«» » » N^ IMAGE EVALUATION TEST TARGET (MT-3) • '-'■' ■%.' ■ '-'H' U\ - .■"\ » 1 % ■ '> !•■•-• r:V'-|':V -.^^H- '■'%') ^ ''l'\-r:\ ■ ';!;*, ' ■';.■ ' •■ ■• ■" ■" .■"■■.* ■■• '' -.■', . ''■>'. "■ :■"■ i,"V *■ .'■;; '';. "■ '• . ■- ■-■•- ■'" ■' '..-''" :■■¥ -. ... . , ■.: ,'■ ;...- , ,* > ■' ■.'."• •'• ■ ■■ -■ " ,,■■■ ■ ■. IGMH i \ a: > Microfiche j ; Coiiection de ■^ ■ . Series microfiches • (IMonograplis) (monograph ies) ; \ » ,'"."■ » • Canadian Inatituta for Hiatorical Microraproductiona / Inatitut Canadian da microraproductiona hiatoriquaa r 1 • r 1 :i: ^r,t:.,;.-t ■: ■; ■z-;? i - i^ ». ■i^'-'V .*. V Tachfiical wtd Bibliographic Notn / Notin tNhniquM M bibliographiqiMs TIm ImtituM hM atMinptad to obtain lh« bait original copy availabia for filming. Faaturat of this copy which may ba biMiographically uniqifa. ««hid« may altar any of tha imagat in ttia raproduction. or which may significantly changa tha usual mathod of filming, ara chadcad balow. ;« '" Y ■' . ■ ■ □ Coloured covers/ Couvartura da coulaur \ D U Covars damagad/ Couvartura andommagia Covars rastorad and/or laminatad/ itaur4i^t/o( Couvartura rastaurto. at/ou palliculAa □ Covar titia missing/ La titra da couvartura manqua a Coloured maps/ Cartas gtegraphiquas an coulaur Coloured ink (i.e. other thgn blue or Mack)/ I I Encre de couleur (i.a. autre que Meue ou noire) □ Coloured platas and/or illustrations/ Planches et/ou illustrations en couleur Bound with other material/ Relii avec d'autres documents .,/ I L'Institut a microf Hmi l» meilleur exemplaire qti'il lui a M possible da sa procurer. Les details de eet exemplaire qui sont peut4tre uniques^ dii point de vue bibliographique, qui peuvent modifier urw image reproduite, ou qui peuvent exiger une modification " dans la mithoda normale de fihnaga sont indiquis ci-dassous. | . . □ Coloured pages/ PagM de couleur □ Pages danuipM/ Pages endpmmagaes □ Pages restored and/or laminated/ Pages restaurias et/ou palliculAes Pages discoloured, stained or foxed/ PagM dteolorfas, tacheties ou piquies □ Pagis detached/ Pages d«tach«es 0Showthrough/ Transparence n Quality of print varies/ ^ "^ Qualiti in4gale de I'impression tott Th« oft! □ Continuous pagirkption/ v^ Pagination continue Orifl tfM-l tion firtt ■ion or HI Tho thol TIMI difffi ontii riglii roqu moti I ~| Tight binding may causa shadows or distortion D D along interior margin/ La reliure serrte peut causer de I'ombre ou de la distorsion le long de la marge intiriaure Blank leaves added during restoration may appeiir withifi the tfext. Whenever possible, these hav* taken omitted f roiti f ilmkig/ II sa peut flue cfrtaines pages blanches ajoutlas lors d'uml restauration apparaissent dans le texte, meis, iorsque cela *tait possible, ces pages n'ont pas^ti filmtes. Additioni^ comments:/ Commentaires supplimentaires: Thii iteiyi is f ilij^ed at the reduction ratio checked belbw/ Ce document eit film* au taux de reduction indiqui ci-dessous. nmr 14X i iwr 4. 1CX 20X D D Includes index(es)/ Comprend un (des) index Title on header taken from:/ Le titre de I'en-tlte provient: Title page of issue/ Page de titre de la livraison t~~\ Caption of issue/ Titre de depart de la livraison □ Masthead/ Q4nirique (piriodiques) de la livraison % "HJT -7ST "snr 24X ax 32X II t I VIM IMI " t Th« eopv ftlmtd hf hM bMn rtprodueMl thonkt to th* ■WMrotHy of : LftraryofthtNationil A^citiv** *f Canada • \ ' ' • - r • ; .", ■' Tho Imoflit oppoorin^ horo oro tho ^mt quolltv' MiMiblo eontldoHn« tho eontfitlon and loflihiUty of tho oritflnol eopy and ifiliooping filmlno eontraet fpoeifteotidfM. L'oximpliiro filmA fut roprodult grico A la gAnlirooijtA do: La MMiotMqua das Archim natioiiaiai du Canada Lot ifno|boa suivantot ont itA ropreduitoa *av«c