be found to require less timber, 
 with much less necessity of adjustment and labour in the notching and dubbing down than the common cross-sills 
 and'string-pieces, and when done, the greatest certainty is had in the accuracy of the work, requiring no alteration 
 or propping up to the grade, or cross-level, which is unavoidable upon the present method of laying down wooden 
 road -ways. 
 
 "The continuous bearing of the iron rail on the string-piece, which rests on the braces of the sub-framing, merits 
 general adoption as a means of producing great strength in a longitudinal direction. A span of 20 feet unsupported 
 maybe made strong enough to bear an ordinary locomotive and car. 
 
 "From the evidences of the several merits of Mr. Herron's plan I cheerfully recommend it for general use. 
 
 WILLIAM STRICKLAND, 
 
 Philadelphia, July 4th, 1840. Architect and Engineer." 
 
 REPORT OF THE FRANKLIN INSTITUTE. 
 
 COMMITTEE ON SCIENCE AND THE ARTS. 
 
 Report on a plan for constructing Railroads, invented by Mr. James Herron, of Maryland,* Civil Engineer. 
 
 The Committe on Science and the Arts, constituted by the Franklin Institute of the State of Pennsylvania for the 
 
 promotion of the Mechanic Arts, to whom was referred for examination a Plan for constructing Railroads, 
 
 invented by Mr. James Herron, of Maryland, Report: — 
 
 That they hav^ examined with much care and interest the drawings and model submitted by the inventor, and 
 have had the advantage of this gentleman's personal explanations. 
 
 It appears to us, that Mr. Herron has fully understood and appreciated the evils inseparably connected with those 
 plans of railway superstructure, so much in use here, as well as in Europe, in which the rails are supported upon 
 isolated blocks of stone, or sleepers of timber. In a climate like that of our northern and middle states, it is out 
 of the question for us to encounter the expenditure which would be necessary in order to obviate the influence 
 of frost. Even our best constructed roads are upheaved by its power, and where the supports of the rails are in 
 the least degree unequal in their character, either in consequence of different dimensions, different depths of foun- 
 dation, or different capacities in the sub-soil for imbibing water the result is a succession of irregular elevations, 
 depressions, and lateral displacements, which are destructive alike to rails, cars and engines, and productive of a 
 jarring and lurching motion, extremely disagreeable to passengers, and by no means free from danger. When the 
 rails are supported on cross-ties of timber, resting upon mud-sills of plank, (which is also a common form of super- 
 structure in America,) a much better form of road is obtained, though still far from perfect. The irregularities caused 
 by unequal settlements are less numerous and less sudden, but there is still nothing to prevent one mud-sill from 
 
 * Of Virginia, and late of N. Carolina. 
 
rising above another from the action of frost, where the ends of two come in contact. The sleepers may be, and 
 no doubt are, often elevated entirejy from the sills, and are also laterally displaced, giving a slightly serpentine 
 form to the rails, which enhances the flanch friction materially. But even if all these evils be avoided, there is 
 still a radical defect common both to this and the former plan, which is thus adverted to in a pamphlet written on 
 this subject by John Reynolds, Esq. in 1837. — "The chief obstacle to durability which pertains to the plan of sup- 
 ports at intervals, whether they be blocks or sleepers, is the alternation o{ flexible spaces and rigid points, which 
 (even if the supports maintain an exact level) produces in carriages moving rapidly over them, a series of concus- 
 sions, as the wheels successively impinge on the rigid or supported parts of the rails. Also, however small may be 
 the deflexion of the rail' between its points of support, those points become fulcra, on which it acts as a lever, to 
 raise or shake the supports next beyond them. When the supports have assumed irregular heights, (which is the 
 usual case) not only are the above evils greatly aggravated, but the rail acts on every depressed support as a spring- 
 beam, tending to jerk it up, or loosen its fastenings." 
 
 In addition to the above, we may add, what is perhaps sufficiently obvious, viz. that the weight of the iron rails 
 may be diminished proportionably with the distance between the points of support; and consequently the minimum 
 quantity of iron will be required when the bearings are continuous. 
 
 The difficulties above alluded to are not merely theoretical. Every engineer who has had charge of the repairs 
 of a finished rail-road will vouch for their practical existence, although much discrepancy of opinion exists con- 
 cerning the most appropriate remedies. 
 
