IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 7 
 
 A 
 
 Ml 
 5> 
 
 
 /. 
 
 
 1.0 
 
 1.1 
 
 11.25 
 
 ■tt|21 US 
 W Ki 12.2 
 
 1^ Hi 
 
 L' U£ 12.0 
 ■■■11 
 
 U 11.6 
 
 <7> 
 
 <^ 
 
 /. 
 
 
 V 
 
 Photographic 
 
 Sciences 
 
 Corporation 
 
 33 WIST MAIN STRUT 
 
 WiBSTiR,N.Y. 14SS0 
 
 (716) 87ll-4503 
 
 « 
 
 »'^ 
 
 '^*^- 
 
 ^>^ 
 
 '4«^ 
 
CIHM/ICMH 
 
 Microfiche 
 
 Series. 
 
 CIHIVi/ICMH 
 Collection de 
 microfiches. 
 
 Canadian Institute for Historical IVIicroreproductions / Institut Canadian de microreproductions historiquas 
 
 II 
 
Technical and Bibliographic Notaa/Notaa tochniquaa at bibliographiquaa 
 
 Tha Instituta haa attamptad to obtain tha baat 
 original copy availabia for filming. Faaturaa of thia 
 copy which may ba bibliographieally uniqua. 
 which may altar any of tha imagaa in tha 
 raproduction, or which may aignificantiy chango 
 tha uauai mathod of filming, ara choelcad balow. 
 
 □ Colourad covara/ 
 Couvortura da coulour 
 
 [~n Covara damagad/ 
 
 D 
 
 D 
 
 D 
 D 
 
 D 
 
 D 
 
 Couvartura andommagia 
 
 Covara raatorad and/or laminatad/ 
 Couvartura raataurto at/ou pallicuiia 
 
 r~n Covar titia miaaing/ 
 
 La titra da couvartura manqua 
 
 r~n Colourad mapa/ 
 
 Cartaa gtographiquaa mt couiaur 
 
 Colourad inic (i.a. othar than blua or blacic)/ 
 Encra da couiaur (i.a. autra qua blaua ou noira) 
 
 r~~| Colourad piataa and/or illuatrationa/ 
 
 Planchaa at/ou illuatrationa an couiaur 
 
 Bound with othar matarial/ 
 RaliA avac d'autrea doeumai^iti 
 
 Tight binding may cauaa ahadowa or diatortion 
 along intarior margin/ 
 
 La ro liura —rr^ paut cauaar ua I'ombra ou da la 
 diatoraion la tong da la marga intAriaura 
 
 Blanic laavaa addad during raatoration may 
 appaar within tha taxt. Whanavar poaaibla. thaaa 
 hava baan omittad from filming/ 
 11 aa paut qua cartainaa pagaa bianchaa ajoutiaa 
 lora d'una raatauration apparaiaaant dana la taxta. 
 maia. loraqua cala 4tait poaaibla, caa pagaa n'ont 
 paa «ti film^aa. 
 
 Additional commanta:/ 
 Commantairaa tupplAmantairaa; 
 
 L'Inatitut a mierofiimi la maiilaur axampiaira 
 qy'tl iui a At* poaaibla da sa procurar. Las details 
 da cat axampiaira qui sont po'Jt-Atra uniquaa du 
 point da vua bibliographiqua, qui pauvant modifiar 
 una imaga raproduita. ou qui pauvant axigar una 
 modification dana la mAthoda normala da fiima^ > 
 sont indiquAa d-daaaoua. 
 
 □ Colourad pagaa/ 
 Pagaa da couiaur 
 
 Pagaa damagad/ 
 
 □Pagaa 
 Pagaa andommagiaa 
 
 Pagaa raatorad and/oi 
 
 Pagaa raatauriaa at/ou pallicuitaa 
 
 Pagaa diacoiourad. stainad or foxai 
 Pagaa dAcoioriaa, tachatiaa ou piques 
 
 Pagaa datachad/ 
 Pagaa dAtachias 
 
 Showthrough/ 
 Tranaparanca 
 
 Quality of prir 
 
 QualitA in^gala da I'impraaaion 
 
 Includaa supplamantary matarii 
 Comprand du matirial suppMmantaira 
 
 Only adition availabia/ 
 Sauia Mition diapon;b!a 
 
 n~1 Pagaa raatorad and/or laminatad/ 
 
 FT] Pagaa diacoiourad. stainad or foxad/ 
 
 rrX Pagaa datachad/ 
 
 r^ Showthrough/ 
 
 r~| Quality of print variaa/ 
 
 F~| Includaa supplamantary matarial/ 
 
 I — I Only adition availabia/ 
 
 n 
 
 Pagaa wholly or partially obscured by errata 
 slips, tissues, etc., have been refiimed to 
 enaurc the beat possible image/ 
 Lee pagee totalement ou partiellement 
 obacurciaa par un feuillet d'crrata. une pelure. 
 etc., ont Ati fiim^es i nouveau da fapon A 
 obtanir la meilleure imaga possible. 
 
 Thia itam is riimad at tha reduction ratio chaci«ed below/ 
 
 Ce document eat film* au taux da reduction indiquA ci-daaaoua. 
 
 10X 14X 18X 22X 
 
 28X 
 
 30X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 
 
 
 12X 
 
 1«X 
 
 20X 
 
 a4x 
 
 2IX 
 
 n 
 
 32X 
 
Th« copy filmMl hm has b««n raproducod thanks 
 to tha ganaro»ity of: 
 
 Library of tha Public 
 Archivaa of Canada 
 
 L'axamplaira film* f ut raprodult grica A la 
 gin^rosKA da: 
 
 La bibiiothAqua das Archivas 
 publiquas du Cdiiada 
 
 Tha imagca appaaring hara ara tha bast quality 
 possibSa conaidaring tha condition and iagibility 
 of tha original copy and in Icaaping with tha 
 filming contract spaeif Ications. 
 