lo plus grdnd ooln. compto tonu'do lo condition ot do lo nottotA do I'oxomploiro filihA. ot On - oonf on^itA ovoc lot conditiohi du controt do fHmof -T Orlflinol c o p lo o In printod popor eovoro oro fNmod tho loot P090 wHh printod or Muotratod improo- olon. or tho book eovor whon opproprlato. All othor orlglnol eoploo oro fMinod bog l nni n g on tho firot pogo wMh a printod or Muotratod Improa- •ion. ond onding on tho loot pogo with printod or HhMtrolodlmproooion. Tho laat roeordod fromo on ooch mlcrofleho •hoN eentoin tho iymbol ^»> Imooning "CON* TINUffO"!. or tho symbol ▼ Imooning "END"). whiohovor opplioo« lAopo. plotoo. ehorta. ate., moy bo fNoMd ot . difforont roduetion rotloo. Thooo too lorgo to bA ontlrolv Ineludod in ono OKpoouro oro fHmod boglnriing in tho uppor loft bond eomor. loft to right ond top to bottom, os many fromos os roqulrod. Tho following dlogroms llluAtroto tho mothod: . ■ ■ * I .. u -* ■ ■ ■ ■ S^ \ 'W'- 1 2 1^ ..3'-; .■ ■ ' ■ porlo dir plot. plolroo originouii doht lo eouvorturo on joot ImprimAo sont filmAs on ebmmoncsnt imior plot ot on torminont sqit por la pogo qui eompdrto uno omprointo -^^ in ou d1llustr4tion. soit por lo second in lo ooo.Toui loo outros oxompioiros' sont MmAs on common^ont por lo pogo q^ui eompofito uno omprointo Ion ou dlNuotrotlon ot on forminont por ^pogo qui eompprto uno tollo nto. Un dos symboloo suivonts apporoltro sur lo domiAro imAgo do choquo mieroficho. colon lo cos: lo symboio -^ signifio "A 8UIVRE". lo symbolo ▼ signifio "FIN". dimpi lo U irtos. plohchos. toMooux. otc. potivont Atro filmAs A dos toux do rAduotlbn diff Aronts. Lorsquo lo document oot trop grond poiir Atro roproduit on un soul oMchA. 11 oot filmA/A pqr^i*^ do I'onglo supAriour goueho. do gouchA A droito. ot-do hout on bos. on pronont lo nombro d'imogos nAcossoiro. Loo diogrommos suivsnts INustront lo mAthodo. J3 32 X ,• i" 1 2 A / T ^J> 6 '(''V.i-^ • ■">>., / \ •/ lanaiiaq j|JofW}j of ^ipil ^ngl.ie«PS. ' IJJCbRPOK^TED 1887.. ADVAPICC P^OOF- (Svb,eet to reti»hn). ^Thii Society, an a body, does not hoU itwif wponnible for ti.e (a. U •Od opinions iUteJ in »ny of its public^ioBH. . ID FILTRATION OP PUBLIC WATER SUPPUKS. Bt R. jB. LcA, AiSBoo, M. CAit. Soo. C. B. (To Be read Thutida^, Jan, 19 few decade.; but which Jm, neverthele«, P" portant branch of cn}iinj.ring. ^ „,trictcd arcn., in doae From thia-crowdlns fgether of people mw^^^^ p„,-.„it, toind«.tn|c«..bU*n.en^.^^^^ ,. UpUcated problem- |an.ou, the n.«^tm^^^^^^ ^^ ^^,^^„,,, ^ -.uoKm^tte" P^P'f^^'jlJ J'*^^ neighhbourbood of citie. In thickly popnla&d diHrict. and in i » ^ ^ ^^^ ^ ^„_ „d towns the wa4 of ^--^^^^.^^i^r lUta7. method of tional. menace to tl^i Health of *« '"^•^ i, „tiii,iog animal pre^^rving the ba^co between gro-Jh^^^^^^ ^^^ ^,,,^ Uc«pU»tfoZi.no ongcr^^^^^^^^ ^^^^ .„^p,y .„a begin to i«rvcn| ^^ ^n^Ti. e-dent that among the moat receptacle, for *••»»• ?*T \i- mnnicinal * engineer m«y b»ve -^^'^^^dbeT^^ to deal are thoJ connected with tn ^^^ ^^ ^ degree of pnri| nepemry i- ^^^.^^nih. acco«pli.b«e«t of thi. The proper Lthoda to he e"P^«^»» ** ^^eal Fi««P»-. object depen^ much upon bo^^^^^ jpj^ .„d with ,o\b.t.eert|indegrwoffcm.har.tym ^^^^^ ^^ ^ ^^"^'X^:^^ror:>^ -^^^-^-^ enioneer en}*ge«f »n ■'»'"' 7 1 . . „^ iloB loth&oonoection: ..^ _^ _., ^ ^^ ..■■..■ / ■ 1 1 ti ti d •< ot ai -.f ■..,.„■ . **«.■. r^ \^ .