^y^ 
 
 %.>. 
 
 o., **A^ 
 
 ^ 
 
 w 
 
 IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 // 
 
 ^/ 
 
 
 
 :/ 
 
 
 "/a 
 
 <? 
 
 °% 
 
 /a 
 
 ^> 
 
 '■c*/ 
 
 
 I' 
 
 '^ 
 
 
 1.0 
 
 I.I 
 
 1.25 
 
 S '^ IIIIM 
 
 mm 
 
 iJA 1111.6 
 
 6" 
 
 i J lUlUglUJJlUt. 
 
 Sciences 
 Corporation 
 
 23 WEST MAIN STREET 
 
 WEBSTER, N.Y. 14580 
 
 (716) 872-4503 
 
 I 
 

 ;;. 
 
 i/.A 
 
 Si. 
 
 CIHM/ICMH 
 
 Microfiche 
 
 Series. 
 
 CIHM/ICMH 
 Collection de 
 microfiches. 
 
 Canadian Institute for Historical Microreproductions / Institut Canadian de microreproductions historiques 
 
 
Technical and Bibliographic Notes/Notes techniques et bibliographiques 
 
 The Institute has atiampted te obtain the best 
 original copy available for filming. Features of this 
 copy which may be bibliographically unique, 
 which may alter any of the images in the 
 reproduction, or which may significantly change 
 the usual method of filming, are checked below. 
 
 □ 
 
 Coloured covers/ 
 Couverture de couleur 
 
 □ Covers damaged/ 
 Couverture endommag^e 
 
 □ Covers restored and/or laminated/ 
 Couverture restaur^e et/ou peliiculie 
 
 □ Cover titib missing/ 
 Le titre de couverture manque 
 
 □ Coloured maps/ 
 Cartes g^ographiques en couleur 
 
 □ Coloured ink (i.e. other than blue or black)/ 
 Encre de couleur (i.a autm nna hiahu n.. ^^ 
 
 Encre de couleur (i.e autre que bleuts ou noire) 
 
 Coloured plates and/or illustrations/ 
 Planches et/ou illustrations en couleur 
 
 Bound with other material/ 
 Ralid avec d'autres documents 
 
 D 
 
 D 
 
 D 
 
 Tight binding may causa shadows or distortion 
 along interior margin/ 
 
 Lareliure serree peut causer de I'ombre ou de la 
 distorsion le long de la marge interieure 
 
 Blank leaves added during restoration may 
 appear within the text. Whenever possible, these 
 have been omitted from filming/ 
 II se peut que certaines pages blanches ajout«es 
 lors d'Lne restauration apparaissent dans le texte, 
 mais, lorsque cela 6tait possible, cas pages n'ont 
 pas 6t6 film^es. 
 
 Additional comments;/ 
 Commentaires suppl6mentaires: 
 
 L'Institut a microfilme le meilleur exemplaire 
 qui) lui a ete possible de se procurer. Les details 
 de cet exemplaire qui sont peut-dtre uniques du 
 point de vue bibliographique, qui peuvent modifier 
 une image reproduite, ou qui peuvent exiger une 
 modification dans la mithode normale de filmage 
 sont indiqu^s ci-dessous. 
 
 n 
 
 n 
 
 D 
 
 Coloured pages/ 
 Pages de cctuleur 
 
 Pages damaged/ 
 Pages endommagdas 
 
 Pages restored and/or laminated/ 
 Pages restaurees et/ou pelliculaes 
 
 Pages discoloured, stained or foxed/ 
 Pages ddcolorees, tachetees ou piquees 
 
 Pages detached/ 
 Pages d6tache«;s 
 
 j / I Showthrough/ 
 
 Transparence 
 
 □ Quality of print varies/ 
 Quality indgale de I'impression 
 
 □ Includes supplementary material/ 
 Comprend du materiel supplemen 
 
 pplementaire 
 
 I I Only edition available/ 
 
 n 
 
 Seule Edition disponible 
 
 Pages wholly or partially obscured by errata 
 slips, tissues, etc.. have been ref limed to 
 ensure the best possible image/ 
 Les pages totalement ou partiellement 
 obscurcies par un feuillet d'errata. une pelure, 
 etc.. cnt 6te fi!m6es d nouveau de facon a 
 obtenir la meilleure image possible. 
 
 This item is filmed at the reduction ratio checked below/ 
 Ce document est film* au taux de reduction indiqud ci-dessous. 
 10X 14X 18X 22X 
 
 26X 
 
 30X 
 
 12X 
 
 16X 
 
 20X 
 
 24X 
 
 28X 
 
 32X 
 
The copy filmed here has been reproduced thanks 
 to the generosity of: 
 
 Library of the Public 
 Archives of Canada 
 
 The images appearing here are the best quality 
 possible considering th^ condition and legibility 
 of the original copy and in keeping with the 
 filming contract specifications. 
 
 Original copies in printed paper covers are filmed 
 beginning with the front cover and ending on 
 the last page with a printed or illustrated impres- 
 sion, or the back cover when appropriate. All 
 other original copies are filmed beginning on the 
 first page with a printed or illustrated impres- 
 sion, and ending on the last page with a printed 
 or illustrated impression. 
 
 The last recorded frame on each microfiche 
 shall contain the symbol •^►(meaning "CON- 
 TINUED"), or the symbol V (meaning "END"), 
 whichever applies. 
 
 Maps, plates, charts, etc., may be filmed at 
 different reduction ratios. Those too large to be 
 entirely included in one exposure aru filmed 
 beginning in the upper left hand corner, left to 
 right and top to bottom, as many frames as 
 required. The following diagrams illustrate the 
 method: 
 
 L'exemplaire film6 fut reproduit grdce d la 
 g6n6rositd de: 
 
 La bibliothdque des Archives 
 publiques du Canada 
 
 Les images suivantes ont 6td reproduites avec le 
 plus grand soin, compte tenu de la condition et 
 de la nettetd de l'exemplaire film6, et en 
 conformity avec les conditions du contrat de 
 filmage. 
 
 Les exemplaires originaux dont Itc couverture en 
 papier est imprimis sont filmds en commenpant 
 par le premier plat et en terminant soit par la 
 dernidre page qui comporte une empreinte 
 d'impression ou d'illuntratinn, soit par le second 
 plat, selon le cas. Tous les autres exemplaires 
 originaux sont filmds en commenpant par la 
 premidre page qui comporte une empreinte 
 d'impression ou d'illustration et en terminant par 
 la dernidre page qui comporte une telle 
 empreinte. 
 
 Un des symboles suivants apparaitra sur la 
 dernidre image de cheque microfiche, selon le 
 cas: le symbole — ►signifie "A SUIVRE", le 
 symbols V signifie "FIN". 
 
 Les cartes, planches, tableaux, etc., peuvent dtre 
 filmds d des taux de reduction diffdrents. 
 Lorsque le document est trop grand pour dtre 
 reproduit en un seul clichd, il est film6 d partir 
 de Tangle sup6rieur gauche, de gauche d droite, 
 et de haut en bas, en prenant le nombre 
 d'images ndcessaire. Les diagrammes suivants 
 illustrent la mdthode. 
 
 1 
 
 2 
 
 3 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 I 
 
4 \f> ®* 
 pi«w— ■'••'possible =» 
 .^^a^seuss-"'^' 
 
 earner ^ ,, ^^ ^.^^^^ resnveil) 
 
 ADVANCE PROOF-(.'^'"'v><"" '•emioii). 
 Tliis [irncif is sent to you ii)r discussion only, and on the expr ss 
 undtr.'-taiidini.' lliiit it is not to ho tised for niiy other purposj what- 
 cvi'i'.— (-fc 'ec. 47 of the Conilituiion), 
 
 (iiiiiadiai; Si-'cictil ^l ti,hn\ (Jnjineers. 
 
 IN'COUI'UllATKD 1887. 
 
 TRANSACTIONS. 
 
 N.ll — Tlii- Society, us u lioily, does not hold Itself rcspoiisilile for tlio fr.cts 
 and o|]inions ftiiti'd In imy of its pn hi lent Ions. 
 
 SI'KCIAL TRACK \VO\lK FOIt ELECTRIC STREET RAIL- 
 'AYS, ESPECIALLY IIKPERIJING TO THE .MONTREAL 
 .\\n TORONTO SYSTEMS. 
 
 Bv !•;. A. Ston-e, M.v.li. A.lf. C.^n. Soc. C.E. 
 
 Special work is the general teriu applied to all track work not in- 
 cluiicd in ilic ordinary .straij;lit track ; its ecnistruction tor electric rail- 
 ways iia.s undergone _s;reat improvements during the la.st few year.*, and is 
 still iuiprovihg. The introduction of electric power for the purpose of 
 city passeiiijcr tr;iffie gave ri.se to the present substantially constructed 
 c.irs, which, witli their additional weight of motors, brought about radical 
 changes in the construction of the track. 
 