 Laa imagaa suh^antaa ont 4tA raproduitas avec la 
 plua grand soin, compta tanu da la condition at 
 da la nc'ttatA da I'axampiaira film*, at an 
 conformity avac laa conditions du contrat da 
 filmi 
 
 Original coplaa in printad papar covars era filmad 
 baginning with tha front covar and anding on 
 tha last paga with a printad or illuatratad impraa- 
 sion, or tha back covar whan approprlata. All 
 othar original copies ara filmad baginning on tha 
 first paga with a printad or illuatratad impraa- 
 sion, and anding on tha last paga with a printad 
 or illuatratad imprassion. 
 
 Tha last racordad frama on aach microflcha 
 shall contain tha symbol — ^> (moaning "CON- 
 TINUED"), or tha symbol ▼ (moaning "END"), 
 whichavar appllas. 
 
 Laa axamplairas orlglnaux dont la couvartura an 
 papiar aat ImprlmAa sont filmto an commandant 
 par la pramlar plat at an tarmlnant soit par la 
 darnlAra paga qui eomporca uifw amprainta 
 d'imprassion ou d'illustration, soit par la sacond 
 plat, salon la cas. Tous laa autras axamplairas 
 orlglnaux aont filmia an comman9ant par la 
 pramlAra paga qui oomporta una amprainta 
 d'imprassion ou d'illustration at an tarmlnant par 
 la darnlAra paga qui comporta una taila 
 amprainta. 
 
 Un das symbolas suivants apparaltra sur la 
 darnMra imaga da chaqua microflcha, salon la 
 caa: la symbols — »> signifia "A 8UIVRE", la 
 symbols ▼ signifia "FIN". 
 
 Maps, platas, charts, ate, may ba filmad at 
 diffarant roduction ratios. Thoaa too larga to ba 
 antiraiy includad in ona axposura ara filmad 
 baginning In tha uppar laft hand cornar, iaft to 
 right and top to bottom, as many framaa as 
 raquirad. Tha following diagrams iliustrata tha 
 mathod: 
 
 Las cartas, planchas, tablaaux, ate, pauvant Atra 
 f limAs A daa taux da rAductlon diff Arants. 
 Lorsqua la documant ast trop grand pour Atra 
 raprodult an un saul clichA, il ast filmA A partir 
 da I'angia supAriaur gaucha, da gaucha A droita, 
 at da haut an bas, an pranant la nombra 
 d'imagas nAcassaira. Las diagrammas suivantx 
 illustrant la mAthoda. 
 
 1 
 
 2 
 
 3 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 8 
 
{All righl* resrrveil, 
 ADVANCE PROOF-(N«?'/ec'/ /.- revision). 
 This proof is sent to you for discussiin only, aiul on tlio express 
 uiKlcrstiindinu; thiit it is not to bo used for any other purpos.! whiit- 
 cvor,— (.See SVc J7 o/the ConitifuHon), 
 
 ^^aiuidiaij Socictij of Hiuil wngincci":.. 
 
 INCORPORATED 1887. 
 
 TPANSACTIONS. 
 
 N'.li — 'I'liis Socletyi as a body, does not hold ilsplf resporisilile for tlic finlH 
 and opinions stated in any of its puhlinations. 
 
 NOTKS ()\ HRTArNINO WALLS T\ MONTRK.VL. 
 
 I!y H. Irwin, M.Can.Soo. V.K. 
 (Tobr rend Thursday, Ootohor 25tli, IRDd.) 
 
 Ilavin<; ohiainrd somo information as to the conipleU; or partial I'ail- 
 urc oCsoniD rvtiiiniat; wnlls in Montreal during the last few years, ilio 
 writer liiinks that it m:iy not be iiltoj;ellicr uninteresting to brini; this 
 inl'oriiialion under llie notiee of the nicuibers of this .Society, in order 
 that it miiy, il' possilde, induce some of tlioni to j^ive the results of ilieir 
 experience in the construction of walls, particularly of those that have 
 proved roo light, as the failure of a structure jrenerally teaches more 
 than its succesi. 
 
 Before laking up the subjeet, however, the writer thinks it only fair 
 to state that, neither the Chief Engineer of the Bnilway on which he is 
 employed, nor the writer liimsclf, was in any way responsible for the 
 design of any of the walls herein alluded to as having failed. 
 
 The various easc^ will be taken up in the order of the numbering 
 of thn figures. 
 
 The wall shown in seeiion by Fig. 1 is of dry masonry, built of 
 ston(^s from three ieet to eight feet in length, by from ton inches to 
 fourt(>on inches thick, the beds of the stones being fairly flat, but the 
 back of the wall, in some |ilaee8, seems to bo rather (xkh', the >lones be- 
 ing too small ; the top front course was built with large, ttat stones 
 from five Ieet lo eight feet six inches in length, but the writer has not 
 been able to find out wliat proportion of through stones were n.'^ed. 
 The filling beliind the wall is principally clay, a small proportion being 
 earth and satidy clay, and was dumped from cars nninini: on a tem- 
 porary trestle. 
 
 The bank on which ll:: tilling rc.-ts has an average slope of about 
 '2^ to 1, and was not benched before the filling was begun ; .it the tem- 
 porary trestle must have iarL'.ly helped to kec p the eudiankment fVum 
 sliding, as it was built fur a doul'le track and was well biacc'il Inngitu- 
 dinally. 
 
 Indeed, possibly, this trestle, so far as it wei\t, was better than bench- 
 ing, as the jrreasy clay would easily slide from one biiieh to another, 
 unless the benches wiic cut down very ileep : however, no doubt it 
 Would bavelieen belter to haV(' benched iln' pan of the slopr below the 
 trestle. 
 
 'J'lio liank under the lillitii; is of gravel, whicli ;ibsorl..~ the melting 
 snow in the spring very i|uiekly, so mneh so that ihe eatehwater ditch 
 on the upper sidu of the slope bcldoiu has any water in it, ami there is 
 never any si^n of water passing down between the original sh)pe and 
 the clay filling. 
 
 The Wall at the foot of the slope was built with a face batli'r of 1 in 
 12, the batter of the back being 1 in 4 ; and shortly after the filling 
 was finished it pushed the wall gradinilly forward till the lop overhung 
 the biise by about I in 12, ihe embankment rising behind the wall. 
 