«I ^Th^^rd, t\tom Its^foundation in 1869, alwayn devoted • ^roat *»! of ^tention to the condition of the water sappliea of the Sute. Ill 1886, Ihe time being partioaliirly »ppropri«ie, it appointed a body of(expeVt»tot^oexcIuMTediityofconduciins a aeries of obserratinoii and e«periii|t5irb..with the object of finding the best mcth*Hl!i for puri. fying both ttatorand lewage. These esperinientH are atill'in pro- greis, »nd the arnraal roporta of the departnitnt, giving the reaults of theirinir^'»tig;ition8,arefl5Aeodingly valuable to eos'ineen and other*" interartefjl.itf SDch qiuxtioaa. * In %t*rtn ^lid in a.fcw other large European cities h!ivin!» water Works dc*partiainta provided with the necessary soicntifid cquipmdnl •nd manageDient, many cartful eipeiiments have been mude oii tJic working of the large water-filter bods of the systeoiH. The UultH of BUoh experiments as these have, an especial value from the fact that they are conducted on a large wale, and under conditinnM which exist in actual practice, ^n the other hapd. these same circumstances render them.ioss reliable as a mean« of determining the true principles npon whroh the procow of filtration •depends. The object of this paper is to describe, as fully as reasonable limits will permu, first, the circumstances Unc(er which water iiupplies become polluted, and the nature of thU pollution ; aiid second, the process of purifymx It agiiin in larsro quantities by sand filtration. ' Of ^urse pure water is profcrab^ to purified water, or, as has been Mid, irith water "innocence is better than repentance." Unlbrtu- nately, however, water whoee natural stete is above suspicion is often exceedingly diiBoiilt to procure, except at a cost which is practically prohijMtive. Consequently, many oities aad towns, especially the larger^" ^n^are forced to os^ sucH water* w maybe praotioidly available, ^d to make the beat of them. But >hie best is by no kueaDS to be heia lightly. By the metlKxla to be described later it is possible to ■0 oliange the nature and oliaraoteristics of poUuted water is to oonvett It to the appearance, toato, and probably absolute wholesp^eoeaa of Che most innocent of mountain torrenta* / Water has the unfortunate capacity of re«d!ly disMUving many of the subatanoea with which it may come in oontaot; so/that outside of the laboratory, chenioally pure water is praoticaUy inknown. Some of these foreign elemento may not only be quite h^mleaa, but may •etuaUy improve the quality of the water. It ia, ^wover, with the oUjow, whioh make ihe water contebiog thorn VnaighUy in appear- nMf diiigraeible to taMe or wnell,.or dange^us to health,>.in 8 r^ P I-.:., 1 1- ^ : « . \ -x vl * t> V -.,v~ .A •rt lUble to receive more or 1^ ■»""»" i / / ,, T^h7, mav b. colored by the ^^^^'"'^'^^ o*« 2. Thewtlr. of mnnj stream!, bwme.tarbid wiUi J «,pondcd m.tter«»rter ^^'^ ~'^'; .V ^^.^^r. art UabVe. ab r 3 Tbe w.t«n.of lakes, poods, sod »^*0» "T X]ff.^ and other oerta^n seasonsof the year •« -^-^:t|^';!i;r;^^^ eye. htt- . minute water plant, whioh ^o^^ ^^^ ^^ diaagrewibVe iMtea .•« irhieh are capable of imparting ^ the water fliaagre- oioof^ , , .„rf^ ..tera *W haw disohanied ^olo %, Anyoftheaecla«««of.urfawwale^^^^ them a greater or loss quantity of ham«i» sewag^ . icaa -». ■ oircumsuoces, totery ^^^^^'^'T^^'^r' r^, Jpiy, the proper In determining the quality of a ,giTea ^«ter ^py. method of prooedure is as follows :-- ( JP^ ^^^^^ 1 To make a local examination of the ^ -h^l, p«,bable «>oroes of P^^^'^**-","''' JLJI.'TJYK^ made of -mpl- of 8. To make ^biological examination gn^B «,, ^,i,^of assla- poMiUe presence of organisms, wnioB w im dement of daiiRer. ^ L.Waea. It'maf *• «t»«^ i» 1 ' .'-''■^(' • / . 1 1 K , X • / ■ I . ^ ■ re i ■'••■«. \ III i .