 Mr. Herron's object appears to have been, to devise a plan in which all the parts forming the structure shall be 
 fully and adequately supported; while at the same time they shall be so connected together, that no portion will be 
 liable to independent displacement, either laterally or vertically. He has proposed several modifications, all of 
 which he thinks may be applicable in particular situations. Those which the committee consider decidedly prefer- 
 able, both from simplicity and efliiciency, are designated on his drawings as Nos. 1 and 4. In both of these, he 
 uses a continuous line of timbers supporting each rail throughout, which are joined at the points of contact by a 
 new scarph, peculiarly well adapted to the purpose, and strengthened materially by the manner in which the iron 
 rails (which maybe either j^ or bridge form) are attached to the longitudinal timbers. The whole is supported 
 and stiffened by a system of diagonal cross-planks, which have a triple use, as they afford a considerable breadth of 
 bearing, act both as ties and braces to prevent lateral displacement, and being loaded by ballast will counteract any 
 tendency which the bearers might otherwise have to become warped by changes of moisture and temperature. Sand 
 or gravel is to be rammed under the longitudinal timbers, so as to give them a firm and equable foundation through- 
 out, and the whole roadway, when finished, is to be filled with the same materials, as high as the base of the iron rails. 
 
 The v/rought iron chaiis proposed by Mr. Herron for thejoints of the rails, are of a form new to this Committee, 
 and if they can be manufactured by machinery, (which, if good iron be used, appears probable,) will possess advan- 
 tages over most of those now employed, as they will clasp the rails, without the intervention of wedges or screw- 
 bolts, with sufficient firmness to prevent any deviation at the joints, yet not so closely as to hinder them from 
 expanding longitudinally by changes in temperature. The rails are to be fastened permanently at their centres to 
 the longitudinal timbers, so as to cause the expansion to take place equally in both directions, thus reducing the 
 spaces necessary between the ends of the rails to a minimum. 
 
 The Committee do not think it necessary to enter into a minute description of the proposed plans, which could 
 not be made fully intelligible without accurate drawings; nor do they wish to be understood as fully approving of 
 all the parts, some of which appear to be attended with practical difficulties, though not of an insurmountable 
 character. The principles aimed at in the design, have their fullest sanction; and they would gladly witness an 
 experiment carried out upon a working scale, with timber kyanised, or otherwise prepared so as to be secure from 
 decay. 
 
 Of course there are many points of importance embodied in the general scheme, which have been previously sug- 
 gested by other engineers, who have at various times adopted similar contrivances. Other details are certainly 
 original, and the whole combination evinces a degree of judgment and ingenuity which we hope will not pass unre- 
 warded. By order of the Committee. 
 
 July 9, 1840. WILLIAM HAMILTON, Actuary. 
 
 Cyfl 
 

 
 
 
 
 ^'' 
 
 >J>3 
 
 7' 
 
 >» 
 
 !!^ 
 
 --ji 
 
 >? 
 
 ^f* 
 
 
 ■-» 
 
 ^) 
 
 '•> 
 
 •i? 
 
 r^ 
 
 
 
 p 
 
 ^' 
 
 >> 
 
 
 
 
 S5|: 
 
 
 
 
 
 
 
 "r»v3» Js 
 
 :^i - 
 
 > t5>=>r>; 
 
 
 
 *l^ 
 
 
 
 '^i> ' >• 
 
 
 
 o'^^arfc^lo^- 
 
 
 a6ae!»-> _ob'j 
 
 
 
 ~jSC 
 
 :-S>^^^ -^^^^^^ -^^S*:,*:.-^ 
 "^ "^=?5^^ >;-SQ^f>^ .^„^3ii:f~. 
 
 
 
 
 '0 r)-yg:>. 
 
 
 ■-»^x>s:>- '^ ;^3S^^>' 
 
 _^^ --':^:»- .:3» ■.:>.;> >-= 
 
 -^^ 
 
 
 
 
 
 
 
 
 
 ty>':-^CS^ 
 
 
 yj?' "i -:> zx Ttj ■, 
 
 1 - '--vs^^^^- ir*' 
 
 

 ■■■> .f^ ■ ,, 
 
 ^,^a?s^ .^-,=^> 
 
 i> '^A^yQ^ 
 
 
 
 1 ^^IJ".?^' ?5.-,,>; 
 
 
 
 
 
 ^'^^^' 
 
 
 ^-;. t:^/ 
 
 ^^^:^ ' 
 
 >t?':^S 
 
 
 
 Mi^S^- 
 
 
 
 
 
 ■5"v;5:>S:. J-3g».?'-