W;i11 slioiild li!ivc lii'i'ii fitioiifi; fliioui;li In i-'sist ovortuminn. ''Ul that, at 
 till! ^I'tiunil level, ilio lioiizniitiil t'oiiijmiiciit d' tlie ULrii*! would h,. 
 4,600 IbR., fHjr loot run, wliilo the vcrtiunl iniiipoiii'ut of tlio rusultiint 
 jirewnre, diviilcd by 2 to give tlie rrictiouul ri'sistinoc to ."liding, waH 
 4,790 lb!<., willinut takiiiL: tlio vibnition uuiiHi'd by trniiis into uoroiint. 
 Had tbc wall been built nfilic diiuensiona givi n abovt^ mh clcrivcd tVoiii 
 " Trautwini', "' it would bave appi^nred to be <|uite!<tron,ir enough, uc- 
 (onlingto Wcyrauoh's theory, to renlsl the ex(r;i thrust from vibrations. 
 
 It Hreuis also that ihis wall would have been quite strong enough to 
 •icinin the filling behind it, had it been of good uiatirial such as wouid 
 stand at a slope of 1} to 1, 
 
 The wall shown in Fig, H was also of dry rubble, built of the same 
 class of stones as No. 2, the embankment behind it boin^' of the sauio 
 nature as that in Case 'i. 
 
 The filling was made in the winter by train on a temporary trestle. 
 The wall failed completely early in the following .summer, and n part 
 of the same wall, which graduiilly stepped down to a height of only two 
 feet, was pushed down lor part of its length, and the lowest part was so 
 completely covered by the filling that no attempt wa!< made to dig the 
 stones otit, as an extra strip of land was bought to give additional room. 
 
 Unfortunately the writer did not sec any partofthi^ wall until after 
 it had given way entirely ; Imt in a part of the wall left standing the 
 writer noticed a large flat bedded stone which had been eight feet nine 
 inches below the top of the wall, and which had been pushed forwards 
 four inches beyond the course below it. This, together with the fact that 
 the wall, when it failed, wtis completely burled by tho filling, seems to 
 .shew that the stones were pushed forward and fell over each other, 
 rather than that the wall failed by overtarning, csjieaially na a wall 
 immediately adjoining it and built in cement (shewn in Fig. 4), 
 which W!is a littln higher and not quite so thick, did not fail ultogether, 
 though it was built with a vertical face. The courses of this wall, 
 built in cement mortar, could not, of course, wlide over each other 
 before overturning. 
 
 The sliding noticed in this case, as well ;i.s that montioned in Case 1, 
 seems to contradict the i^tatcmcut to be found in Mr. M. A. Howe's 
 book on retaining w.ills, edition of 188t5, page 48, that "experience 
 " and theory ]irove th.it if the resultant cuts the ba.sc within the 
 " middle third, the wall is perfectly stable, and will not yield cither 
 " by sliding or bulging, and also that the wall has a factor of .safety of 
 " at least 2." 
 
 •This statement has, however, been omitted in the edition of 1891, 
 aud the writer has contluded that Mr. Howe must have fouud that it 
 was not correct for dry walls, at Iciist when they wore stibject to (he 
 vibration froiu passing train<. 
 
 Shortly after the failure of this wall the slr)pe of the bank was found 
 to be from If to 1 to 2 to 1 at a place where the bank had ooinpletoly 
 covered the w.ill. 
 
 I'l diggini: away the debris, ul a plaC'" where nn extra land could be 
 acquired, consilcraldc ni ;-ses of snow were found <iuitc hard and frcsli 
 in the months oC August and September. Tho clay in the bank was 
 also (|uite damp aud gveaHy, and requireil very strong timber to retain 
 it while the new wall was being built. 
 
 The writer noticed in one plice that a 6" 15" stick fifteen feet 
 lotisi, with its greatest depth against the bank which was about eleven 
 feet high against if, was badly cracked. , 
 
 This stick was well braced ai the foot and at a point about eight 
 fcfit up from the foot, and carried a length of seven feet of the bank. 
 Unfortunately the writer was so busy with other work that h(> had no 
 time to take proper notes of the shoriug of the bank. 
 
 f 
 
'it 
 
 1 
 
ftq/ 
 
 .s^oj?_ 
 
 
 ^.o 
 
 /.o 2p 
 
 
TRANSACTIONS CAN. SOC. C. E.. 
 
 VOL. VIII. PLATE I 
 
I 
 
it 
 ii 
 
 I/) 
 
 z 
 
 
 CVJ 
 
 ■S97 06L'ir-*S'0^09S'G 
 SOJO'QS'e'i/^O 
 
 
■*-*j«**»«ftA.;- 
 

 SWJWH'S'M 
 
 %- 
 
 
 8- 
 
 
 ^ 
 
L 
 
 h 
 
 o 
 
 < w 
 o o 
 r > 
 
 < ^ 
 
 = w 
 ~ o 
 
 ^ P 
 
 
Jl 
 
 
 ^ 
 ^ 
 
 ^ 
 
 u 3 
 
 Q. 
 
 f/> E 
 
 ad 
 
 
 
 5) 
 
 V-<— ^ ^ 
 
 
/■/(?. 4i' 
 
 
 ■ HtmttftTMt CtmfMf/nr o^f *3QSC4la»^ 
 
 TRANSACTIONS CAN. SOC. C. E., 
 VOL. VIII. PLATE IVB. 
 
|gEaBB*S*«*?»f?&.c«BIB^B»Wffi:~*»iHf;*»-;K:-v^^^^^^ 
 
i :: 
 
40 
 
 F/q,5 
 
 30 
 
 20 
 
 /^eiffJzeiz/T^ /Sty^ect 
 
 
 
 /Sdalo> of T^&eU: 
 
 3 
 
 f 
 
5/ 
 
 ^O 
 
 30 
 
 20 
 
 to 
 
 a 
 
 FtQ.S 
 
 TRANSACTIONS CAN. SOC C. E,. 
 
 VOL VIII. PLATE V. 
 

 I 
 
„ .f. 
 
 Fro. 6. 
 
 TRANSACTIONS CAN. SOC. C. E„ 
 
 VOL VIII. PLATE VI. 
 