 Besides electricity as usul in the trolley system, other motive powers 
 liave been tried to t:ikc the place of the horse, such as gas and com- 
 pressed air nio'ors, cables, electric conduits and storage batteries ; hut up 
 to the present time, the trolley system lias demonstr.itod its practical 
 superiority over all others. 
 
 Tfie track which had answered all purposes for the old comparatively 
 lightly cimstructcd horse cars beeami; utt rly u.seless for the motor cars. 
 As the special work is subjected to the greatest wear, and coiise((ueiitly 
 requires the most frctjueut renewal, it changed (iirm completely. T!ie 
 old cast-iron curves, with their sho. ., lightly constructed switches and 
 po.ir joints, had to give way to the heavier steel construction, bearing a 
 greater resiinblanee tu that nf a steam railroad. 
 
 Special track work should be of good stibstantial construction, with 
 the greatest care paid to the designing of the parts which wear most 
 rapidly. It is most important that track, especially in the central part.s 
 of a city, should rcnuire renewal as .seldom as possible, for such renewals 
 are very expensive, apart from the actual cost of the new track work, as 
 traffic is interrupted, causing great inconvenience and sometimes loss of 
 bus.ness to the ])uhlic, and generally demoralising a whole route ofcat.s, 
 and sometimes the greater part of the entire .system. Special work 
 slioul i be made in such a manner as to cause the least possible obstruc- 
 tion to vehicles, no part rising above the level of the paving more than 
 is unavoidable; the necessary recesses, grooves, etc., should be as narrow 
 and .shallow as possible, to prevent wheels of vehicles from catching. 
 Flat surfaces .should have a rough top, to prevent hoises from slipping 
 upon them. All jjicecs should he tiiiisbed so as to facilitate the paving, 
 no long, unnecessary projections being left on bolts, etc. The curves 
 .-hould he of a.s great a radius as the width of the streets will allow. 
 Tlie sharper the curve the greater is the wear on the track, and wheels 
 of ears, the slower the late of mution. the more power n quind to drive 
 the cats, the more uneven llie nioiion and the greater liability to de- 
 railment. 
 
 The track may he laid on longitudinal striiigirs. on ero.ss ties, or 
 directly on coiicreie with tie bars coiniectiiig t]w rails. The old tracks 
 of strap rail were laid <in stringers, and the rail generally called stringer 
 rail. (Figs. 1 and 2.) The greater part of the new construction is laid 
 on ties, and in many respects is similar to steam track wiik. A combiua- 
 tion of these two methods, consisting of planks laid Umgitudiniilly od 
 
 1 
 
cro>s tics, ill ordor to f:ivc ii innie evon Mufacc, has boon trieJ, but the 
 results do not seom lo liiivi.' boon so Fatislactury as were expected. In 
 several streets in .Montreal, where permanent paviii'.; Ii:is been laid, tin- 
 rails liave bei II brid dirrctly uii concrete, and bimnd tn;;,'lb(T by flat tie 
 bars Willi tjiiradud ends and dmible nuts. Tliis, with tlui concrete be- 
 tween tlio ties, and pavins. makes a very snlid bed ; however, it docs 
 not seem to have so uiueb elasticity as traek hid on ties in macadam. 
 The rails used in Toroiiln and Montr.'al an^ " Girder " r .ils. Tho.se 
 first liid have a heiLdit of Ci jn. with a fl ,ii-e of U in., while those 
 laid later are (i^' in. high with a flanL'e off) in. ; the web of the rail is 
 not directly below the e. nire of the head as in tlu' " T.e " rail, but nearer 
 the -rau^'c line, while a flan.-way Ij in. wide at the top is provided for 
 by a prcjeelini; lip. Thes.' rails aveia-e 75 Ib.s. per yard. This type 
 of rail (Ki.'. •.) is used on all strai^'ht pieces and out.side rails on curves 
 in the spee.al wor!' ; the in>ide rails are m;ide of a .section very similar 
 to thi.s, the principal difterence bein.n; that the lip is much heavier, 
 beinuone inch in width at thi' top and risin- 5 lOiiis in. above the level 
 of the head of the rail : this ]irovide.-. an ifficieiit guard lor the ears in 
 running .ound a enive, the groove is ;] in, wider than in the ordinary 
 girder rail. This rail weighs 84 lbs. per yard. fFig. 4.) Another 
 section (Fi,g. 5) is, however, coming into use, and will no doubt largely 
 replace these sections fnr'peeial work : it i> the same as the guard rail 
 section, except that the groove is filled up with .solid metafto within 
 0-16ths in. ol the top of the head, thu.s providinga double bearing for the 
 wheels as both flanges and treads of wIm els rest on the metal, so that the 
 cars pass over all ]joinls without jolting, and the wear on the lea.stdur- 
 able parts of special work, viz., points, is greatly diminished. This 
 section gives a rail ol 89 lbs. to the yard. (Figs. 1 "to G.) The peculiar 
 .scciions of those rails, with their thin flanges and webs, and niueh thicker 
 heals, cause a variable amount of:. lu^diih'ss in the section, the head 
 having received the least amount of rolling proportionally and taking tlio 
 
 
 /-., J- 
 
 r.g e 
 
 longest time to cool is not so tough as the web and flanires. Tests on 
 pieces tak.Mi fiom the guard rail (I'ig. 4) have given" the follovvinii 
 results : — 
 
 Head :— Tensile stren-th— i;4,;!0ii lbs. per S(i. in. 
 
 Elastic limit— 75 \w.\ cent, of tensile .strength. 
 
 Klongation on 4 in — 3>.; per cent . ; irduction in aiva-'i vum- cent., 
 with an even and uniform whili-h gray fracture, moderately line -rained! 
 
 Web:— Tensile sireiiglh— ill, 250 llw. per .xi. in. 
 
 Klastie limit— 75 per cent, of tensile strcn-tb. 
 
 Elongation on 4 in,- -27 per cent. ; reduction in ana —20 percent., 
 with a fine grained light gray fracture. 
 
 The necessity for the incr.ase in the weight of ihe new rails over the 
 old is made apparent wh.u it is considered that the weight of a motor 
 car averagesabnut (i tons, wldlo the weigh! >.f the old iiorsc cars averaged 
 
only about 2 tons, and whereas Ixirsf cars run at tlin riito of obout 
 miles per hour, eleetrie ears f reijuentlv have i spocJ of 1 ;"> miles per hour. 
 'I'ee mil (5(! lbs.) is also uscl lately tor this work, but its use is 
 Kciieraliy coafined to macadamised roads in tin) .suburbs, as its height 
 is iKit suitable for pavini; purposes (unless raised on chairs), although 
 oiherwise (piite a.< efficient. (Fiy. (5.) The j;iriler rail beinj; >o hijrh 
 iid.iiils ol' block ()avini:. and by tlic lip on the inside provides a ^tood 
 odj;e ft r tlio pavers to work to, \vl '1st the narrow fjroovo ott'ersi si very 
 sliiilit liiiulrance to vehicles. 
 
 In lee rail sjiecial work, the inside rail on ''Mrves i> neneriilly guarded 
 by a Mcund r:iil liein;; bolted to it, the tv/o rails being held apart by 
 cast iron fillinj; pieces ; the space between these rails is afterwards 
 filled with eenient to within an inch from the top, so as to cause aS 
 little obstruction to traffic as possible, thefiuurd rail is slightly elevated 
 ab'ive the ruiiniiej rail. FreijU^'Otly rails are used in paved streets of 
 insufficient height to admit oi'a paving block between the ties and the 
 hc.id of the rail ; when this is the ease, the ditfireiice in height, has tobo 
 made np by the use of cliuirs ; ibis leads to rather eoniplieated Joiiit.s, 
 and re(iuires a longer time to lay than the methnil of direct spiking to 
 the ti('s. 
 
 .M.\IN UIVISIO.N.S OKSI'KCIAL W.)RK. 
 
