■> IMAGE EVALUATION TEST TARGET (MT-3) k <>/« ^ 1.0 I.I mtli, 121 ■^ Itt 122 iM 12.0 IIS lU lit IL25 III 1.4 I 1.6 Sciences Corporation 33 WEST MAIN STREET WEBSTER, N.Y. 14580 (716)t7il-4S03 CIHM/ICMH Microfiche Series. CIHIVl/ICIVIH Collection de microfiches. Canadian institute for Historicai IMicroreproductions / Institut Canadian de microreproductions historiques Tachnieal and BiMioflraphie NotM/NotM taohniquM M MbliographiquM Tha Instituta haa attamptad to obtain tha baat original copy avaiiabia for filming. Faaturaa of thia copy which may ba bibiiographicaily uniqua, which may altar any of tha imagaa in tha raproduction, or which may aignifleantly changa tha uaual mathod of filming, ara ehaekad balow. □ Colourad eoyara/ Couvartiiro da coulaur r~n Covara damagad/ Couvartura andommag^a I I Couvartura raataurAa at/ou pallleuMa I I La titra da couvartura manqua Covara raatorad and/or laminatad/ Couvartura raataur4« Covar titia miaaing/ La titra da couva □ Colourad mapa/ Cartaa gAographiquaa t* covtiiaur □ Colourad ink (i.a. othar ttian blua or black)/ Encra da coulaur (i.a. autra qua blaua ou noira) □ Colourad plataa and/or illuatratlona/ Planchaa at/ou illuatrationa an coulaur □ D D Bound with othar matarial/ RalM avac d'autraa documanta Tight binding may cauaa ahadowa or diatortion along intarior margin/ Lar«liura aarrAa paut cauaar da I'ombra ou da la diatoraion la kmg da la marga intiriaura Blank laavaa addad during raatoration may appaar within tha taxt. Whanavar poaaibla. thaaa hava baan omittad from filming/ II aa paut qua cartainaa pagaa blanchaa ajoutiaa lora d'una raatauration apparaiaaant dana la taxta. maia. loraqua cala Atalt poaaibla. caa pagaa n'ont paa At* filmAaa. Additional commanta:/ Commantairaa tupplAmantairaa: L'Inatitut a microfilm^ la maillaur axamplaira qu1l lui a it* poaaibla da aa procurar. Laa dAtaiia da cat axampiaira qui sont paut-Atra uniquaa du point da vua bibliographiqua. qui pauvant modifiar una imaga raproduita. ou qui pauvant axigar una medHieation dana ki mAthoda normala da f ilmaga aont indlquAa cl-daaaoua. D Colourad pagaa/ Pagaa da coulaur D Tha to til Tha □ Pagaa damagad/ Pagaa Pagaa andommag^aa Pagaa raatorad and/oi Pagaa raatauriaa at/ou paillcuiiaa Pagaa diacolourad. ttainad or foxat Pagaa dicolorAaa, tachatiaa ou piquiaa Pagaa datachad/ Pagaa dAtachiaa Showthroughy Tranaparanca Quality of priii Quality in4gala da I'impraaaion Includaa auppkimantary matarii Comprand du matirial suppMmantaira Only adition avaiiabia/ Saula Mition diaponibia pn' Pagaa raatorad and/or laminatad/ Pagaa diacolourad. ttainad or foxad/ Pagaa □ Pagaa datachad/ Pagaa Fy'l Showthrough/ pn Quality of print variaa/ pn Includaa auppkimantary matarial/ rn Only adition avaiiabia/ ofti filmi Grig bagi thai alon otha firat aion or ill Pagaa wholly or partially obacurad by arrata •lipa. tiaauaa. ate. hava baan rafilmad to anaura tha baat poaaibla imaga/ Laa pagaa totalamant ou partiallamant obacurciaa par un fauillat d'arrata. una palura, ate., ont it* filmAaa i nouvaau da fapon A obtanir la maillaura imaga poaaibla. Tha ahall TINl whit Map diffa antii bagii right raqu mati Thia Itam ia filmad at tha raduction ratio ehaekad balow/ Ca document aat film* au taux da rMuetion indiqui ci-daaaoua 10X 14X 18X 22X 2IX 30X y 12X 16X aox 24X 28X 32X V Th« oopy fllmMl hcra hat bam raprodueml thankt to tlw ganwotity of: Ubrary of th« Public Archive* of Canodo L'axamplaira fUniA fut raproduit grica i la g4n4roalti da: La bibiiothiqua dat Archival publiquas du Canada Tha imagoc appaaring hara ara tha bact quality poaaibia eonaidaring tha condition and laglb*llty of tha original copy and In liaaping with tha filming contract apacif icationa. Original coplaa In printad papar oovara ara