 /d cafleJ of Jf^e-eiJ. 
 
 > 
 
jn 
 
 "0 
 
 •^ 
 
 H 
 JO 
 
 > 
 
 z 
 (/I 
 
 > 
 o 
 
 H 
 
 V) 
 
 ^5 
 
 o 
 o 
 
 < 
 
 o 
 
 J 
 
^ q! 
 
 d _; 
 O = 
 W > 
 
 z 
 o 
 
 I- 
 <J 
 
 < 
 
 z 
 < 
 cc 
 
 »- 
 
 
 •S07OQ89-M 
 
 rr 
 
 05 
 
 
 051 
 
 *^5- 
 
 ^ 
 
 
; !i 
 
< 
 
 AcoonUni; to '•'IVautniii.'," iIiIh w;ill, if l)uilt\vllli tlic hidio lutliT 
 (111 fi'iint mill liiii'k, aliiiiild Imvc :i lliiikiii'ss ol' Ion IVcl oiio iiK'li iil tlic 
 Imsi! mill live t'ftit xix iiiclii -^ iit tin; li>|i, iiistvii'l nl" «i'Vi'ii fci't m'VciI 
 iiiclios at t'c i;roiiiiil Irvi'l ami tlii'ei> Urt iit ilic io|), us if wii.- iKniiully 
 built, iiml Triiiit\vim;'> iliiiii'iisions iirc Ims'il on ilic assuujptii)ii tliat 
 till' tilliiii; liiliinil tlir w:i!i wiiiilil lie iif ^iumI iiiatri'i.il such as woulil 
 staml lit a slope of 1 ', t.) 1. 
 
 Ciinsiilmiiij: (liat ilif fillini.' ivinilii iioi st inil at a sliicjinr slope tliaii 
 - to 1. wliil'! nioi«t, tlio writer thinks thir, aeiMiiding t'l ■' Trautwiiic," 
 the wall should have lieen at least ten ll^el -even inches at the liasjaiiJ 
 six feet at ihe top. or mIi mt fifty peieuiit. thicker tlnui it was built, and 
 that it would have hoeii better to have built ;i wall in cement mortar 
 about ten per cent, thicker than (bo wall which failed. 
 
 Tiie diaiiiaiu :> A fihows that, >if tin: wall had been liuilt accordiin; 
 to Ti'autwino's d'yicnsioiis, itshnuld have had, by Wcyr.iiich's theory, 
 a bietor ot siiiei^ of one and a half a-iinst the stones .sliding on the 
 courHc above mentioned as bcin^ eight feet iiiin inches below llie top of 
 tlio wall ; but this theory makes no allowance lor vibration eiiused liy 
 traiiLs. 
 
 Diaurani H A is drawn for the wail as it was aetiially built and for 
 (he eour.se above mentioned. 
 
 Ah the wall was made with extra large stones, its weight is taken at 
 t>.8:J of solid stone, or 135 lbs, per culiic (hot, the clay fillinjr being 
 taken at 12(1 lbs, jut cubic fool, and being assuned toeonio only t<i the 
 top innur corner of tiio wall. 
 
 The diagram shows ilial (he liorizon(al couipiineut of the thrust is 
 4,070 lbs. per foil run of the wall, and that tlie vortical component of 
 the rosultiint pressuie is 0.70I) lbs., which, divided by 2 to got the 
 frictional resistance to sliding, gives 4,850 lbs. for this resistance, 
 therefore, according to the theory mentioned above, tlie stones at this 
 joint should have iiecii just ab"Ut ready to slide without tlie help of 
 vibration. 
 
 As this wall did not fail til' ea'ly in the summer after the tilliii" was 
 tiiiisbfid, when the iday was gi'tting thawed out, it is probiblo that it 
 wouhl have been just about .strong enough had the Hlling been of uood 
 uiattrial, deposi(ed in layers and prop' rly rammed behind the wall. 
 
 Tn viewof the i'aci that the cl.iy in the bank was still unite plastic 
 and contiined sunie snow, which nii^iil b • e.vpeeted to melt not gradii- 
 iiHy and m.ike the clay still worse, and benring in mind that a failure 
 of the wall might derail a train and -end it down the bink, and that, 
 if the w.'dl gave way, heavy cliiiiiis for damiim'S would have to be 
 met, it was decided to ndiuild the will of a very heavy section 
 almost the same as that shown in Fig. 4 15, wliieli was adopted in 
 rebuilding tiu' wall stiewii by Pig, 4, the new walls in both these -jases 
 being much higher than iho'^e which I'ailed si as (n allow the bank 
 to lie widem'd niit, 
 
 l"'ig. 4 H, drawn aecording to Weyraiicb- (Keory. slmws lliat the 
 resnltiinl pressure cuts the wall at ground K'Vel .seven inclios outside 
 the middle third ; this would leiiye ilic w.ill >lahl.' enoiiL'h tliem-eticailv. 
 a> tlie tiniiiilation was uool. 
 
 The diau'r.ini shows tiie Inrizontal compunent of the thrust t i he Hl(,- 
 500 lbs. per lineal foot of ihe wall, and the vertical component of the 
 resultant pressiue tn be 48,000 lbs., which, divideil by :i to L'et the 
 frictional resi~taMee of dry stmo , would '.'ive only 24.450 lbs., as the 
 resistance to the horizontal pressure ; but as the wall was biiiit with 
 Portli.nd eeiiient mortar, there would he no dan;;er of ihe cdurses >|id- 
 inu on each other, 
 
 111 drawing this jMessnic diagram the weiiiht oT the wall was taken 
 at l(!0 I'.is per eiihie tiiot, the woijbl of the lilliiej wis takoii at 120 
 
 b 
 
Ib>. |ii'i'( uliiu I'liiit. and nil aJditiiiiiul 2 I'l'l w,i> mIWiI to tlir lioi<;lit 
 ul'tlir liiiuU 111 iiKikii alliiwaiicf tiir llic li;iiii liia'l lirini;- mi I'lnsr lii'liiiiil 
 tlu' wull. 
 