>A. V ^. i- 'V i tbj* Of rftny ■■\ to aUade f>rie% H^ th« oonititdtion of oi]granie noAtter and to the ehange* it is liable to andergo. * To b^ia wilih, it inclades M those eombiititions of KhjB ohcmioul ele- meodi) 4rhose /or Illation depeodji upon the prooeases of life ; and which, ih«refore,,oeeor either in plaoU or animals. Its history is cjclicil, oonsisting of a eon$trueti»0 phase or period of ;innowlly^nd a deatructive phase or period of decay ; the death of the plant or animal. far oiin^ the dividing lino between tlie two phased. The cycle begins by the appro- priation of inert, purely mineral siibatanocs from the earth by tbo green plants, which derive the necessary energy from the sunlight ; and ends with the complete disintegration of the more or lei^ complez structures which oonstituted its oi]ganio charaoter, and the return of the elements to the earthi .^ With regard to the nature of the change^ it osay haiv undergone, it is only with those in the seoond or destructive phaso that wo are eoneemcd. At the bejiinniagjof .this phase, at tjw death uftiio plant or aninial, we find that sli organic matter is composed mainly pf jtarbon, oxygen, hydrogt n, ai\tep is the reduction of the nitrous acid to niirie aoid. The nitrc^us ani nitrip aoids dd not rcm.iin free but eombine with come base present, as soda or potash, to form nitrites »bd nitrates, the latter being piirely mineral substances ; so that iho final results of the dceompositionl process are oarbonio noid, water, and ni- trates. Thus the dead inor^nio materials needed for the formation of organic atrueton'g are only borrowed ; and ultimately are returned to the earth again as inert as when they were taken froui it. Returning now to the cheniical analysis, wo find the results siiven in lome suoh form as the fnllowiW which is the one used by the Massa. ohuMtto Sute Board of Health:— * ♦ Hi. flrtU i> thk Ubto iiMMM tmm p«r lOojm. Vkm Mtow^tS^Ml^Mr^ for tlw ■-- ,i »v- '' l^ow it hu been ibuod thai •▼ery aocurate, and at the aame time oomparatiTely easy method of determioing the organie matter in water by a chemieal analysia is to determine the amoant and Condition of the nitrogen present Thus, under the head of Attttiminoid Amnionta, are entered ■monntA which are proportional to that part of ihe nitro- gen which is derived from fMi organic matter, ». e., firom organio ^ X matter which haa not yet begun to deoompoM. Theie oolumna, there- fore, represent the posflibilitien of putrefaction still ekistinK in the water. The amounts unt'or Free Ammonia represent decay began; un- der Nitrons Aoidi (or Nitrites) doeay still farther advaneed ■; while under Nit^tes the amounts entered represent the nitrogen derived ftom that portion of the original organic matter whieh has pa<«ed through all the stagfes of decay, and which has been converted into purely mi^- - oral matter i^in. ^ • The importance of the determination of the ohlorine is, ihat an ex*. oensive ¥inoUnt points to oonUmination by sewage which always con- tains a considerable proportion of coiumon salt. The actual amounts of the different substinees as they oeour in water supplies are ejoeediUgly minute, as will be seen by referring to the above table of analysis, one of which # samples (Q) is a highly polluted one. Henoe, in th^mselvea these subsUnoes are of very little ♦^ iiiiporUnoe. It is. in the history pf the water which their presence in- ;p^ dicates that their aignifloanoe lioi. "pius the chemieal analysis can ^ • toll w not only what is in ihe Water, but also a gnat deal about what is going on in it. It is only within recent years, however, that tha mcthodB of organie analysis have been capable of producing such r«- 6 Aver«K« aur* fMc wftUr. Private well. Mystie'Lakc. he MOM time latter in water oodition of the lid Annioniai 1 of ihe nitro* from orgtoio ' wlamns, there- i^ io the w^ter. f begoo; ua- ranoed ■; while Bo derived from pa<«ed through ito porelj mi^> is, that an ez-. oh idways eon* theyoeonr in . by referrtng to (Q) is a highly re of very little bir presenoe in. lal analysis ean ioal aboat what rerer. that the lacing meh !