filmad beginning with tha front covar and ending on the iaat page wMt a printed or illuetratad impree- aion, or the back covar when appropriate. All other original copiea are filmed beginning on the firat page with a printed or llluatratad imprea- aion, and ending on the Iaat page with a printad or llluatratad impreaaion. Lea image* auhrantae ont At* reproduitea avac la piua grand coin, compta tenu do ia condition at da la nettetA do rexemplaira film*, et en conformit4 avac la* conditiona du contrat da filmage. Lea exemplairee originaux dont la couverture en papier eat ImprimAa aont fiimAa en commen^nt per le premier plat et en terminant aoit par la demlAre page qui comporte une empreinte d'Impreeaion ou d'iliuatration, aoit par le second plat, aelon le caa. Tous lea autrea axempieires originaux sent filmte en commen^ant par la pramlAre page qui comporte une empreinte d'impreaaion ou d'illuatration et en terminant par la darnlAre pege qui comporte une telle empreinte. The iaat recorded frame on each microfiche ahail contain the symbol ^»- (meaning "CON- TINUED"), or the aymbol Y (meaning "END"), whichever appliea. Un dee aymbolaa auh^ants epparattra aur la darnlAre image de cheque microfiche, selon le cas: le symbole -^ signifie "A 8UIVRE", le symbols y signifie "FIN". Maps, plates, charts, etc., may be filmed at different reduction ratioa. Those too large to be entirely included in one exposure ara filmed beginning in the upper left hand corner, left to right and top to bottom, aa many frames aa required. The following diagrama illustrate the method: Lea cartes, planches, tableaux, etc., peuvent Atre fiimia A das taux de rMuction diff^rents. Lorsque le document est trop grand pour Atre reproduit en un seul ciichA, 11 est film* A pertir de Tangle supArieur gauche, de gauche k droite, et de haut an bea, an pranant la nombre d'imagea n^caaaaira. I.es diagrammes suivanta illustrant la mMhode. 1 2 3 32X 1 2 3 4 5 6 (All righti raerved.') ADVANCE PROOF-('S'"'i/«c' '" revinion). This proof IB sent to you for diiscuRsiui) only, and on the express understandinij! that it is not to be used for uuy other purpose what- ever, — (.Vee Sec. 47 of the Constitution). ^anadiaij ^ociety of ^m\ (^ngintens, INCORPORATED 1887. TRANSACTIONS. N.B — This Society, a^ a body, does not hold itself responaible for the facls iind opinions stateil in any of its publications. A CUBIC YARD OF CONCRRTK. By Henuy F. Perley, M. Can. Soc. C. E. To hr read Thursday, %th November, 1893. With the disappeiiranee of timber from the move settled parts of Caiuidii, and conscqutnlly its increaiie in cost, other materials will eventually have to be used iu the eijiLsti'uction of our public and other works. Heretofore, timber has tidieu the lead «sa constructive uiaterial, by reason of its existence everywhere thruuithuut the I>ominion, the ease with which it eould be obtained, and its apparent cheapm>ss when compared with tither and miiuh more durable materials. Except in Canada and the United Slates, the use of timber has been to a very great extent abandoned, and in its stead, iron, steel, ;-tono and concrete lire used, the two first ]irinei(ially in superstructures, and the latter two in foundations and sui)orstruelures as well ; and of all these materials it would appear tliat eonereti', has pnivcd to be the cheapest and most advantaj^eons to use in the construction of break waters, wharves, dock- walls, sub-aqueous foundations, etc. ; and the day is not far distant when it will be fully comprehended that all large and important works in Canada will, to a very great extent, have to be constructed with that material. The object of this paper is not to point out where concrete can bo advantageously used, nor to show