 Special work may be diviiicl into foiii' classes considi.'red with respect 
 to its use and its position when in place, viz.: — inter.srctioiis, p;issiiig 
 sidings, e.rossDvers and turnouts, and niiseellaneous combinations, 
 
 1. I III ur sect ions. — By the term intersection is meant the special work 
 placed at the iiitersc(!tion of twn or more sticets, and may as.sume an 
 almost endless variety ol' lorms as regards number and direetio of 
 curves and tho aligniiiciil of the main tracks. The work must bo so 
 constructed as to guide the oars in whatever direetinn re(|uireil, with- 
 imt any other e.xtenial assistance ihaii the moving of the tongues in the 
 switches by the motor men ; the cars must ride as smoothly as possible, 
 i.e., there should be no j.jlting ; in places where a groove is to be 
 crossed that would cause the ear to run iiueveidy, the floor should be 
 raised so as to give a hearing on which the flau:;es ma_\ run. On double 
 track lines the distance between tracks is usually Iroin Hmr to five feet, 
 but in order that ears may p.iss one another on tlic curves, and not be 
 obliged to wait at the eniN, this distance is inerjased to about seven or 
 eight feet to provide ample elearanee, this CNtra width is obtained by 
 strikimrthe curves from differiiit centres. /.'., the curves are not eonceu- 
 tric. The practice in Montreal and Toronto has generally been to make 
 the inner and outer curves of the same radius when the apex angle has 
 been nearly 90^ ; but when the angle varies greatly IVom a right angle, 
 the outer curve lias generally been maile shar[);'r than the inner when 
 ruiiiiinL' round the obtuse angle. When the centre line ol'strc-'t 
 
 changes direction, or has a "jog" at the inforsection, necossitating a 
 plain or reverse curve on the tliroudi tracks, the coniplioatious increase 
 very rapidly. 
 
li, Pnuiiig SiiHiiij.:, — Tlit'MO HIT UHoil on sin^'li' tiiiek linns vvlicii 
 curs luii in both iliroclioris ; tln-v niiiy b" divi'^Icil iiitn iwu cliisjfs, viz. : 
 •liuiiiiiinl mil ihrowii-nvor >iilin''s. 
 
 Ill the rliaiiumd Milinix (Fi.i:. 10) tlic luvk divirj.'.'s like a V .it 
 eithiT imd, so that tlic (viitic lino butvvoon tlic irackM in tliu sidiip.;s i- ..u 
 line with till' cniiv liiir of tiio sin-lr traek ; this is liio r„nn imu.iiiy 
 ailoi.toil on .-inuli! tniek-^ rmiiiinu' llinmu'li iiiirrow .-tivols. If it is 
 desiivd tlmf oars -hull run ritlior to tliu rii;htoi- left at thuso poiiif.s, the 
 switches or till' sidin-s iim>l Ik- i.mvid.'d with movable ioii^'iirs ; but if 
 the i:us always run in the same diivetion, liny may he yuided in the 
 dii-wtinn rciiuiivd hy a movable tongue heM to llic proper .side by a 
 .•^prin-. .HI that a car iiieinj;' a .switch is alway,- -iiidcd to the san.e side, 
 and .1 ear trailiu- it eompressos the .spring, and pa-os on, liie toiij,'iie of 
 the >witch fdlliii_u- hack to its proper poMtion. (.S,-' Plate l''iu'. B.) This 
 
 ;j;iiidin- of til,,' ear in one direeti however, may ho provided for mueli 
 
 more .■•iniply by means ,.f a switch willumtany movable part, commonly 
 called a blind .switch. One .-ide of the switch i~ slrai!.'lil ami th.e other 
 curved, the front of the switch coincides ajipro.'cimately witli the en 1 of 
 the curve of the switch, whil-t the curve to the opposite .side bcuiiis 
 near the b ick of the swit(di, as shown in I-'ij;. IS. If the cars alwav- 
 
 run t,j the linht (a.i in Montreal and Toiont,,) the switch is made loft 
 ''"i", (.<., the p, c. of the cm-vc tnriiini; to the left is in front of the p.e. 
 of tliecur.c turning to the ri,t;ht hy the leii-th of the switch (appro.xi- 
 niately) ; thus a ear approaching the siding travels straight along on 
 tlie tangent past the point of the switch, and is then curved out ol its 
 path to the side hy the curve in th.' rail hehiiid, and when leaving the 
 Riding run.s over the curve of the swiieli ; this is the best airangomeii t 
 tor .'^uch sidings, as it is the simplest, most durable, and causes lea.st 
 delay to the cars. 
 
 In the thrown-over siding (Fig, 17) one track is continued strai-ht 
 lliroiigh, whilst the other is thrown over to one side of it ; this is Miit- 
 a lie tor siiigh.. track lines on a wide sirect, or in places where the track 
 'SOU otu; H,]^, of the sirect. If ears are to he run to ehhw si.le, 
 switches with movable longm-s an- neec-s.rv; but if the c:irs always 
 keep to the .same side. tl^. t.mgues must b' proveled with sprin-s, or 
 blind switches used ; with the latter the problem is not so simple'.is in 
 he diamond .siding, and ui order to .s.dve it the main track has a slight 
 reverse curve placed in it (i.xtcmiin.j from the lirst of the switch to a 
 si'ort distance inside the curve «ross ; by introduein- thii. the ..leneral 
 arrangement for the diamond .siding holds -oo. I. (.See Kiu. l!l.) The 
 radius h„. the curves of pa.ssing sidings in .Montre.il and ■foronlu is3(i0 
 Jeet to iiisidf gauge lino. 
 
dilli'il <'(iimiTliii;i ti';i(k'*, me used mi ilniiblo lliiplt liiiO'* lor t\w pur- 
 jiiiM' dl' traiiH('c'nin;_' curs from oiii! truck to the other, and consc- 
 (|Uuntly iiri' plai;c.i <•( the torminationa of rojjuliir tnwU'* ami at [)oint» 
 wliicli ;iiv ui:i(li5 tciiipHiMry tcruiinii to iiccomiiiodiiti! wpociul traffic. 
 
 Tiniiiiiit< (Ki^'. 112) lire u-^id wlioii ;i doulilu track runs into a 
 »iiiiilc> track, ilic centre liiii^ of tlie sin^li! track l)"in^ on line witli tliu 
 tciilru line iil'one of tlio trucks of tliu iloublc track line. 
 
 These cro-Hovors and turnouts, as well ag all special work, nhould 
 eli.injje tlic ilirecti m of the ear's rantiiin I'roni one line into another with 
 ill" Ica.-l amount of reslHtance po«ible ci nsi^tent with the date ;,,'ivpn ; 
 those in Montreal and Toronto have 75 feet railius curve and about 25 
 feet of tanijeiil, the latter varying with thi' distance between tracks; 
 tills irivcs a crossover of about tiO feet hetweun extreuio ends of switches. 
 (,'rossiivers and turnouts aic said to lie eitlnu' left or rliilit hand, aecord- 
 ing to the diieetion in which they curve from the ti nek, as seen from 
 
 ^ - r ' 
 
 
 -^.^ 
 
 
 ■ , ■ 
 
 1 
 
 T"* 
 
 
 —^^ 
 
 -«..-^- . 
 
 *~- 
 
 -,i_ 
 
 i 
 
 • 
 
 f-- 
 
 • 1 — 
 
 .rj-- 
 
 ■ — - 
 
 -i_ 
 
 X 
 
 !** 
 
 
 1- 
 
 i ■ 
 
 r.t 
 
 
 . •« 
 
 
 
 the switch when leokinv' towards the cros--. Fig. 20 <how.s ii right hand 
 crossover. If a crossover of either hand is suitable at a certain point 
 of the line, one of the snuiu hand as the .side to which the cars run 
 should be chosen, /.e,, ri!j;ht h;ind crossovers are prelL-rable for .systems 
 on wliieli the cars rini to the right and left hand, on those in which the 
 cars keep to the left : this is on aeeouaf. of the f.iei that (Mrs running 
 always to the right will trail all switches of riu'ht hand crossovers and 
 faci^ those of loft, so that they cannot possibly take the wron;^ track in 
 the tirst case, while they .jay be suddenly thrown out of their course 
 in the second, and aciiidents result. 
 
 In addition to |ieruiancnt crossovers il is ;dways iieccssiiry to have 
 tenip'rary oucs during construction, which are laid directly oi: top of 
 the paving wherever rcijuired. These aro so coustructod as to be easily 
 and nuieldy laid in plaeo and roulily moved from one part of the line 
 to another by a small gang ol men. 
 
 4. Miecvlluiit'oiix Comliiiiaftoit. — Besides tlie work already men- 
 tioned, there arc several kinds of diamonds made to till various 
 rctiuiremeuls. There are also sjiccial combinations for car houses, etc. 
 The simplest kinds of diamonds are tlio.se used when electric lines cross 
 electric lines, and only rci|uiro the running rails. When an electric 
 road crosses a steam road, the steam road track recjuires guard rails for 
 greater safety, and the electric line should also he guarded either by an 
 additional rail or plate. 
 
 81IH-DIVISI0N8. 
 
 Int<'r.scctions, cross-overs, etc., are compo.sed of several pieces, which 
 may be divided into the following subdivisions, viz. : — Tongue switches 
 (single and (iouble curve), blind switches, mates (single curve, double 
 curve and combination), curve crosses (single curve, double curve and 
 combination), diamonds (forelectiie and steam crossings), split switches, 
 stub switches and lengths of rail (curved and straight). (See Pages 18 
 and 19.) 
 