•• ■•1^.:: . \ ■alts; when the flnt aitempti at water pbrificaVion were made, Tery little was .known of the organic mattekr in solution, and the object aimed at was simply the dari/ieation |cif the #ater, or the re* moval of suspended matter Tisible to ttiecye. This was the condition of things wh^o James \ Simpson, in 1839, oon- structed a sand filtration plant for one of the London water eoinpaniee. Baeh of the beds of this system oonsiatod of a bWad shallow basin or re8enn4r with water-tight bottom and sides. The depth #as about 12 feet, and it was filled to about half this depth with tlie fllterin.; ma- terial, whieh consisted of uniform layers of small stones, gravel and sand, the stones oo the bottom and the finest sand on the top. Through the bottom layer of stones and gmvel extended a number of branch drains leading into a brgci^ central drain which wan connected to the outlet (Fig. 2). The inlet to the filter bed opened above the surface of the sand, and both it and the outlet were provided with gates. The process of filtering consisted in fiooding this bed of sand and gravel, and drawing off the water from beneath by jmeans of the nystem of underdrains, which were built with open joints. The rate could be r^ulated by tlie gates or other apparatus on the inlet and outlet pipes. As filtisation prognssed the Burlaeeof the Nind became gradually choked up by the formation upon it of a layer composed of material removed from the water. When this layer became so impervious as to prevent the water passing in sufficient quantities, the filter was stopped, the water level drawn down below the surface of the bed, and the deposit layer reniMnd, .together with from } to 1 inch of sand. Whep the snrfaoe n^raptoothed and levelled, the bed was ready to be put in action nffSii. The fceqneney of the scrapings depended upon the condition of the water and the rate at which it was filtered ; and when the sand layer . bad become reduced in thickness to what was iwnsidvted a proper minimum, the whole amount removed was retraced at one time, either by new sand, or b^ the ser^pings after they bud been tlioroughjly washed. • The results from the use of dieae filters^ were lo satisfactory acoord* ing to the ideas of purified water then io vogue, that in the following years peveral others were built.in Bufi^nd, and a little Uteir on the continent, espeoially In Oermany. Some of the most important of the continental filters built during this period were designed by the Bng« / Ush engineers Gill and Lindley. They were all built oo the same general lines as the Simpson filter deseribed above, the details vsrylng ■omewlmt With the ii^dividnal notions of the designers. J •fflPW»W"P!|g!P ■-f . "1 ■■%- % ( "■' . )S "Oer dedgn. A I«|^ earlier, in I8T0, the Engliah ehemiat. Wanklyn and PVnnIr knd .-.c^ted new and in.pra,ed method, of otganie ^^L'^^X ' . --■ o ■' ■ * 1. ' # ^( ,.' ' \. ■'. M' ■ ♦• '. ' * ." ■' /; ■-•----;: — . - > 1 ■ ■ • « fl .- / • .# » ■ 3> ■ -.»■■.' ■ "■(- ■ ^-n/"-" 1 •■• ^^ ^ Mi s' ■ . ,iJ«U t^iijyuL^^tjSSMK HI iigh*' llMDroftai 1^ ■■■■■■li J4- 1 .r» attor. The nil to report leludod the ^ Baropean t of Earope. i> were oon- lB92i only I N. Y., in ^er dedgna Md I^nk*- Ijsw whieh . ' » .f . "^ *"•" ™'^ ooooeming the prooeMoe offer- menution ^d decomposition. It wm -uppo^'hu d^uld 4 ^Z7^ ^f'-^-^JonS-nio .-tier by contact with otJr 6rg n^ ^r?.l^^ Ji'" ^'«>''«'o held that ordinary alcoholic LZ;„Tr'^^°n? ''^ '*'"^'*^ ••"> aecayinJeaH ceil.. m«eado^ by the action of the HHtTg .nd groyioK clN L we know M ■" • :•"«''•"»««<» ^' Ac pi^nce ofciccoBpoaing ible or animal jiikttAP in ..*.. tA ^^ •. . .': 9 «MM.».ki. «. • • —"•«• ««••' »ao preaonce at uccompos fm;'"!.'^ ^ "!'