 Auconlim; tu '• TiMiiiwinc,' a wall ol llic liciL-lii shown in I'ii^. I M^ 
 with tlio uivi'M huiclnirgc. wiili the f'aei' liatli'i- iraiisl'unui'il liv 
 
 Trailtwinii'.s tlwid, wliicli, iu tliis msc, ail'U 2 inclii'S to the liici' 
 
 at ;j;rinin(l level urn) tuki'.s utf iJ fuct 5 iin-luv- IViiin the tii|) width 
 iu Ironl, ami will' the Imek chaiii;ed to tlie baltir slinwn, whieh is 2\ 
 iiiehes per loot l()r the lower |iait, uud leavin;^ the same i|uantily of 
 mai-onry al the baek, woiilil he 14 feel !{ ineliis in w.ilth at the l)ase 
 and 5 ioet 5 inches in width at the top. 
 
 These dimensions wcri^ considered oxues-ive iis they are alioiitijri 
 per eent. greater tiian I he siand.ard whieh was msimI lor tillnigs level 
 with top ol Wall, and ii was decided to Ijiiild the npper 7t leet of the 
 wall of tln^ same thickness as the standard, hut to i;ive ih ■ wall a I'liee 
 liattcr of 2 inches per foot insicuJ of 1 inch, according to the .stan- 
 dard, thus uiaklfii; the wall Hi iiiehes thicUer iit the ground level 
 than tlie standard. 
 
 Wecj) holes were left in the new walls at intervals of about fi feet, 
 and the buck of the wall was packed with small sloiie ,nul spawls to 
 secure good draim.gi: behind the wall. 
 
 The wall shown in l''ig. 4 was built lo retain a portion of iho .sune 
 bunk a.s No. .'>. Hcinj; rather hiiiher than the lii.;ho.st part (d' the 
 dry wall just dealt with, and as it was to be built with a vertical face 
 because ii adjoined a propeity line and abutted ai,'ain>t a brick alablo, 
 it was built with cement mortar. 
 
 It was pushed lorward gradually fur some months before and alter 
 the failure of tin' wall No. .'!, and would probably have failed alto- 
 j^etiicr in the course of a year. 
 
 It was pushed forward about 5 inches at the top, when it was de- 
 cided to take it down and replace it by the wall shown in Pig. 4 15. whieh 
 has been already di.scusstd. 
 
 The reason-; for t.iking it down were firsliy to sceuie. more space by 
 building a higher wall, and secondly to avoid the possibility of dam- 
 ages caused by a total failure. 
 
 According to '• Trautwine," a guod rubble nia.sunry wall of the sur- 
 charge shown in Fig. 4. and with thi' same batter at the back, should 
 be It feet ."> inch s thick at the base and 5 leet (! inehes thii^k 
 at the top, instead of tj feet 11 inches at the base and .'i feet, 
 at the top, as the wall that faije 1 was actually built, and ii .seems cer- 
 tain that if Trautwines dimensions had lii^eii adopted the wdl woul<l 
 not have moved. 
 
 The thickness of 5 feet (J inches at the top ici(uii'ed for a wall 
 with a vertical lace might have been reduced to 4 feet (i inches 
 without weakening ilir wall il tht,' lioiit lac had been given a bailer of 
 1 inch per foot. 
 
 In the case of ihe new wall, with u front baiter o|' J inches per 
 foot, the uoighbouring prujirictor was allowed to run the loofcjf his sta- 
 ble baek. and to eoniieet it with the fae,- of the wall, so as to keep snow 
 and rain Irom getting iiilo the \' sha]>e'l space beiween the two wall>. 
 
 Fig. 4 A, drawn according to Weyrauch',- thecjry. shews th;it the 
 rcsuUani pressure cuts the base a I one-fourth of iN width from the 
 front corner, and the vibrations caused iiy trains shou.d throw it siiil 
 nearer the Iroiit corner, so that it is possible that the small forward 
 movement of the wall was due to eXce.ss of Jtres-ujc- (.m the louiidations. 
 The horizuulal component of tin' thru-i is about half the vertical coai 
 Jionent ol the resultant pies>ure; but the courses could not have slip- 
 ped on each other, as the wall was buill in Torthiud cetuent mortar. 
 
 The now wall shown in Fig. 41J has shown no sIl'iis of f'liluve. 
 
 (i 
 
 I 
 
It is tnii; it iM'vni lliicliT lliiiii tii. >(;iiii|iir(.l O.I'.K, rctiiiniM'^ wiill, 
 uliii^h limy ut lii'^l -^i;^!:!. us it did ti tiic wiitcr, mi'Iii tuci iliiul;. Ii 
 liiii.ot III' lioi'iii' ill iiiiiid, limvovrr. tliiit llii> ■^luiulard wa- dtsiuni'd imi 
 .-imply tip I'i'taiii an laitii liauk, lnit tn raiiy and Niaiid 'lir tlini«t and 
 vihriitiiin.s of a liravy liain inniiin;: cImm' up tii il, riilioroii t:in.i;iinls ur 
 hlmrp uurvrs ; and i( lias ijuiic tlii< I'm- many miu's and in a variety id' 
 lilacfc without liny lail'.ircs tlial llic wiiltr r\ir Inanl id'. 
 
 'I'lic wall »hti\vii III \'\ii. -11! lias Imwivir til btand a .-iindi.ili;r id' 
 1.27 ti) 1, iiml tliL' lliiust id a sid't cday Ijaiik, su tliat llic writrr tl.inks 
 it ih fairly propmtiniud. | 
 
 ]''ig. 5 slii'WH till,' old ivtaiiiiiiL; wall mi Scii;iit'iirs strcol liclwivn 
 St. Auloiiir i-tin'l and l*()r(dusti.i' .stivct, wIiltc it ninn diiiuiMiiilly np 
 till' fat'o id'lliu liill, 
 
 'Jill' dinit'M.-iiins wim takuii liy llji' wiiui wliili' ilio wall was liidii;; 
 taken iliiwn a clmit. linn' a^u. and si'i'in rallirr tmi small ronsidi'rin;; 
 llir aniiMiil id'su'i-hariir, yrt its pai'tial failiui.' .-r'His tn liavi' lircn duu 
 111 llii' Want (ifpliijii'i call.'. Had tlir joiiils lii'un all wrll raki'd out and 
 iilli'd with i^oiid ci'nu'nt iiinrtar, it would donlith'ss havi' .stood as Imif: 
 as it wa» waiitud. 
 