that, U'.ough entailing a larger initial expenditure, it eventuiiUy becomes, where depreeiatimi through natural decay and woar and tear, and the cons('(|uent cost of repairs are taken into account, cheaper than timber ; for a solid structure of concrete is better in every way than the wooden boxes tilled with loose stone, con- structed by private as well as public enterprise, which pass and serve as wharves, breakwaters, bridge-piers and abutments, etc., ihroughout the length and breadth of our country ; but merely to make a few remarks on the composition and character of concrete as a constructive material, without any reference to its use, or its cost, in both of which cases the question of locality becomes an important factor. Concrete is an homogeneous mass, — in fact, an artificial slonc, formed by the admixture of lime, or cement with gravel, broken stone, sand, etc., in fixed proportions, the strength and durability of the compound being directly in j)roporlion to the (jualities of the lime or cement cm- ployed, the nature of the stone, sand, etc., selected, and the manner of their admixture and mode of deposit in situ. For the purpose of this paper, the subject has been divided into four parts as follows : — 1. Cement. 2. Components (stone, sand, etc.). 3. Mixing. 4. Deposit. 1. CemE! T. In concrete, cement — for the use of lime will not be considered — • takes the leading place, and on its goodness the goodness of the resulting massdipends, always provided that the other components are good and the mixing, etc., has been lionestly done. By cement is nipnnt tlmt sub^tntico, ciihci ii naturnl or an artificial production, whioit po,sses.se.s tin; propiirty, when luixod witli water, of setting in coinpuMtivcly short poriodH, oillicr in the air or in fresh or Halt wiitor, and uliso of attaining; •iruiitvr .str(3nj^th iind Holidity with an advance in ii{;i', and tlu'so prop(Ttic.s arc poswsHud in their Cullest oxiuut by what arc clasHod as PnrlUind Cement». It is true tlmt certain ceniontfl bearing local nnnics, but not ninnufaclurL'd in the sanic manner as the Portlands, jKisscs-i the sanio properties, but only to :i certain extent, and often that extent is so small as to preclude their usu in any important work. Portland cement is an artifi(3ial product, resulting from the mixing of certain classes of limestone or chalk with clay of a suitable cjuality, in fixed proportions, iind calcining and grinding the product. In tjio southj[of England the hard eluilks and clay de|osit from the beds of several rivers are used, but in the ncprth of Kngland and on the Conti- nent, limcsloDe :nd field clay are the eom|onents, the proportions not varying far from 72 per cent, of stone to 28 per cent, of clay.* / On analysis, a somple of good cement should show the following oon- stituents — Silica 22.23 percent. Iron oxide 4.32 « Alumina 7.22 " Calcium oxido 60.59 " Magnesia 110 " Sulphuric acid 1.00 " Carbonic acid 0.80 Water combined with lime 1.05 " Insoluble and other matters 1.69 " ^ 100.00 In the above there are three constituents, which, when they exceed a certain percentage, are objectioniible,viz. : magnesia, when it exceeds IJ per cent., and sulphuric and earTwnic acids, when they exceed IJ per cent. Another objectionable element which cannot be detected by analysis exists in the shape of/rcB lime, but it is as.serted that the measure of carbonic acid is the measure ol' the amount of free lime.f Lately in England there has come into use a cement made from ' ' slag," or the rel'use from the blast furnaces in the smelting of iron, it having been found to contain generally all the chief ingredients found in Port- land cement; but all slags are not equally adapted, for those which dis- int^rate