 1. Tongiii'-tiiritclies. — The ti:i::'i'' switch is perhaps the most 
 important piece in any combination of special work, as it is subjected 
 to greater and more fre(|uent shocks than any other piece, its duty 
 being to change the direction of tiie car's motion from one line to 
 
itnotl.i I-. Wlii'ii iimK' nf iiinliT mil, it incHn.triicti'd <irtlu' i;ui''.| rail 
 Hoi'tiim to ensure' the p rHot ji;uiilimci.' of tliii vviioiU. Whin in.i(li'ii(' 
 Ti'o mil, II -(mini is I'nrincd eitht i- by boltinii on iinnthrr picco of mil, 
 or by lairvlm; up ih" c.istins; "ii ih.' !<icK' to lorni thi' r.c|iiirc(l -imni. 
 Till' switch i;eiicru'ly riiii-ists ol' Co-.r niniii parts, vi*. : — tln' tiii.'oc :t 
 <M«lin^ ami two pioccHot' mil. Tin' hmgUK in inmlc of st".'!, iind ..honld 
 b<!of a Mibutaiitiiii gizc, havinji :i ci'o«i* Moction near the [miiit, pin)iir- 
 tioiiod to rodst violent shock-' ; nt iIk' shww r'liw the point must be 
 mthfi- sharp to niisiirc t\w ear " taking " it r^iujily ; if blimt. thi> e ir 
 ni.iy iMiiint the toiLoif, and drop ;^t;iiii. .•nisini; a s vi'ro jilt. If ilm 
 top o: the tonjiUi) rincs above tlio level of tlio ha 1 of ih.' rail, it is 
 >lopc.i it both ends so as ti allow tin' ri-if and fall of tin' cur to be 
 inipcrii'ptible. Tiie pin miit be Ho plae •d a- to niakn if iinpn-sibii' for 
 u wlii'ii to toiiidi tiie tonu'Ui! bohiml the pin, and «i) throw the switoli 
 bci'iire the bnek wheels luve rcaidied the pdint, If the t"iii;ii(MVeru 
 made Ml lonu' that the distance fnmi the centre of the pin to the toniinc 
 fiiiint ■.vere ;,'iciiter than the wheel li,|.i; iif the cars (ahoiit 7 f,.|.|) ilij;, 
 would be iinpo>:.ible ; this method, liowi'vcr, would nec's-itate ;, tri'i 
 expeii-!vc switch, and (ne ditticully is easily overennie by louiMlinL' the 
 liiiek nf the t ini,'ue and placing' the piu sutHciently fur back. The pin 
 sbiiuM also be placed so that the wheel- do not run over il, and .-i cause 
 it to heeuuic loosi;, ,and ,-iliould be sii fi^tened In tlie casliri,; that the 
 tomruc may easily be reimjved at any time. The top of thccastioL.' on 
 whieii the tongue .>|;des and the bottom of the toui;iie sheul 1 hr. truly 
 even, as if nul, dirt will ccllecl b.twecn the two, and after a slmi t lime 
 the tipnuue will tilt when a car runs over it, and may cause the too'iie 
 to t'i:usv to the uppo-iile -ide, m- th" back wh'ol may strik ■ the point, 
 either of whieli may be' sutlicicut to ilivow the car "rt' the track. .Simile 
 curve switches ure those curved only on unc siiK' ; double curve switches 
 lire curved on botli sid*s. 
 
 2. ///('/I'/ ^'(('iV('/icK. — 'Phe blaid -witch is u,-ed in p',. ice of the iiue'iie 
 switch when cais always run off the curve at that point and never 
 inter it. It closely resembles llie niiitc in licn'ral eunstructioii. In 
 order that the L'uidauce uf the car facing' the switch may not .ilteucilier 
 depend on the fact (bat the car will naliirally take the straiLthi track 
 
 in the direction in which it is viiig, ntln'r than turn into ihc curve 
 
 a ridLie i, left afmu; the floor on the str.:i-ht trick which actsasa "au-e 
 line. 10 make it practically impo.ssible for the car to enter the curve. 
 
 ;{. Afitf^t. — The mate is the piece opposite the switch, on which the 
 wheels of one side uf the car run wbili; the wheels on the other side 
 lire being pulled around by the switch : its sole use is to provide a 
 surfice tor the wheels t.i run upon, ami has nothin<,' to do with the 
 change in direction of the car's motion. It is made of two pieces of 
 rail, and sometimes there is a casting. One piece of rail extends over 
 the whole length, aud is straight if fir a siii'.de curve mite, and curved 
 if for a double curved mate : tlic other piece is shortt r and always 
 curved, the head terminating in a point this point should be so 
 designed that the gauge at the point is ipiiie slack. S'o that a wheel 
 facing the mate may not strike upon it. The widtli ul' the point >l,ijuld 
 not be loss llian |l-iiich, as if made .■,li,iipci' it will wear t,i this. In 
 girder rail tlie solid flnor .seeti'in m ikes the be>i ,, aie. as ii j.i'ovidcs a 
 wide floor for the wheels to roll upon, and the depth of the fl'iorbi'low the 
 head of the rail being le.s.s than tiie deplh of the fia .ge uf the wheel, it 
 quickly wears so as to pnivide a double be iriiig lor the wheids, m, that 
 the point is pas.sed without llie wheels iliopping heavily ujioii it. If the 
 
 luaie is nut made of ti; ' tlo ir seo.i but ■,( the or liii iry gov! t r.iil as 
 
 used on the slraiglit track, ur il' of Tee rail construction, a seel castio" 
 is nee ,s.siry to carry th; wheels over the poml fr.mi the long rail on to 
 the short one ; tins casting is more dfficient if carried up on the in.dde 
 to provide a guard ; for in case of the gauge being too slack, the ton.ruc 
 may have a tendency to Jerk the car off the track. i'lii, cisting iiiusr. 
 project considerably inside tile gang..' iiiij of tiie short r.'ii, llie pith of 
 the rear wheels on a truck not coinciding with that of the front ones 
 but lying about i-incli inside, as may be ckariy seen ou any worn mite' 
 
 6 
 
4. i'lntc Cruinf. — I'liiTP or'iSK is tlii^ imiii'' j;ivcii in tliin work to 
 tlit^ picci)'ci>iTO!«pontlin!{ to tlio t'rou in ntciim ruilrrml work ; it differ* 
 oonsiilcnibly rniin the tro;», however : "im, nt Iciist, of tlio rnils in a 
 curve erojis i- vnn lally piu'viil to n very sliiirp nirve, wliii't the I'm^ in 
 ,-trui-lit on inlle r lr:ii:k ; tlie Iro.' Iia< win:,' nilN, iiml !i wlie"! en^itin^ 
 n fro^' runs frmu one |iiccc of r;iil iioriiis tli" nlianni'l nn t" aiintluT rnil, 
 wliilst in the curvo immss n whool ironorally run» llm imtirulenjtth of the 
 ein4-< (in one pioco of rail, lln' olmnnei for llie flnn'^es bein^' Hliapel out 
 of till' li' ;ul of t!ie rail. .\(M'nnliii;; iis one or lioiji riiil- are onrve I, 
 the eross Ik wiiil to he a 'inj.'lo "f donhle curve erosn, 
 
 r». Piumoiiih, — Diani mils ire luii'le in various ways, iieconlinu to 
 th'! re((uireinont'< t-liey are to Si'ive. A siiuplo sing'p track ilinmiml 
 for llie erns-iiii'^ of two eli'etrie lints eonsists of two main parls, each 
 part heiuL; inailo of live I'ieoi's of rail, one Ion"; pi' w with lour short 
 pioriH buttinj; up iiuuinsl it, two on each siile ; the lonj» rail Ih usually 
 ninite to foriu part of the ti lek on iluiMtitet having the frroater auiounr 
 ol tiattie. Wlun 'in elect: • mad crosses a ste.iin roiiil, the ilitiinonil 
 is usually all nni'le of tee nil, ol' the sinne soetiou :is the rail 
 of the strain mail. If the rails of the sle.un r^ nl are not to bo cut, 
 tile ili.imo'iil is luade in three purts (Pago Ifl, Fig, E), t\ >ui-iile aud 
 one insiile the steam trnelj, the whole being so eonstructwl as to lift tho 
 tilreet ear before reaching the rails of the sle:iin track on to the fl itigo.'i 
 of the wheels, anil ri'.iming across on them In the other side, and then 
 dropping gruduiiUy lo the ordinary hvel again, .so that the only placo 
 where any jolt can occur to :i eir while crossing tiuch a dinmond is 
 when it cro-.sis the (dnmnel ol' tin' .-team truck rails, notwithstanding 
 the fiict that th'' raiLs of the steam tiack are not eut to the smallest 
 extent I" provide a passage fur the flanges of the street-car wheels. 
 
 II. SpUt Swltilu'i'. — Sjilit swllihes are us d t'la comparatively small 
 extent on this class id' work. They are more especially adapted to 
 snbiirliin traffic when Tc- rail is used, ritle^r ih m crowded thor- 
 onuhiares of cities. They arc especially suitable when cars alw.iys run 
 lo ibe s:inie side, when the switch may be maile to work automatically 
 by means ol' a spriiiL', and in this way they have been found very satis- 
 laelorv. 
 