•** •*"'**• '~™ *^« -■«» «••«« "bowed ^irir . nyde^to Mnctlo. of the org^ic matter-^don, a, mnoh .. M • |«r«nt-thcre,oltwaa eonWdewd wry diaappointinff, and a, indi- -..-^that iW, method of .Itratlon. wh.le oa^We oTimpnoW^ 'Z •pgaranco «„d taatc of the water. w«a of alighrhygiooio value ' Not many yeaf. later. Lowerer, them, idea, and theorie. wero^broken ^ S '!"• ~»"'?^*'- °f P"*«»r. Who demdn.tr.ted t. at the p«>. ^ITn^nrT"*'*'"" '"** putrefnction wore depend. nt upon the •p^bloofc.««ngd.««e. A newTiew wa. ^ow fc.lcen of organic ^^S;* . '• T**' ?••-«««-«•« the pnAable preaenoe of g*™/ • ^ IJ'*! .» -"7' r'»y ''- '-w deemed of much leas import- Moe than b.«l«ery by Dr. Robert Koch of hi. ■Z'Sttl"^"'^''^^. "''•"«"' "•'■«'« ^^ eKt^memlmT nl2r T '^ •"•* *• •""'"""■ '*** " •'»'<'•» thor.incroa*.d in nnmber under oircum^ance. (ktorable to their qrowth. it wi F«g«>«.« the knowledge of the mibjec*. But wUh t|.e .drent of ^i^olIZ"" "^^"^^^^ ''•^ *o a g„^t c£teafrc..^ed; :L K.^Jfr;'"'* *° <>«««n»ineth.nmnbcr^fger«.,to.tad; 4«ir habit, of hfe. function, rto.. „d to cU«fy them h^to ^^^^ h ,am«oerwh.cb, c^n«d«riogtii«ki«d o« on»i«« d«.lt i^mZ qaite mtrvelluun. "^ . ^ ^^ Bcide. placing the germ theory of dimM on . fcrn M» thiii A ■;^ ■' M^ -t ^i ^ 'f ■v.- ■ i / .■^^^ , 's,|i.>*^«i • discovery of Kooh's marks the beginniog of the /period daring wbioh it has beep possible to deal with the subjects tof the parifieation of water and sewage io a rational and soientifio way. Numerous ihycs- tigators at onoe began the study of these qu^tions under the new and vastly improved oirpumstances. Inasmuol^' as the results of many of those experiments have a direct bearing ^pon the subject' under consideration, a brief description of the n^ur^ and some of the char- acteristics ofthe bacteria will be given before-i|>roceeding further. BAOTSaiA. They belonrf to the lowest and smallest foLn of life. Structuimlly '^ they are composed of a single cell with a iall; possibly of .ocUulose, and oonteots eonsisting of apparently structureless protoplasm and a nucleus j and are thus comparable ^o the bone, blood, nerve cells, etc., which represent the ultimate/ structural composition of - the animal body. ' / They are of such cftrcmo minuteness/as tolw vuible only to high powers of the microsonpe. In their greatest dimensions they vary from ^^ to 2 micro millimetres (from *UW to riiW of an inch). A sphere Js of an inoh^in diameter iould contain more than 600 1^"* •*''''**'' larger ones, and it Wns they yary of an inch), re than 600 sand of them, at it is little soorcely sua- iin9d for par- Bed as plants ^ are plants. > be by means by fissiun, — itself in the ) this mnlti- prise a great aircmcnts for Tal term boo- B their form, ion, eto. 2..Th(e6aeiKtVor rod shaped. . : - 3. The apiriUa, or spiral shaped. ' '» They occur usaaUy as separate individuals, but may also occur in pairs, tetrads, or in a row like a chain. Besides these forms they aw often found under oertaip' circumstances ,in irregular groups or masses, held together by a transparent glutinous material which they secrete through their cell walls. These stifeky; jeUy-like masses are termed aooglcea. Unlike the algae and other green plants they cannot exist upon purely inorganic matter, but require il^r. their nourishment matter already di^niied in s^ form. Moisture is also a necessity to their proper growth ; and thus according to their habitat or preferred environment they are classified as : ' 1. Saprophytt$, living on dead^ animal or vegetable matter or on water containing the^e in solution. ' 2. PanwiVe., subsisting on a livitfg host, in the body of which they grow and muluply. in some cases without any injurious effect, but in others causing disease