and fall to powder spontaneously are wholly unfit, but they have been made use of by unscrupulous manufacturers, especially in £urope, for the purposes of adulteration. The following is an analysis of good slag given in the " Proc. Inst. C. B., Vol. 105, p. 221." Silica 24.19 percent. Iron oxide 0.93 Alumina 16.30 Calcium oxide 46.53 " Magnesia 2.08 " • Sulphuric acid 2.05 " Carbonic acid 0.65 " Water combined with lime 6.45 " Insoluble and other matters 0.94 " 100.03 It will be noted that there is a deficiency in calcium oxido (lime), but this is made up in the process of manufacture, which is freely quoted as follows : — " The slag, as it emerges from the blast furnace, is passed through a stream of water, which reduces it to a spongy and readily crushed material known as ' slag-sand,' which is ground to a • G. F. White—" Proc. Inst. C. E., Vol. 02, p. 185." t A pamphlet on Portland Cement, by W. W. Maclay, C. B 2 S I ,1 flne powder between mill-stonog. Ah no slug in itwilf contains a guffioient amount of liuiu to pioduou coinontitioUH uction, thu requisite anio'int in miidu up by tliu iiildition of 'J5 \)ur cent, of the niixtuio of ulnkcd lime, obtained f'ruui puruor liit liuiux, which in the act ul Miaiiiug fall into an cxtrcmi'ly Hnu ijowder, tiner than can be produced by nny nieclianiual means. " Alter the mixture of the Hlag-sand and lime, a chiir^o i» passed into a metal cylinder nlMiut 4A ft. in diauiotcr, partly tilled with iron or steel balls from 1 to '1 ins. diameter This cylinder rovolvetthoriKonlully and slowly, and in coM.seriucncu of tlic eru^hin^; and pounding action of the balls, the friction betwetm them, and tlio very coinploto inter- mixture of the ingredients eaiised by the rotary motion, the particles of lime and sla^; iiro most intimately uniteil and reduced to a smooth silky powder, re>cmblin<; to the touch the finest flour. After remaining in tlie drum lor about one hour, the contents arc withdrawn, and the cc- meni is ready for use. "The undermentioned results were obtaining in testing samples of slag-cement at the Koyal Testing establishment for Building Materials, Berlin, Germany. 'IViisilo per s( 7 (lays. IllS. (147 427 Htrpngtli . ineli. 2S (lays. Compressive strenn;lli per sq. inch. Remarks. MLxture. 7 (lays. 28 (lays. The briquettes were one day Nent. 1 ct. ;! sand. lbs. G92 .■)09 lbs. lbs. 4,2«!» ill the air, the remain- der in water. " In colour, slag-cenicnt is lighter than Portland, and, owing to its fine grindinir and partly to a lower sjxicific gravit}' both of the slaii aud the lime, tlu^ weight per eubic-f'dot seldom exceeds 75 lbs." * The process of manufacturing Portland Cement need not be described, as it is now well known, but it ditt'ers materially from that adopted for slag-cement, and it is at once apparent that it is a much more extended and therefore a more extensive process. The goodness of Portland cement depends : — (1) On the proper constituents of the uinterials employed ; (2) Upon their being propi'rly mixed in the risrht proportions; (3) On the exact amount of calcination ; (A) The degree of fineness to which the clinker is reduced by grinding; (5) The tborouglnv'ss with which it has been sieved to obtain only the finest particles, and the rejection of all coarse parts ; and (G) A c;irelul air-slaking tor at least one month, to permit the cement to cool and purirc itself of frec-lime.'t" Uniform finen(!ss is almost an absolute necessity, and in passing a sample through a sieve of 70 meshes per linear inch, or 5776 inches per square inch, not more than G per cent, of residue should remain on the sieve. All coarse particles, such as small lumps of unground or partially ground clinker, arc not of any cementitious value, indeed