 7. Sitili NiciVe/cA'.— Stub .switelies aio snitabh' for yar'l purpo.scH 
 ami sidings oidy occasionally tiscd ; they an- cheap, winch is always a 
 pnitit in their lavonr. 'Plie use of a stninl prohibits their use in city 
 tbori U!.;hfares. 
 
 .S Itfiiiitlix til liiiil --Kails for nil special work should be accu- 
 rately cut to tile rei|uire'l lengths, and e:iretnliy lient lo the pro/ier 
 temi'lat.' if for use on a curve, or accurately straightened if re((uirod 
 for siraiglit Hack. Il'|.art of a rail i.s l.o be .str.dght and the reuiaiiidor 
 curved, the rail mn>t not only agree with stiaiglit edge and template 
 for the rei|Uireil lengths, but it must be tested, to determine whether the 
 .straiLdil part is tangent to the curve, fir if not, the piece will not fit 
 correctly alien I'Ik-iiI in the work of which it forms pari. 
 
 TlIK DKTKIl.MINATUl.N OV NKrKSS.\RV Sl'Kri.M. WDItlv. 
 
 Having laid down th" routes of any street railway system nece-sary 
 lor the accomiiiodation of the pres'?it traffic and thatof thenear future, 
 the special work rei|uirod becomes apparent. It la u.ost important that 
 curves likely to be rci(uiied in a few ,vcars, but not necessary at tho 
 present, should be laid, if at all possible, duriiu construction, as the 
 tiddition of a single ciiive to an intersectiou in some oases necessitates 
 the reeon-truelion uf the greatiT p.irt oftlie whole intersection. 
 
 8IJHVKV.S. 
 
 A c. ireful survey must be made of the iiiter.seetion of streets requir- 
 ing special work, aud all tneasurcineuts of lines aiul angles taken which 
 are necessary to plot with the greatest accuracy the centre lines of the 
 proposed tr.icks together with the street aud curb lines. 
 
 7 
 
n.oTTINII. 
 
 TlieMi luciihiiti'iiiuiitM nf pliittcd to u Kiiitible duflu (h»j tO U'ui to 
 1 inch), iinij till' iiioit siiitiilili' nidii I'lir tin' rcijiiiri'il curvon ili'tcriiiiriod, 
 which aro ui-imlly Ircuii 10 to 7B t'l'd riidiu-. ('15 ui'd 5il It. uri' iikwI 
 coiiiinon in >^fl)ntrt■lll and Tiiroiito). 
 
 The uit«iupt is wiinitiiiii's luiidi' to oiiiic tlicto curvi'n ujt on -tcaui rail- 
 Mud work ; but whun it in rcuH^Mibcrt'd tliiit thi' l('u«th oC ninut of tho 
 curvi'K is iboiil 8lt tit., it will bo ■<im'Ii Imw limiicd llic -|i;u'o is in wliidi to 
 iiltrnijit iiiiyiliini; ot'tlM' kind; liowi'ViT, an iin|ir(ivinicnt uiiiy be: intro- 
 duc'ii by tuiilcin^' tliu switclms ;it the uiids cd' curviiiot' a ionni'r ruliug 
 than tb'' inain (lul't ul' (liu curvus, sucli as -.isiii;; 75 It. riidiuit switches 
 01' 46 it. radius cur?cn, lhi« cas"* the curves Cor 10 I'l, at cicii end and 
 nii'itH all inai'liwil iTquircnicntH, any liirlber ."ti'ps in tlM> dirrctinn 
 would sitni to kun towards '■ hair splittin':." 
 
 Itniiubt bt'i'i' bi' inriitioni'd that i-lthougli iho.so curves would ajijiuur 
 very shiirp to cnjiinocrs auru.stuun'd to ^U'arn railroad work, .v:'t ibcro 
 is a caio o» record ol'a 5(» ft. radius curvi' on a trestle bcjn'-' u«i'd on a 
 steam railway, and opi'iatcd sueee-xfidly, tlir -peed on it bciiii; Iroui S 
 to 10 miles per hour. (U.S. Militoy Ua.lway, Petersburg, \a.; see 
 Trans. .\iii. !?oc. C.K. 1878.) The Maidiattan Kb'vated Railway in 
 New York city lias ourv.s id'lM) I'ct radiu«. 
 
 There should be, if po.ssibl', soffieient spic b'twcen the in--idi' rail 
 of the curve and til.' euibsion;' lot a veliiele lo |ia>s a ear easily ; thi*, how- 
 ever, reqiiii'i's very wide strei s ; if this eannot be .lone, the rail should 
 be at about two leel from the curb sinni' ni ilio eorner. for ii at say 
 four feet, tliere w.mid not brMifficieni roitu I'nr . ;raiul vehicle to pas-, 
 but the attempt minlit be made , an accident etfue Th,. viidii of 
 tlie eurves should also b' .letrrmined with a view tosuffieielit r'lom for 
 the .switches : if this is not looked to, special sbi'rt switches may be re- 
 quired, wlieh is not desirable. The interseclin- jmints of the ijau^e 
 lines should also bi" eavefully observed, a> by the slif^ht altifatioii of a 
 radius, combination piecC'* of complie.ited eonstruetion atid (d'an iition- 
 duralilc character may olten be avoid"il. The radii haviiif,' bten fixed 
 the ,i,'auj;' lines alone may be laid tlown to a lar;;o scale (say l tWi to 
 1 inch), and the calculations proc-edcd with. 
 
 C.M.cUl.VTlO.NS. 
 
 The data on whieli the calcul iiions are based aio : — the irauuv dis- 
 tance between tracks, an^'le of inter-eetian, ladii of curves, and ome- 
 tiliies diftunces I etnei'ii apexes an I d''flectioti anirles. 
 
 First, the tan^'cnts and Idijtlis of all eurves are found ; m.vt. the 
 distances between the ends ol ihe lurvcs aro dctcrminud. 
 
 Ill the case of a double track liianelicff, with inner and outer curves of 
 the same radius and e(|Ual ci'titral di-taneis, this di.^tance, ,i (Flj, 25) 
 is given by distanc between /'.r.'.«, „ =: (,i:au),'e ■■ central di.s- 
 
 trnce) tan 
 
 centre aiiirle 
 
 II' the radii are equal, but the eeni-al distances on the two streets 
 are uneciual, the distances ri'i|uiieil may be louud as folbiws :— 
 
lH U " «nu}/u. (Ml'"' Fig 24) 
 " D mill i> ■= ('iMirnhii>tanci'i 
 " K =r:|iili;;lc df illtirM'Ctioii 
 Hinoo '.111) inil'i ol' iIk' insiilr uii<l iniisiilr curvfi urc I'ljual, tlif tin- 
 jfi'iiti (Cell' till' wuiii' imuli') nrc cumil. 
 .'. (li»t,im,'n 'i)fiwiuii I'J'.'i = (lisliiiic'o bctwtMii U|K!Xi's. 
 (both mfiisurc'il pur^illrl to f;iui;^(' linen) 
 .-, ,( — (O D] ) cici.i' a ((I I Ih ) rut a 
 h = ((/ I D-j ) coiti'o a > {II I />i ) cot a 
 '• = (^' )■ />! ) oo-u' a — (tf I /ti ) cot a 
 </ = (f» , Z),. ) cohl'c a — (. t' I />! ) rut a 
 Wliiii both tliij fuutrul JUtniutH mill niclii v:"y, iho 'listnnoua bo- 
 twi'cii I'.C.'i, iiiu toiiiiil hy aililiiii: umi >ul,iriutiiii; tliu I nu'ths of tbo 
 tatiuiiitH, nittk'iiK iillinviumo tor ihi' iipcx :ii)):lu i! ilifferini; very much 
 fioiii a rii;lit inij;!!', 
 
 Nixl, tliu miiiib I' (f iiiic.-i into wliicli to ilividu the interncctiou in 
 (K'tcimiiieJ, and thu piopor loii-tliM Inr ^witchiw ami iiui».u,s flxeii. 
 
 Thi' jioiiilH wlirri tin; imrvi'^ inti r.^iot tbu stiai'jht j-aujjo I'ivk aro 
 ui'xl fiiimiil ; ihirt may l.n ilo.i,. by liihci' ot tin- two Ibllowiiig iiiotbods : 
 Tiikitii,' Kii;. 25 with listaiiC'.-! a- mark -il. 
 
 1st Mi'tbiid. Coiisidiir ilic point i\, 
 //, = V (y/i"i 0')- — //,- 
 
 StII 111 
 
 , •. ((| = Kill 
 
 I Hi / 
 
 Siniilaiiy li'i- /;. //_■ = ^IJi: — {/f-; — D — 0)- 
 
 Jl. 
 
 mil (I- = 
 
 . • . a, — mil \ ., ) 
 
 anil Ml oil tor oilier poiiil>. 
 