 Thr writer nndei'stands that il Was rcniuved to make wav fur the 
 widi'iiini: of the street which entailed the Imilding ol' a much higher new 
 wall. 
 
 Wlicn the old wall w;i,s removed and the hank iiu h.ick, its face stuid 
 );eifectly well at a slo]ie ul' abiiiit :} to 1, apparonlly aliiiust vertical, the 
 material nf the hank hoiiii; mostly compact line gravel and sand, which 
 would stand at a very sleep shipe so liini; as it was imdislurbcd and 
 protected from tin weather ; in lint, it looked firm enough to rei|iiiii' 
 a icry thin wall if proper ]iiecantions were taken to prevent water from 
 lodginu' hehind il, and if the lilliny replaced at the back of the wall were 
 well rammed so as to jirevent '.my movement of the old bank behind. 
 
 It will be noticed that the hank retained by this wall is of a totally 
 diftercnt nature from those already alluded to ; the former is an old 
 solid undi.'iturbcd gravel bank, the latter new clay fillings dnniped from 
 cars with no attempt to spread the material in layers or eonsolidate it 
 behiud the walls. 
 
 The writer hopes that ihe City Surveyor will furnish the Society with 
 a section of the new wall on Seitriieurs street, uiviie.' the profile of the 
 bank for .some distanee b.nk from the w.dl. 
 
 Aeeording lo Trautwine. a rubble masonry wall with a siii'chari.'e 
 similar to that shown in Fii;. 5, and witli the given batter, should he 
 5 lijol5 iiielios thick at the ^rnuud line and lifeet 9 inches at the top, 
 instead -f '■> feet I! inches ai the base and 2 feet at the lop; 
 Iiowevor, it must he bone' in iniim thai 'rrautwine's dimensions 
 are given for a bimk of elean dry sand which would exert far more 
 pressure tlian the bank in i|iieslii)ii. tin; surf lee of which above and 
 behind the wall was covered with the old sod an I bound together with 
 the roots of trees, and the mateiial of the iiank was. ^^ already stated 
 vi ly compaet. 
 
 The FIl'. 't\ is drawn aeeordini; to Weyraueh's llieory fur a sur- 
 face surcharge shilling back it the same aiiu'lo as the natural slope and 
 liir material standing at a slope of Ir, lo I. The wi'i;;hl of the hank 
 is taken at 110 lbs. per eiibie loot and ol'llie wall at lliil lbs : and the 
 
 bank is assumed lo eonu within :i loot oltlii' I'l |. ul the w.'ill. 
 
 The diagram shows that the resultaiit pi'e-.-uii' cut- the leis • at about 
 one-third of it- thiekne.ss IVoin the I'lui'.i eonier, so thai the w.dl should 
 have been . safe Irimi overt iirniin;. 'I'he hoiizonl.il eonipoiient ul the 
 thrust is 1,G0II lbs. per foot inn. »liile the verlieal eoniponcnt of the 
 I'e-ultant pressure, divided by 2 In 'iet the fiietional resi.staiiee. is 
 -t,.SI:{0 lbs.; bill as lie wall was built with mori;ir, there would be -iilli 
 
 7 
 
iiih( n>i>.|;iii(i' Id >liili!i'^ iil llic yihUy imli ( il. in tlii> r;iMi', ll^ llio 
 • tiiiiiv Wire snmll mA ikiI li\ii|i in ri!s;ul;ii' I'oiir.-VN, llir liiiili'inv tor llii- 
 'tnni's to >li(ii' III! liicli iitlur mIkhiIiI haw lu'vu small. 
 
 l-'ii;. t! is II sti'timi (if tlif wall in IVimt oT ilic Ai'i'lilpi.-lio|i's I'alacr 
 on liiiifhui'lu lii'M >inrt, IroMi iHini'nsioiiy uivin to tlio wiitcr l)y tln^ 
 ioiiiiai;l(ir, wlio laiilt it aliont ltd yiiirs an", ami iv|iiiiriil tlio |Hirtion 
 (Vnni ilir t'tuiancc 1 astnly to Maiistielil ^t^L•('t, a li'nj;tli ot'aliont two 
 linniiit'il and iwrnlv loit, iibont nine yoiil's ai.'o. 
 
 'I'Im' wiitci was I nalilcil to pi-t tills infoiiiiailon ilnuuuli tlic kiiidnrHH 
 III iMc. A. Ilolhit, llifaiconnt.aiit.Ht tlic I'lilai'o. 
 
 'I'liO iioiti t ilio wall IV.iin tin' ontiMiiPc wotorly U> Catliedral 
 
 <(iici, a Icnulli iil'alioiil 57 H'ct. Im^< not liccii I'l'palri'il Hiiico If was 
 Imiii, loll Ijiopi's out at llic lop 4 inclius ai. llui niiildlc of Its lonj;!}!. 
 Till' iinis 111' ijiis piirilon ail' wi'll Hod in liy tlio wall on ('iitliodnil 
 siri'i't and by llm nlnin at ilicintriuici', so tlial llic iinsiipport<Hl loiij^tli 
 l.s iinih^r "lO fri't. 
 
 Tlie wall U all Imilt In coninion linio inortar. and, -o Car as tho 
 writci niiiombei.s, llic part npaii'tid only Imd llic Inint stones tnkuii out 
 and 11 placed with new innrtiir ; tlio .stoni'H soeinod to have licen puslicd 
 I. lit by till' trot ziiij,' and thawing of wuIli lod;;('d belilnd tlieni. 
 
 Tlicrc nil' no wicp IioIi'h In the ca.»tcrn part, bnt it has one drain 
 'oi'hind at the ci'iiln; wliich discharges into tho ^iroct stwor. Thi" (;nn 
 tractor said that he wisliod to have put iiioro drains bcliind, Imt was 
 not allowed to do ito ; and also said thai he considered surh drnins 
 naich better tlinii weep bole.s, which arc liable in this country to get 
 I'rosfcn up in the spring froiii the cold of the inner part of the wall, and 
 that they let the f'riiBt into the body of Jie wall. 
 