it is better louse more coarse sand in concrete than such particles, because they can only be regarded in practice as sand, as they reduce the eflfec- tivc proportion of the ct^ment. To show the eft'eet of coarse and therefore inert particles in cement in the manufacture of concrete, take a specification which requires concrete to be made with 6 parts of brok(»n stone, 2 parts of sand, and 1 part of cement, tlie product being known as G to I concrete, and suppose that a sample of the cement supplied, on being sieved through a standard sieve, leaves a residue of 20 per cent, of coarse particles, then the actual available amount of cement is 80 per cent., and the mixturu becomes 7^ to 1, in.st«ad of 6 to 1, as called for. • Proc. Inst. C. E., vol. 105, p. 221 et .seq. t Vide General Scott's paper in Proc. Inst. C. E., vol. 62, p. 70 et seq. Tho antlmritios* cnnsultol in (lio pti-p^imlion of tliis paper nffreo on the iiocoHsity ot'dciuaiidiM^ II pioporly ^rounil cement, for exactly in the drfproc that it is not ho, in \>h viiluo n»i conicnt thrown awny, ax tlic partirlcHoflmnl clinkfr xvliicli will not pass through it 2500 mesh Hievc, will mil <.'iv(i iiiiy ^ood icsiilts. In Goriuniiy — and it inny hero liK stilted par piirrvthi.if, lln'l llu" (iurnian Porlliind ccmrntH give, a!< a rule, better rosulfH limn llione nl' KiiL'li^li iniikc, fine finding beins required — tlio recofvii/cd (ilniiiliird i.* tli:it the residue iiniS't not exceed 20 per cent, on a Hicve oi !tOO nienlies per srniare centimetre, or 580(1 ine»lie» per K]. inch, but an nrliele wliicli will mily leave 10 pereont. Ih Hupplied. In Kiijjland tlic^ lenidne left on ii Hieveof'2850 niesheH per 8q. inch viirieR i'rom 15 to 2? p<'r uenl. The weight ofeement per cubio liiot euiinotbe taken as an indication of its strength, for a lienvily burnt elinker, if conrHely fjround, will produce a heavier article jier eubic font ilmn tlie » .nt of equul fineness obtained by sifting. The difference in strength of coarse and fine cements is not ascertained by testirg them neat, for of the two the coarser would generally appear to be the stronger, and it is only when mixed with sand that this can be seen." For concrete, as for mortars, cement cannot be ground too fine, and as cement should be sold by weight, and not by the barrel (whic)i Ifist in the English market is an uncertain quantity;, it is by weight com- bined with fineness and testile strength when mixed with sand, that the relative economical value of different cements can bt obtained. TENSILE STRENGTH. Anomalous as it may scnin, Portland cement, except in rare cases, or under very exceptional circumstances, is not used neat, and yet here- tofore al! te.sts to a^ee^tain its strenju'lh have been made with blocks, specially prepareil from neat cement alone. Happily a change has taken place, and the German system now obtains of making the blocks (briquettes) out of a mixture of cement and (normal) sand, in the pro- ' Grant, " Proc. Inst. C.E., " Vol. 62. 6 '^4 portion ofono part of the furnior to tliroc pnrtH nfthe lutUir, thus approxini:tting tho i-n»uliini{ iiiit«i* to tho inortiir actually uiod un works. KogllNh uianurscturcritanil Erif^lUli cn^iticurH did not at first take kindly to thlH radical dop:irturu trinn th ' nld groovu,)iut i' Iism bocn aocvptoJ, and, it tniiy bu said, witii udvantu^o to tliu iimit. When (.'ciucnt ii luixed with sand Un strrntttli is ridiici'd, or, in otliur wurdii, nciit oouivnt is strongnr than ony niixluri' Miut can bo niiidn with Hand, and thur- forn, to comply with ii stikndani fixed an thu tunHilo Htreni^th of a oiix- tiiri* of oou of cenu-nt and three of >((n;irf iiieli in arcii, thin* facilitating; tho nionufuoturu and tCHiin^ of a ^'renter