 — ( '^' \ 
 
 2 d M.'ibud.— I'or .1, (VCN a,-/.', ^ G .. a, = c-rg 'V 7e;i~(77 
 
 //, - A'l sill (I I 
 For /?, «-■ = /T« ' ( :^-i-^') 
 
 //., = A' -ill ,^. 
 Similarly tor other poiiit>. 
 
 At a distiiucr s, the sproad »■ = 2 .s sin _- (.See Vvz. 26) which ij 
 
 r,. It 
 
the iliatancu bctWL'cu two points at a ilistanco ,s Croiu tlio iiitersiotimi 
 point, oiiu on the straight gauge line and the other on the taiii'i'iit to 
 tlie curve at tlie int^^rsijetiou puiut. 
 
 The straight kni^ths of the figure (h'ig. -5), i.r., the ilistance.s along 
 'he straight track between the points ^1, li, etc., arc found by means of 
 tlie lengtlis //,, 7/j, etc., and tlie distance between tlie ^V.'s. Tiie 
 are to any point from the I'.C. isgiviu by : - 
 
 arc = radius - *c.m. a. 
 So tliat (lie curved loiigtlis, i.e., iho distaiieos hetweon the points 
 JJ, B,- F, E. ''!., are found by tiiking the ditfexonees between the 
 arcs to tliese points, wliile the distances beyond ^1, iJ. etc., to tiie other 
 end of the curve are found by takinir the differences between Uw total 
 lengths of tile eiirvcs atnl tile ares to tliese points. 
 
 Th.' foliowiiiir tables have been calculated by means <if the preceding 
 formulae : — 
 
 Radius i)f inside ijauae — lo' I)" flndiiis of inside ganjre = 50' I)". 
 
 Oaujje == 4' H\". CVnlral'distanct --= ■!' 0" iGange^-= 4' S\. Cenlral disi, =4' 0", 
 
 Points PerpiMi- ■\'.'J:}^"} Arc Irom „ , 
 
 A A Fi 21.117 I 2.-)"(W 21.sr2 | 
 
 Ji I 2(i.(J(l7 ; M^ 15' 2.S.4I17 ' 
 
 \ ;i:!.»(i.s I 4;t"0(;' .•!7.:!!t« i 
 
 .'» l8..-)47 I 24-21' i ly.lKi ; 
 
 '•• ,28.105 84^26' ! 29.S70 
 
 10/.." 
 
 17 f 
 lUi' 
 14^- 
 
 
 Angl<> 111, 
 
 Cl-lltl 
 
 Itv lire 
 
 i-iitit' n ,. ( Spread 
 
 point. " "'■ 
 
 22.204 
 2X.!!)li 
 
 ;iri.s.s|) 
 
 I'J. 511(1 
 29.014 
 
 2:!" 57' I 22.S(!3 j !)[■;" 
 
 .i4-20' ; 39.995 14 A" 
 
 41" (III' i .^.1.144 I IG-i'." 
 
 2:!"tl.V i 2lt.l.'{7 , 9^ 
 
 :i2''4(i' :ii.29;i i;ttV 
 
 When the iuteisectioii h:is curves hranehing in lintli directions, as 
 shown by Fig. i;j, the points where the eurves intersect lis A', L, etc., 
 have to be found, in order to deterniine ttio different lengths the 
 problem tliii-; lieeomes '-to determine ilii! interseetion point of cwo 
 curve-. Iirancliiiig in opposite directior from parallel lines.'' This may 
 be solved by idtlier of tlu' two following methods, the secfind of which 
 is niucli llie more readily applitid. ( e" Vl^ 27.) 
 
 Lei yi'i = 
 
 •' a = 
 
 '• h = 
 
 " C -T 
 
 •' L'- 
 
 " L = 
 
 " /J 
 
 " 7' = 
 
 1st M,.il 
 
 t — 
 
 r.t IT 
 
 -. radius of curve with upper /'. C. 
 " lower /'. (J. 
 distance h. iwe. ii /', C.'.s measured parallel to gau;;e lines. 
 ' ■' centres '■ per|)cndieiilar '' '• 
 
 " " " '' in a straight line. 
 
 " nf iiiteis ■ction point from npjur /'. (\ m :i~ureil 
 p.ira. to uange lines. 
 
 .-inglc lielwe.'ii a line perpendiriilir t<i -;iiigo liri. s and lirn? 
 joining centres. 
 
 angle at upper ei litre hetween ra lius to intersection pnlnt 
 and line jcjinin^- centres. 
 
 angle at lower centre betwem r^icljus Im intersection |iiiiiit 
 and line joiniii;^ centres. 
 
 angle at centre subtended by ,irc between lower /', l\ mn] 
 inler.-cction ]ioiiit. 
 
 anule at centre 'aibtci. led hy ,irc h tween upper /'. ( ', urn! 
 intersection jioint. 
 
 lod ■ — .''."' I //'■'-- A','. 
 
 I'.ireiilar ineii-iip'. 
 
 10 
 
i 
 
 (x 1- of I- (h-yf = it,' 
 
 .-. x' + 2ax + <»» + 6'^ - 2 6\/fi/-»^ i- ffi' - •«' = A;,' 
 which becomes 
 
 4 x' (rt- + 6-) 1 4 ax (a- + W v A',' - /.',') = '■',' (26= - ff,' - 2,a' + 2 rt,') 
 + ;?,■' (2 a- + 2 A" - It;') - b'' (/,« I- 2 a') - a* 
 Corollary. When A', = lu = /,' 
 
 then xr i „.r = \ I /;-■ (4 /.« _ /,2 - 2 a') - a' | 
 
 These formuliu iire very liibourious to use in pruotico ; however, as 
 in the majority of c:ise.s /,', --• l>',. the oomllary is tiio more frequently 
 recjuii'ed. 
 
 Having found x, the .in,'ies /{and '/'arc ;;ivcn by 
 
 sin 11 - ,, - 
 
 sin Tr= 
 
 /.', 
 
 and tlio spread at a disumce s - 2 s sin ( .,- j 
 
 Tiiesc forniulaj apply also when tlw twn curves brancii off in tiie same 
 direction, with the exception tliat the spread is given by 
 
 spread = 2.v sin f ., ) (see Fipr. 28.) 
 
 2nd Method ; 
 
 eos [/ - 
 COS £ = — 
 
 tan -j- 
 
 (■ = /) see P 
 
 Ui 
 
 2clii 
 
 : n > -- - R l '' 
 
 2 .• lb 
 T-r U-0 
 
 B^ L 
 
 sjiread = 2.<. sin ( -,, j 
 Corollary. Wiien lU = /.', = R 
 
 then I' = fj 
 
 sec V ■=■ sec L = 
 
 c 
 
 spread = 2.'< .sin U 
 When two curves liranuh in the same direction (Fig. 28) the above 
 applies with tin; following e\eeptions : — 
 
 T - 180° - (7/-^) 
 
 (J - l{\ 
 — '> — ) 
 
 Having fixed these points, the straigiit lengths arc found Hs before 
 
 by means of the perpend i cellar beichts to the intorsoction points of the 
 
 sinf^le curve ero.sses and the disi moes lo ilie diumond by means of thu 
 
 tangents. The arcs to tlie intersection point.s ol tho double curve 
 
 erossesaro given by : — 
 
 U 
 
For arc to intersection point on curv(^ with upper F.C., 
 
 ari' = li\ cm. T 
 For lire to intersection point on curve witli lower PJl, 
 
 arc = It cm. fi, 
 so tiiat tlie (listaucus aionn tlic arcs between llie points are iriven by 
 taking' the differences between tlio ares. 
 
 in I''!';. 13 it may ho noted tiiat when the la.lii of all the curves are 
 equal, the ando for tlie points A, .V, t) and /' = intersection unslc 
 ~ !I0 = . 
 that for the points A', L, iV and /', /,', = /,'_. 
 
 " " " L, N't and /-". ,1. h aiid con<e(iuently and r are 
 
 the same. 
 
 that the angle ffor tlie point .V = the anirle /, (or the point O, and 
 vice versa. 
 that L y = L 0, X U == H.O r = iV /', and I'T = P s. 
 
 The followiuij; table has been calculated by the precedins; fbrnnilM 
 from tile following data : — (refer to Fig. 13) J>| = 4' 9"^ JX^ = 4 
 0". « = St) * 3H', gauge = 4' 8^", radius of inside gauge lino ol' all 
 curves = 45' 0". 
 
 Point* 
 (F,;:. 1;!). 
 
 Perpendicidar 
 
 from np|>t»r 
 
 l\V. (X). 
 
 .\iigle at centre 
 subtended by 
 are hranchiiig 
 
 Anj;le al eentre 
 
 ubt ended liy 
 are hram'hing 
 
 Spiead ut 
 2 IVet. 
 
 
 to left. 
 
 10 rijrlii. 
 
 
 
 A" 
 
 ,5.:!l 
 
 21^ 2V 
 
 (W m' 
 
 
 125" 
 
 /. 
 