and death. It is not known whether these re- suits are produced by tboir action in obtaining their food or by the products which are thus set free. Tliese injurious members df the ' parasmc cla» are the so-called pathogenic or disease-producine ' bacteria; such as the well-known germs of typhoid and cholera^ Some species of bacteria are able to exist either as saprophytes or parasites, and aro called /acu/to deoompositioa is evident that beneficent one. organic growth t the pftraaitie 1 the otiiers, are rlien tlie condi- oonditioni will irilj cause their to the typhoid iobiut ; yet, it and then oon* and virnlenoe^ I sewage is the in such germs ; a cYideitly the lard for water the siind filter n the chemical titutoQf Civil <: "Filtration who have not Qcee diawlved in pntresocnt irciist through tliis purpose." soon became lally removing i water. Not >e then have ind operating >vo from 99 to her dtfvioes for been shown to these earlier resolts whgae Investigation into the manner in which it was possible for^oom- ' p.rat.veiyporomi„aterianike the «i„d bed to hold back .uch\in- .flnncaimal bodies M the bacteria revealed a paradoxical condition \of affaire, v,i, that these germs, while constituting the most danseroig dement m the pollution of drinkin. water, were^at the same til the\ ehef .gents ,„ .ts purification, ft was found that the purifyini action was partly mechanical and parUy biological, the circumstanoi attending the ktter not being very well understood. The manner in which It t«kes pla^ and the means by which it mav be enhanced, will be referred^to while describing the construction and operaticft. of* modem filtration plant. i^ «mwu oi » Indercribing the material of the bed and the best method of dhi. posing them, we shall begin with the aand. aaifs. I. t 'iL" '^ T'* ^T"^^' ^ •^"•* purification take, place ; and it IS observed that the efficiency and economy of the process are de- pendenC to a comiiderable extent upon the siie of the sand grains Orspbloal r«pr«Mnta« f» in«ohpk«l Mriyri, «, . „ayj, ^ ,^ M I I t ! i 's .: •I i and the thiokn( n of the bed. It U the smaller grains which determine ' the " effective sise " of a sample of eand ; since, by filling 'up the spaces between jthe larger ones they fix the diameter and length of the channels throagn which the water most pass. Jyijawrenoe. "as the result of ezperlnientin!; on the rate at which Water flows through Tarious sices of sanA^ the " eflisctiTe sice " is ttiken as that of the gratn which has^lO pel^ent. by weight of the sample smaller than itself, and 90 per cent, larger. This sise is obtained by a process ofmechanicol analysis described in the Report for 1892, which also gives what is termed the " uniformity co-efficient," the latter being the ratio of the size of. the grain which has- dO per cent. Biualler thitn itself to the <* efTective sis^." (Sec Fig. ) If We look more closely into the pnrifyi|^ action 6f the sand, in order to be able to .understand just how it [is affeotcd by difference in the " effective sisc," " uniformity co-effici^tnt," thickness of the bed, etc., we shall see that what takes place is as follows :— When water is first let in to the filter, it rises to a depth of 3 or 4 feet above the sarfaoe of the bed ; and it is either held there for some honors, or filtration is allowed to proceed at once, the first part of the effluent being wasted. In either case, the sand grains at the surface sodh''be^ojne enveloped in a membranous film composed partly of^the zoogloea form of the bacteria, and partly of the more or less finely divided organic matter which the water holds in saspension. This sticky jelly-tike substance, extending around and between the sand grains, entangles and holds back the