number of briquettON duiing u day, and the u*e of Mnaller unil mure eiiHily operated tentini; ii])paratuB and appliunocH. In the preparation of bri(|Uett('N, mII arc agreed thiit whether of noat eoment or of an admixture of crinent and Hand, only » certain percentage of wiitr, siiy 20 per eent. \t\ wi'it;lit for neat ecinont, and 10 per cent, for sand and cenient — is ri.'qnired ; and tlint the mixture Rhould bo prc< (o i>in|i|i njipliiil iil an even itnil rc^tiiitf Hpuud, and \w MtjiytyMn tliut tlii< stamlikrd s\»wi xiionld lii< till) Ibi. up- piiuj in in NocoiidM, iir it iittld hIowci', and cortuinly notttlowir thun 100 lbs. in 30 woonds, nn uiiiniiil n/ llif length o/' liiiif irkirh a teit would ornapjf, u» if the Huvin.; ol' a low scuondH of ti;ni^ in of uioro importunoe than the diituraiinut inn orilu' I'orriTt tonitiii' Mtroni;lh ol' tho ocmont undi'r cxaniiniition. Tho Ibllowiii); liiblc! Ims been ixtiaclcd I'roni Mr. Kiiija'H papor, for the purpo.su orNhowinu; ihiit tho upeiil witii which woii;ht in applied ii an important iuotor, iind that it is poMMiblc to obtain roHultx tVoni tho Haiue Huinple ot'cuincnt which ant totally at vurianuo with uaoh other and, it luny bo Nsid, aru not IruHtwurlhy. "Summary nt' Kosiilts ni' KxporinicntH to dotcrniino the differonoe obtained by appl^inj; the weight to tho briquette, when tcntinf? for TenRilo 8trenjj;th at different speeds : — Niiiiilier of liriqiirtleii. Sp^od ' Average UcHiill. 129 129 " T»5 lbs. furn, 1(10 in 1 100 " 15 iOO "U, 100 " ;io 1 luo •• :)o IflO " (III 10" " oil 100 " 120 lbs. .■)C0.75 .'■>n(!.4;) 452.20 4.'IO.!Ki 00 no 40 40 417.27 40:i.(l5 11(175 4()0.H7 From the foregoing results it will bo Heen that the increase per cent, duo to increased speed of 'ipplyinj; tho strain is a,s followH : — Taking the lowest speed of 100 lbs. in 120 .seconds oh a starting point, by applying the strain at tho rate of — 100 lbs. in 60 seconds, the inercuM is H.DfiOper cent. " •' :!0 " " 7.488 " " 15 " " I2.41ti •; a II J II i< 23 14'' " It is plain to sec from the above statement that it is possible to give a fictitious strength valuo to a cement, and at the same time to justify the mode by which it was obtained, hence tho speed with which weight should be applied should be a fixed factor. The question, what is the Htandard tensile stronuth of Portland cement in Canada ? remains unanswered, and each engineer, follow- ing the practice of his Knfilish brother and American cousin as well, is left free to fix, in his specifications, the minimum strengtii per square inch he requires. To show the want of uniformity on this point, tho following statements arc fakcii from " Proc. Inst. C'.E,, vol, 62, p. 216." It was highly desirable that cngiiiceis' and architects' specifica- tions should be more uniform The vagueness and the curious variety of divergence which characterised them (tho specifications) were remarkable and it was difficult to imagine that this state of thin<:s could be allowed to continue much longer, displaying as it did a backwardness on the part of tlione conecrned, in comparison with German engineers, who had been so much later in cntcrini; the field, and who, from being disciples, now appeared as teachers and examples. The evidence of this was furnished by publislied rules, adopted by the Society of Architects of Berlin, the Society of Builders of Berlin, the German Society for the manufacture of bricks, earthenware, lime and oemcDt, and the Society of German cement manufacturers. 8 i III 21 »p«ieiflaRtionii cxantinuil tliero wuro 13 varUuicit oI'IcnI fur flno- ncM, 10 vttrinticn for wi'inlit |wr htiwhol, ikiiil l'.