 10.;!;m 
 
 20° VJ 
 
 13-' 21' 
 
 
 13;" 
 
 .V 
 
 i:!.in4 
 
 15° 17' 
 
 19'^ 50' 
 
 
 14.'," 
 
 N 
 
 ic.sr.i 
 
 2»^ 19' 
 
 19" 49' 
 
 
 19|f 
 
 
 
 17.162 
 
 2«" 43' 
 
 22° 25' 
 
 
 19 if 
 
 P 
 
 22. ins 
 
 ;:!" 23' 
 
 2(;° 29' 
 
 
 231 ;V' 
 
 Note :— 2 (90"— 86° 33' :i = (;» 51' 
 
 = diftercnce between left and right angles nf fj and /■" 
 = " " " of A'^ and right angles uf ^^ 
 
 " " right ijf A'' and left of O 
 
 To determine the Af '. of a branch -off curve IVoui a curve main track 
 
 f,g >• 
 
 Ijet a = deflection angle of main track tangents 
 
 Let /3 = angle between one of 1 hese tangents and tairgent to 
 
 branch-off curve. 
 Let (^ = angle between line joiiung centres and perpendicular from 
 
 centre of oiain track curve to tangent of bramdi off curve. 
 
 Let (I = distance between apexes. 
 
 Let yf I = ra<liuH of main track curve. 
 
 Leti^'. ---■ " " branch-off .. 
 it is required to determine the point /'. 
 Taking x + .1/ as shown by Fig. 2!) ; 
 
 ;e = 1/ ll[ Ian . — // 
 
 12 
 
= (H- Jt\ tan .^— n, eot /3 
 
 = 'I + Ji\ ( tan [[ — cot /3 j 
 
 J a '■ sin l3 — 111 
 
 and cos f/ 5s 
 
 -.[11 + li\ tan '^ I sin /i - /ili cos /8 - Hi 
 
 1{\ - 111 when curves bnuicli in tlic same direction as in Fig. 29. 
 lt\ + Ri " " " opposite directions. 
 
 This c{etoriiiines the point I' with respect to either P.C. 
 
 To determine the intersecting points of the gnuge lines when the 
 main track curve lies wholly betwec'n the P.O. of the branch-off curve 
 and the nearest interseetiiiii points. 
 
 1' is the point to 'je dctoruiiiied (Fig. liO). taking lengths as marked. 
 ..v" + ,/ ^ ]{" 
 y =z I) - (./■ — (() tan a 
 
 .■ . X' i- \Ij — {x — a) tan a\ ' = IV 
 which becomes 
 
 j:' nee 'a — 2 .»■ fun a {h -1- a fan a) — li^ — A' — a tan a (2 & + 
 a tan a) 
 
 when the main track curves in the opposite direction to that of 
 the branch-off, this equation becomes 
 
 .(.■^ sec ^ o -*- 2 ./■ tan a (1/ — a tan a) = R' — Ir + a tan a (2 h — 
 a tan n) 
 
 = sin ■—- for both cases. 
 II 
 
 (fi ■, 
 — — J when main track and branch-off 
 
 curve in same direction 
 
 or spread = 2 x fin ( — ., ) when main track and branch oti' 
 
 curve in opposite directions. 
 
 If the distance (h) from the P.C. of a curve is known, the deflection 
 (d) i« the curve at that point is given by 
 
 ,/ = ,■ rcfi ( ■•••'■"-■ '-)(.Sec Fia. 31) 
 
 r'A 
 
 
 In order to iii:ike tenipl.uti'R tn whi.-!. t]u- vn'h sre bolit, ealculations 
 are necessary for flat curves (over UOft. ); but those of a shorter riulius 
 may be trammolied out. To calculate these templates, the defleetions 
 
 13 
 
lit cvory 3 inches from zero up tn liiilf (lie Iciifith of tlie ro(|iiii'ed tem- 
 plate are cHlculalcJ by oiii' of tin' nbovo formula!. Those iloflections 
 are hiid otl' on a board, a curve is drawn tliroui;li the points so fouml, 
 and the boiird is then eut to tiie curve. Of course tlio trammellin}; 
 process is preferable wiienever practicable. 
 
 Calculatioihijur Ci'iit--son:r-<. —Takiwj: lerijitlis a.s shown by Fijr. 2(1. 
 
 2 It rir/i a + lawjenf si a a — I) G 
 
 First, a length may be fixed upon approximately as desirable for a 
 tangent ; witli this lengtli, .solve for a (mcst easily done by trial), hav. 
 ing fund a approximately, assume an oven value for it (say to near- 
 est 10 minutes) for simplieily, and with this value solve tlio Ciiuation 
 again for the length ol' t.mgent, deuriiiining it exactly, which will he 
 very close to the desired length (practically the same). 
 
 The distiinces from ;'entre P.C. to interseeting point of inside gauge 
 is given by 
 
 .c — D co.'.vc a -- 0' — .', ) I 'in " 
 
 Tlif total length betwctn "Xtreuie ei:d I'.C.'s iagiven by 
 
 // = 2 Jt xiii a -r tani/fiif (vw a 
 
 The distance from end P.C, to nearest intersecting point measured 
 alonu main track is given liv 
 
 a 
 
 Z — (Ji .^ 1 silt a '■- ■£ c(i-i a 
 
 1 S.M r,) 
 
 iJy making tangent = 0, tli,. conditions for a rcver.se curve are 
 
 When 
 
 and V = 2 /i' •-■('// a 
 
 a crossover is recjuired f.ir a width between track.i, Di , the 
 only charge necessary in a crossover designed for a width D 's in the 
 length of the tangent which is dmnged by a length = (/>, ~ /;) 
 eoscc a. 
 
 To determine a reverse curve (short tangent between curves) between 
 two tangents not [larallel, at an intersection. 
 
 A A- and li.ll. are the tw<j tangents not parallel, representing the 
 centre lines of a street with a deflection at the intersection of another 
 street, the centre lino of which is reprosi I by (J.O. 
 
 Take distances as shown in Fin. '"52. 
 
 Fix upon a point which will be convenient to Ibrni one end of the 
 curve, and let its distance iiom an apex be /). 
 
 Then, ll\ rm:-: ,- laniit'nt sin -■■ A'j reis = n .tin a — b .iin 
 {a — /3) ■■' l{< i-ers (a-ld), as in the ordinary crossover calculations, 
 fix by trial and then s<ilve for the tangent, 
 
 tangent = . ] n xina — '■ sin (a — /9) ■ A'... ivrs (a — /iJ) 
 sin V ( 
 
 -vers 0(11, -,-n.,) \ 
 
 Having determined upon the angle ^, nnd limml tie- t iiiji'ni the 
 othtr Icngtli.s arc cisdy IouihI, 
 
 n 
 
CalctiUifmi" for IHunmnd Sidiwi. — Consider end A, Fig. 16. 
 
 D + a 
 
 vers a = 
 
 ^R 
 
 total ienj^th between extreme P.O." = 2 li ,iiii a 
 
 cof anjile at centre subtended by arc from riirht hand P. C. to interseo. 
 
 tionpoint=^'zii^'^-^)=^o. S 
 
 R-i a 
 
 nnsile of curve cross =2/8 
 
 distance from right hand P. 0. + , intersection point = ( R — J O) sin /8. 
 Tiiow! calculations apjily when the curves Ijegiu at the same point to 
 brancii to either aide as in Fig. 16 ; but when the curves begin at differ- 
 ent points (for blind switches) as iu Pig, 18, the intersecting point does 
 nut lie on the centre line, and may be found as follows : — (Fig. .33.) 
 
 tun = 
 
 _Ri 
 
 '■OS (j) — 
 a =r no" ^_ 7- 
 
 Ri —R2 —a 
 
 _ Ri- +(hs,;r0y'—Rl'^ 
 2 /,-.. f, ser e 
 
 ■^ Hb. er,ey-Bs'^ 
 2'Ri h sec e 
 
 iw\ /3 ^dO" -0-(f} 
 
 •'■ = ''\ sii, 13 and spread = 2 .v. .s/,i(lJ_'^) 
 
 (•'ih-K/.ition/or /In-on-H-orur n'dln:/ icith hUnd sw^YfAfl.s-.— The calcu- 
 lations are generally similar to those already described for crossover and 
 diamond sidings, except for the curves in the main track ; these are 
 solv"d as I'oUows: — (See Fig. 19, end A) 
 
 « = (li + h <') vi:i;i a i width of switch at back 
 
 rot:il I'eiitre an;_'le tor curve adjoining' switch = a .- /:?. 
 
 WORKINO IIUAWINHS, 
 
 Jlaviiiu eompleteil the calculations tor an intersection, I he detail 
 drawings for oacli piece are made, and sent to the shop, tonother with 
 a print slmwing the whole iuter.sectioQ witli ilio distinguishing marks of 
 all pieces and lengths of the ecmnecling rails. A drawing is also made 
 for as.sembling tlie work in tbi' street, showing all necessary mciisure- 
 meuts for laying out tlie work together with tlie p)»ition and marks of 
 the Various pieces. 
 