 The writer hopes some luerubers will give their opinion on this point. 
 
 'riicrn Is no drain behind the shorter western pari., Oiit there i.s one 
 
 weep hole at the street coiner. If the dlnitdsions given are correct, 
 
 the wall is fully thick enough, and if built In cement and properly 
 
 drained behind, >hould never have moved. 
 
 There is a thin Cement catchwater drain behind the eastern portion, 
 as shown in Fig. <i, which takes away the surface water from behind 
 the wall. 
 
 Kig. 7 is a section of the will on the wcsl side of Bleuiy sircet 
 and just above Dorchester street. 
 
 It failed by bulging outon the face, and was taken down and rebuilt 
 last year. 
 
 Thii'Ugh the eiMiitesy of the l!ev. F. \, lienaiid, Superior of St. 
 Mary'.s College, tlie writer was pi'rmitted to get the dimensions of tUi.s 
 wall, and other Inform itlon a.s t'l Its oonsiruct ion, from the architects 
 who licsigued it. 
 
 It apicais lliat the iiiclillicis at fiisl mIvlshI ibe building of .i 4 
 foot stuiic wall, similar tu that wliieh carries the pii'.-ent new liuiliilii;:. 
 whiih could bi' usid lor a future ixfiiisinu along Hleiiiy street, linw- 
 i VI r, ;lie prop! ieleis did nut wish to go lo >n inneh c.xiicnse, as they did 
 not think of biiilillng any further in ll.e fiituir, and wished to reduce 
 the cost of the wall :is niucli as pos.-IMe. 
 
 It was thercfoi'it deeided to pave the slope as sliuwii, build as light a 
 wall as po.>-sible, and fill the V .shaped spiice between tin; slope and the 
 
 wall with light material, imt lai 'd, so as to allow the water I'allini; in 
 
 the yard to [Mjiiolate tliroiigh the filling and run off throngh the drain 
 at the foot of the slo]ie. the drain beint: covered with Hat tiles laid with 
 open joints. 
 
 The wall was let by eonlraet In llie lowest tenderer, not the eon- 
 tractor for the new building, iind the price was mily sufficient to leave 
 lilin 81.50 per cubic yard lor the rubble backing. 
 The writer was allowed to see the ^pecitication for the wall. Jt wan 
 
 8 
 
I'urul'ully (Iniwii up. iiii.l piMvidcd liir an iiui|i' iiumLor of lit'iiilcr», 
 
 wliirli were to bt.' tlii'Oll^li .-li s wlicii'Vcr llio wall wns 'J' U" or los in 
 
 lliii'kness. Italwi ralltid lor 'J to 1 IVirtlttri'l (ciin'iit mortar, luadi' witli 
 Whilc'H or ».•< good (.'(■mi'iit. It iKhv* not sci.ni tlnit tln^ prosisuni ol' tlic 
 furlli tilling was the nin>c of ilio Iiulgc in the \v,dl lor which tin; IJuild- 
 ii.},' lns)ifctor li:id ii taki n down ; for wIk n it tkilcil, tin: tillin.i,' wa^< not 
 
 saturated with watrr. ninl iIhtc had luci tinii' lor I'm.st to atlici it. 
 
 It will Ih' win llial tin: wall wa.s aljout thick umiupli to retuin tin; 
 lining, tor 8'!)" mulli|ilie.! by (Uglvcs il'C" lor the tliicknusi- at grounds 
 lt'V« I, wbili- tlu; actual tliickntv^s of the wall wax 'S li", and the vcrrtiial 
 |iri»uri: of thi' I'Xiiii 10 Icit in height would have bucn more iIi:mi 
 ('iioiii;li to chunrc the stability of the wall, providing ii were properly 
 built. 
 
 In faet,tlnMVull bulged ont in one pluce wheru there wan no lilling, at 
 all behind it. 
 
 The greate>t bnlgr was at a height of i; or 8 feet above the level 
 ol'ihe sidewalk, or almost as high as the top of the tilling, and this seems 
 too high to bavc been caused by the lateral pre-siirc. since the greatist 
 bulge shouhl be about the centre; of pressure. 
 
 The failure seems to have been due entirely to bad niortiif, want of 
 j)ropcr headers, and through stones and excess of mortar iu the backing. 
 The writer oxamined the wall carefully after it had been itliuost all 
 taken down, ami could find no trace of a tlirougb stone or of a good 
 header at all, and tin.' wall seemed to have parted in two, the cut stone 
 front separating from tin' backing. 
 
 The face was boilt of cut stone with a bed of only about 8 incliiis 
 in depth, and thebiicking was of small stone with altogcllKM- too much 
 mortar in the joints. The innrtar wa.s made of very fine dirty sand 
 with too little linic in it, and what little there was no doubt of the 
 useltfw fal lime now exclusively used iu Montreal, because it is 
 cheaper than good lime made from the black limestone which takes 
 more fuel to burn it. 
 
 The con.so(|ucnce was that the upper 10 feet of the wall compressed 
 the mortar in the backing, while the ent stone f'ront,witb i's thin Joints, 
 Would scarcely eompre.^s at all, and the front was forced out. 
 The writer has brought a sample of the mortar ♦or inspection. 
 The architects protested against the way in which the walls were 
 being built, and notified the proprietors not to i)iy for it; it was taken 
 down not long after completion, and rebuilt with the money retained. 
 
 The writer believes that this wall would have stood perfc'ctly well if 
 a sufficient number of headers and through stones b;id been used, and if 
 the moi'lar had been made of good eement and saiul, in accordance with 
 the specification ; but thinks that it would have been better to have ar- 
 ranged to set the upper 2 tiiot portion 4 inches back from the face of 
 the part below. 
 
 The diagram down in l''ig. 7A. diawn according to Wevraueh's 
 theory, shows that the wall would be iu no danger of overturning and 
 that the courses would not slide on each other. 
 