\ v»riotio« for tenxilu bruskiii); Rlritin 7 ii(lit per nqiiaru inoh an the minimum tent utriiin, tlioro arc wivoral tliin^^H to bo uonHulurud iiud duturuiiDud, Til ; — 1. Whutlior thn tint HJiall Im! with nont peniont ; or 2, Willi A niixturi' ot'Haiid and{cuin(tiit ; if HO .'I. 'riio proportioiiH of Nimd to oenicnt to bo x to 1 ; ami 4. Tlic cciiifnt to be linn I'liou^'h to pa8M tlirough a Hiove of ;«; modboi por H(|iiaro inoli, li'iiv'iii'.;ii nwiduo not oxondiling .r pnr ci^nt by Wi'inhl; 5. Tlio Hand to \m of a Ntandurd i|uality (whioli nujuircH deUning), or to bo Nuub ni* is );onurully UMid ) 6. The quuntiiy of wuUsr to bo u«od in mixing to bo r por ount. of tbo weight of tbo uuiiiL'nt, or of thu uixturu of mnd and oouiunt; 7. Tint bii(|Ut'tto« to bu uiado of ii utaudard foiui (to bo dotinod) in a Htatiid tunipcriituro ; 8. The briquoltos to remain onu dny in air, oovorod with a dump cloth, and 7, or 28, or r dayH in water, ki-pt at .r, ° Fiilir. ; 0. Ill teHting, to uho — 'h nincliinu, and to ati|)ulatu: 10. Thnt the ittrain hhall bo applied nt the rato of x lbs. in x locondx; and 11. To fix the number ofbrlquettoH which muttbu brok«n to obtaio an average of thu strain oxortcd. So mueh for uniformity wliioh in not obtained in Canada, for there will bo m innny variations in tbo results obtained aa there are varia- tionH in tlio malerinlH employed, the miinnor adnptol in making the brifjuelles, and in the tostinf? mncliinos uHod,for no two of the latter using the nnmo make of briquette will i;ive thu same roHultis, and in fact very variable rcsultst are obtained from the .same miichino. As an illustriition of tliix, the following hii.s been taken from a pamphlet''' on Portland Cement. COMPARATIVE TESTS OF THE SAME CEMENT. j Tensilf Streiiglli per square incli, obtained by A. Southiim- H . I'aijii. Ooriluii &'Co. Kiistwooil (1) lianUvood Uibbs. lbs. ■290 lbs. *.'!8!) lbs. lbs. + im lbs. T 57(1 lbs. 625 • llrokoii at till! rati) e retained on a thirty _^mesh sieve. If dirty, it should be thoroughly washed before using, and if soft and fine it should be rejected. 3. MIXING. Where a solid and impervious mass of concrete is required, the com- ponents selected must be mixed in certain proportions. Specifications generally read that the concrete required shall consist ofso many parts of broken stone, or gravel, to one part of cement, or so many parts of broken stone, or gravel, and so many parts of sand to one part of cement. In the.se the word " parts " is an indefinite term, and may mean a cubic foot, or a bushel, or a fixed weight per unit, or even a barrel Or a wheel-barrow load. It is asserted that in proportioning the amount of cement to stone, a given weight rather than a given bulk should be used, bulk being affected by the weight and degree of fineness ns a cement of eighty lbs. per cubic foot will b< larger in bulk than one of ninety lbs., but the system of using a cubic foot as tlie unit of measure is certainly the most convenient and gives equally as good results. Assuming a cement weighing 8.') lbs. per cubic foot, and a mixture of broken stone and sand weighing 283.'> lbs. per cubic yard, or 27 cubic feet, then ; — Nmiilier of (Miliie feet of Will be equal to Cement. l Klone. Uy liuik. I to 9 1 toH 1 to 7 1 toG I to 5 1 to 4 Hy Weight. :i.oo .