 SIIOl' WOKK. 
 
 A bill of the rails recjuireil and the necessary new prints and refer- 
 ences to olil ones having been obtaineil from the Drawing Office the 
 manufacture of the work may be proceeded with. The bill of rails 
 required (made out so as to give a minimum amount of scrap) is "iven 
 into the hands of the man in charge of the rail saw, who proceeds to 
 cut np the rails into the rei(uired lengths, marking the lenuth of each 
 and \vheiberre((uired strai'iht or curved upon the web. The rails next 
 with few exei'plious, go to the rail bender, to be either curved to the re- 
 quired radius, or straightened : they next proceed to the "'marker off " 
 
 15 
 
 I 
 
wlii> carefully iimrks the necessary lines for all niiichine work ri'C|uireil 
 to be done upon them, he also Mtumps the rails on the end with their 
 ili.^tinguishint; marks ; the rails afterwards pass on tothe machines (uiill- 
 iiij? machines, slottors, shapers, planers, etc.) suited to the work rci|niie(l ; 
 thty then <^o to the fittin;; shop to be assembled :iceordiiiir to the druw- 
 iug>. 
 
 In a tOD'^ue switch the long rail has to be properly curved, und 
 tlottcci or bent fur tho tongue to fall into place. The ton,t;ue is loade 
 of haniiuered steel, and the turned pin is shrunk in ; this is dropped into 
 place, and all ineiisurenients clieeked before beiiii: considered ivuily for 
 the iraek. 
 
 In the blinil switch an<l male, one rail is planed so as to leave a ionj; 
 notch on one side, while the other rail is planed to a point which tits into 
 the notch ; tlie iwo .oe strongly bolted or rivettej together and some- 
 times finished on a planer. 
 
 The curve irnsses have usually two pieces of rail, one of which has 
 the upper part so shaped at the crossing point as to allow the 
 second one to drop down on the first, and fit aecurately into the place 
 allowed for it : wliil.' llio seennd Iims the lower p;irt shaped so as to 
 nllow the lirsl rail to p;iss through, the two rails jointing neatly inlci one 
 another. Great care is neces.saiy in the lining to have the angles of 
 intersection o.\-;ictly as required ; in order tn obtiiin the correct angle, 
 the drawing shows the spn^ad, ,r, !it a lixed distance, together with the 
 deflection?. iZi ami (/,_> of the curves at th.it point ; so that this distaneo is 
 me:i>ured along the rails from the interseciinn jioiiit iind the defleetinn< 
 maiked from the gauge line, the spread is tlien nieiiMir.d hitweeii tlu^ 
 point-* ,^0 inirkcd. (See Fig. -li.) 
 
 ClIKi KINO. 
 
 When an intersction has been mid.', it is sum 'times a Ivisable to 
 have it asseiublc'l as a final check before .'^hipping ; for this purpose a 
 large jiiece of ground, as hn-el iis pos-^ihh', is rjcjuired. and mnel: more 
 than is actually oceupied by the work when in | laee shonid he available ; 
 the tangents of the intersection should bo laid out, and a sufficient num- 
 ber of points fixed to aecurately cheek the end of each curve. Having 
 laid out the ground, the pieces are as^euibled, and any errors observed 
 m;iy lie corrected; this last step ensures the work being ab.sohiteiy correct, 
 and i-* the licst cheek nn ilie work that can be aihipted. 
 
 .VSSKMm.I.VO IN TIIK TR.\f'IC. 
 
 In laying an intersection, it makes a great (hal ofdiU'erence whetiicr 
 the whole space recjuired is graded at once and all traffic sioppeii, or if 
 only part of the intersection is graded, Icavini;' part undi.sturlied so as noj 
 to interrupt tratlic. When the work has to be performed in th.' latter 
 way, Lireateare is neee.ssary in placiu'.; th' work, so that the reimiining 
 part when laid may lit up to and line in accurately with the first ])art. 
 If it is necessary to lay out a curve, it is gener.illy most easily ]ier- 
 form'd by tangent and chord deflections or by ordinatcs tVom a chord- 
 In grading a corner when an important intersection is to be laid, e.ire 
 shonid be cxereiscd in ixcavating to the correct depth and having the 
 grading duTie evenly, for if the track has to be lilted say si.\ inches after 
 being laid, it m^ans very nineh nnu-e than the same lift 'in ordinary 
 track, as the wei'.'.ht of rail is .sometimes enormous as compared with 
 the' extent of ground it covers ; aiso, if the work ha?!)ecn earele.-sly dmie, 
 and presents a very uneven beil, mueli more time is necessary tn eouple 
 \ip the joints than wiinld have been rei(uired had the grading bem pro- 
 perly performed. TIk^ spacing of the lies for this work should receive 
 more attention than is sometimes given to it, as it is a very important 
 matter. The ties should bo the very best available, and space I ..'ore 
 eloseiy than those on the straight track. 
 
 The center lines ol tracks for both streets are accurately li.\ed. and if 
 there is no diamond, the t:mls of the curves must be found ; otherwise, 
 this is not esaenl'al. If there is a diamond in the inlio'sec^tion, lliisis 
 laid tirst, bolted up and lined .iccurately. The other pieees havinii been 
 
 k; 
 
jjcattcred about in lUv'w approximiito positions are next druwn to place 
 imil bolted tOKcllicr. Tlio rails am tlicn securely spiked to gauf^e, and 
 lifted (if necessary) to^irade, when tlie intersection may be paved and 
 so eoinnlcted. If there is no (li;iuiond to lay, an end of a curve may 
 be taken .IS the siartin^i iioint. To lay the intersection so as to have 
 the thronjrh straifrht tracks in perfect aii<j;nment requires j^rcat care, ns 
 the joints are usually vuiy close toircther. 
 
 An idi'n of the iiinouiit ol rail that may be used is a sinsrlc intersec- 
 tion, and the eon>e(|iient auionnt of labi ur rei|uired lo make one, m»y 
 be I'orined from the Ci.llowin'.' li'^tires, for one laid at the intersecti'/U of 
 St. Lawrnneo .Main and St. Catherine streets, Montre;il (same as Fiu;. 
 15). It is built of 75 lbs. iuid 84 lbs. uirder rail (Fij;--. 3 md 4). It 
 contains 2,l"t(t liet of rail, and has a total wei-lit of .ihout 211 tons. 
 There are HC built up pieces (switches, mates and curve crosses), and 
 78 len<,'lhs of coniuetinf; rails, makin;;- a total of 1C4 pieces in tlio com- 
 plete inti'rsection. The extreme len;_'th between ends and opposite 
 switches is about 110 feet. TIk^ railiu- of the inside i,'auge lines of all 
 the curves is 45 feet, and llie distance between tracks varies from 4 ft. to 
 8 ft. G in. This intcr.-ection, as well as all others in Montreal and 
 Toronto, was made by the Canada Switch Manut'acturin^' Co., Lini., 
 ol' Montreal. 
 
 Sneli work, when properly conntruoted and laid, represents a larj^c 
 amount of eipital, anil deserves niucii more attention and care than the 
 old cast iron work ; but, unl'iriunately, it seems sometimes to bo treated 
 no better. The curves at intersections are necessarily very sharp, and 
 inord.'i' to diminish the anemnt 'if power reiiuired and the wear on the 
 rails (as well as on iires), ilicy rei|uirc oiling at least once a day for 
 he.ivy traffic, while the rate at which eais run over sp-oial work should 
 be strictly regulated to a low speed. The groove of the rail and the 
 toULine -witches require to be constai.tly cleared of the dirt which in- 
 evitably eoUeets, and ii' not reinoved cause.- '.ifeat inconvenience. The life 
 of sneh wiirk may be appreciably prolonged by such attention, and when 
 one cunsiilers tlie cost oi' renewal and the consequent interference to 
 traffic while doiiiu'>o, it will be readily seen- that it pays in theend. 
 
 17 
 

 Eai ^gjrrr" ' ~b igS5i::Sr::;;;:.v ■}^L&r::z:z. 
 
 •««'<'«•'.*#,' tawttt 
 
 J 
 
 
 \ii: 
 
 
 
 ^3< 
 
 IcQi ^ lio L_,_,..°„^: } 
 
 ** i»«£./^^,»^/ #•.!',•« ^/r r**'^ 
 
 ^■(f,.« S-lf 
 
 If" fl"! • 
 
 _iiL 
 
 
 P-C 
 
 ^1 
 
 II 
 
 18 
 
t,»m»m0 0mm • 
 
 JLlJ 
 
 mr 
 
 .a^y?a^.^jm^.^ 
 
 '■ * £^£jf^d. ffo^fr 
 
 a 
 
 • • « 
 
 L 
 
 ^^»D^ 
 
 19 
 
 E