 In drawing thi.-, the masonry was taken at l(><t lbs. jier enbic foot 
 and the loose earth at SO lbs. per cubic fool. 
 
 It may be added that a part of the wall wliieli bulged slightly is still 
 stalidir.g. 
 
 The writer regrets that the various ea.ses referred to in this paper 
 are niiimportanl works, but thinks that, in describing them as fully as 
 po.ssibic, the results may prove useful. 
 
 They .seem to shew thai Weyraueh's theory, iiili-tliiiindii ti/i/>/ieJ. 
 wo\ild be a good check on empirical rules. 
 
 This theory assumes that the filling is without eohcsion, and that its 
 free surface slope would be a ]>l.me surlaee. 
 
 1) 
 
Tlli- "rilliNlu III 111, Wi.r,.(, .1 lillillllll lll.il I lillill- ■■linlil ;l«-|l||ll', uimI 
 ll,:lt i- Wliill i> 111 111' ::llliri|rll iLMillfit ill till lJ«i' 111 l.ll|tl:itlkllliMl> iril'li' 
 
 l>'> ilMiii|iiii'j I'liiiii li.iiii^ III! ii tri>ll'', 
 
 III llic r.M.' i>r .1 »;i|l ilit"i.i|ri| til I'i'hilii :i ii.iIiiimI Ii ink, |';ii I I'l' wlllrll 
 i> 111 Ic i-iil iiHii y, iiiilliii ,;; I iil i \|it'i'ii'M'i' iiii'l 'jmiil Jiiilfjiin iil will lie 
 lil0|'i r uiiiilc ill ill III iiiiiiiii;; l.iiw I'ui' llimrj iiiiiv lir iK |Kirtiil rrmii, ,iiiil 
 ill, • '.uii;!'! II i|iiiii' III lir >ii|i|i|i'iiicnlril liy a lli"i"iii;li kiinwli il'^ml llii' 
 iialiiii' lit till' biuik. 
 
 A iiallllii' ll.illk. liki lll.lt lit Sl'ij;lu IIIM NHrit,iillls lri|l|ili'S tn I"' 
 
 |i|ii|ii'ly iliiiiiH il. mill 111 lia^r llii' iiiiitiiial riphifil krliiml tlir wall 
 w'l'l! riiiiiiiii il, .-II tluit till' iiiw fail may nut ■j,railiiiil'\ litn.^m iiml iiii;>.'< 
 Inrwiiiil (ill at lai-t llicir will Lo im luucli jirc.iRuru a^^ainst tliu wall an 
 friiin a nrw li.ink ; i( iIm'ki' picraiiliniisarr rnniril mit, a vrrv lii^iii wall 
 iiiiijlii III' u.M'il. 
 
 'I'liii Mi'iti'i' i- Will iiwaii' llial ^iicli ail oiiiim^nl autluuily aM Sir Urn 
 jaiiiin Hakir, wIium; I'lar'tiial inks ai'i! wi ;j;iiiiil, liiiuiH to think tliat 
 tliiiirs is I ntirrly at sia willi ii'^aril to rolniiiing wallx, it'ono may JU(1k<' 
 li'uin lii>lpriik 111 till' lali'tal |irr»iiro of uartliwuik [.iililislicil liy I). \'iin 
 No>lranil in IS-Sl'. 
 
 With all iliuiili'linncr In lhi^ (linliimiilHln il iii'^iiu'i r, the writer tliiiikH 
 that III' lA a iittli! liai'il nil tlirory. 
 
 Miiiiy of tin" t'ailiiirs iiK'iitionwl in hi.-* Imok vvrrr iliii; to hail I'ounila- 
 lintis ; micli (;aM'>i as iIumi ii'ijiiirc i|iiitr ililluii'nt IrratiiR'iil fVulii tlinv' 
 wlt'ii' thr r<iuinlat!on i- ,«oliil, ami .should Ih' ki'|it i|uito so|inrali'. 
 
 Without atti'm|itiiij; lo criticise thu book iihovi' iiit'iitiuiiL'il in f'iill,tlio 
 writi'r may nicnlioii tlmt. on \.tvj^vt* 2G ami .'J I Sir Ijrnjamin ohtaiiis tho 
 jilL'>Miii) aiiiiin.st hvii walUhy lalculaliiig the luoiiii'iit iil .stiibilily of tlio 
 walls al'ti'i- iliiy h:iil hri n iiri>>i'il loiwaid oonaiili'rahly. .Surely this is 
 iidi laii- tn ilmry, >iiiri: till' force ii'i|Ui'ii!il to .>liut a wall anil push it 
 mil lit pliiiiili wiiiilil 111 iiiiieli greater than that liiially leiiuireil to push 
 il ilowii when ieaiiinj; ; anil, he.sides. it is reasonable to assunio that llie 
 [iiisMiri' Wiiiilil he reliivi'il by the llirwanl nioveinent of the wall, unless 
 ihu tilliii:> hail "Il I'lilii'siiin, iiml unless it were kept up Id its original 
 level. 
 
 Again, 111 page -19, riferring to eraek> in elay over luunels, the 
 I raeks bilng |iiiilalily ilong Ihe line nt least resistanee, he a»iiines this 
 In he (.'mill mil's liiie nl li'iist let islaiice, iiiiil iiigues against his theory 
 biitiiiiM' ihe iraeks had a slope of .J to 1. 
 
 There dies not seem, however, to be any Miy good reason lor doing 
 so .siiiee C'liuloinb'.- iheoiy did mil t.ike inliiaecouiil the eohesion of the 
 , lay ; and, lie-ide~, the writrr thinks ihat ihe ease of a iiki.ss of elay over a 
 tunnel is not i|nili' the -aiiie as that of a bank against a retaining wall 
 with a phiiii' t'lee. 
 
 In l•ll||l■lll^ioll,tlle writei wmild :,sk iiis fellow- me iiibeis to hear ill mind 
 that llii> papii' ha- In en written lather leu hiirrieiliy fur proper eonsid- 
 eratioii, and trusts lb. it they '.vill i;i\r as iniieli iiifirnial inn as they lali 
 on the snbjeel. 
 
 10