•1.H7 4.50 6.40 6.7.5 27.0 J to 11.12 1 to 9.89 1 to 8.04 1 to 7.41 1 to 015 1 to 4.94 11 For solid concrete, there arc two proportions In the material* u»ed which should he determined : — (1) The voids in the gravel or stone, and the quantity of sand to fill theni ; and (2) The voids in the sand, and the amount of cement to fill them. Taking stone broken into pieces which will pass through a two and a half inch ring, and be stopped by one 2 inches in diameter, the Bolidi will average about 5'2 per cent., and the voids 48 per cent. The voids in dvj sharp silicious sand will average about 40 per cent.; but where the sand is compressed in water, its volume can be reduced 12^ per cent. When dry sand and cement are mixed together, contraction takes place, and in one of sand iuid one of cement it iimounts to 16.66 per cent, of the original volume, in two tounc the contraction is 16 per cent., and in three to one 13.81 per cent. When mixed with water, further contraction to the extent ot lU pur cent, takes place, and, according to Sandeman, " Proc. Ins^t. U. E.," Vol. 54, p. 251, the ratio of contt-ao- tion of dry materials is 34.43 per cent. Adopting this percentage, to produce one cubic foot of set concrete would require 1.525 cubic feet of dry materials, and one cubic yard will require 27 x 1.525=41.176 cubic feet. Assuming that an impermeable Portland cement ooncrole is required, experience has shown that such can be made by using one part of cement, two and one half parts of ele;in, sharp, silicious sand, and five and one-half parts of sound compact stone, broken into cubes averaging from two to two and one-half inches i-quare. The quantities of each material required to make one eubicyard of concrete set in place can be determined as follows ; Apart will be equal to 27 -f- 5.5=4,900 cubic feet. Cement = 4.909 X 1 Sand = " X 2.5 Stone = '• x 5.5 Total = 4.91 cubic feet. = 12.27 " " = 27.00 " " 44.18 The contraction or shrinkage will therefore be = 44.18 — 27.00, or 17.18 cubic feet = 38.88 per cent., and the results will be as 61.12 to 100, or it will take 1.630 cubic yards of materials to make one cubic yard of concrete set in place. The composition of this assumed "cubic yard of concrete " is as follows : — 4.91 cu. ft. of cement x 85 lbs = 417.35 lbs. = 10.72 per cent. 12.27 " sand X 85 " = 1042.95 ' = 26.80 " 27.00 " stone x 90 " = 2430.00 " =.62.48 " or, it is composed of 10.72 + 26.80 =37.52 per cent, of mortar and 62.48 per cent, of stone. That the amount of materials provided for mortar is sufficient to fill the voids in the stone is determined thus : — c. ft. c. ft. c, ft. Voids in the stone = 27 X 0.48 = 12.96 4.91 of cement and 12.27 of sand = 17.18 Shrinkage on admixture dry 17.18 x 0.85 = 14.60 Further shrinkage or admixture wet 14.60 X 0.90 Excess of mortar 12 = 18.14 = 0.18 Or the excess of uioitar may be dcteiniined in another manner:— 0. ft. Total quantity of cement — dry measure = 4,91 Rcduoiion by wetting = 15 per cent. = 0,74 Net quantity Total quantity of sand — dry measure = 12.27 Reduction lor voids = 25 per cent. = 3.07 Not quantity Total net quantity of mortar Voids in stone = 27.00 X 48 percent. Excess of mortar eft. 417 9.20 = 13.37 = 12,96 = 0.41 As.«iuming that 5,000 cubic yards of concrete of the foregoing pro- portions are required, then the quantities of the componente to be pro- cured will be as follows— a barrel bein;: taken to contain 370 lbs. of cement : lbs. Cemrnt. 417.35 x 5000) Add for waste, etc., 2J per cent. lbs. 2,080,750 52,160 Total = 2,138,900 = 5,780 bbla. c. ft. c. ft. Sand. 12.27 x 5000 = 01,350 Add for waste, etc., 5 per cent. = 3,067 Total = 64,417 = 3,0G8 c,y. Stone. 27,00 X 5000 = 13.'>,000 Add for waste, etc., 5 per cent. = 6 750 Total :r=.- 141,7.50 = 5,J50c.y. Given good materials and poor mixing, the result will be poor con- crcti', and therofore too much care cannot be taken with this operation, whrther it h: done by hand or machine. Poor results are due to : (1) improperly frauginii the materials; or (2) to an excess of water ; or (3) an insufficiency of bibnur in mixing the maU'rials, either dry or wet. or both ; or (4) the want of a proper mi.xing platform. For the mannlaciurt! of concrole by hand, a platform is absolntely necessary, for concrete sh