smallest particles in the water, even the bacteria themselves. The latter are not only prevented from moving farther, but are detained under sufah adverse circumstances as to not only arrest their growth and multiplication, but also to cause their death. Naturally the larger suspended particles, water animalonlie, frag* ments of plants, etc., ore stopped at the very surface of the sand^ and a continuous mantle called by the Gkrmans the SchmiUzdecke is soon , ibrmed and covers the whole bed. Under certain circumstances, as for example when the water contains a Urg^ algae growth, this layer forms a dark gfeenish carpet of a texture like felt, which when dry can be peeled off in flakes.* Ordinarily, however, since it contains a certain quantity of siliy matter, it penetrates the sand for a depth of half an inch or so. But even when formed in thii way there is often almost a distind plane of oleavage- between it and the sand bek>w, which makes it very easy to remove with broad Bquare-oon\ered shovels. i which determine >y filling 'up the ind length of the he rate at which efiective site" is \)y weight of the St, This sise is in the Report for aity co-efficient," hich has- dO per Fig.) If We look order to be able n the " effective etc., we shall see a depth of 3 or 4 Id there for some i first part of the ns at the surface )sed partly of'tbe >re or less finely ispension. This etween the sand les in the water, y prevented from circumstances as lUt also to cause animaloulas, frag* of the sand, and nutzdeeke is soon iiroumstanoes, as growth, this layer which when dry nee it contains a ,nd for a depth of ly there is often the aand below, -oon^ered shovels. -V,..-.V^. : This operttion becomes necessary when the gradual thickening 6f the surface layer prevents the required quantity of water from passingf It will thus be se^n that the surface film forms by. itself an cxoecd- • ingly effective filtering material, but with'* very delicate structure; and as such, should be carefutly guarded against any influence which might cause its fracture'. Several European engineers have concluJed that it alone constitutes the actual filtering medium ; and that the remainder of the sand l»ed serves merely f6r its support, and for steadying thfe flow of the water through the bed. But experiments made at the Lawrence SUtion do not by any means verify this view. Indeed tbeyliave shown that if great care is taken not to disturb the underiyingsan^, almost the whole of the surface layer may be removed without at all affecting the bacterial character of the effluent. It is also shown th.jt a new filter does not arrive at what is called its •« fuH, bacterial efficiency" until it has been in use for a consjderabic time ; even though in the meantime surface layers may have been formed of. sufficient thickness as to completely clog, tlie filters. . By examining the Hand, it has been found that this sub-surface puri- fication only occurs when the sand grains for a considerable depth below the surface haye become coated with a film of the gcktinous organic material referred to above It has also been shown that if, durinii scraping, the bed is subjected to any considerable mechanifeai disturbances, as by spadinir, by which these envelopes are broken/^nd detached, the result is a decided inferiority in the ^ality of the effluent. It is a well known fact, that the longer sand is in Wse the greater is its efficiency for filteHng purposes. / All these considerations go to show that while undoubtedly most of the purification takes place in the surface layer, it is not absolutely essential. The facts stated are chiefly of importance in so far as they indioate the true principles upon which the process of filtration depends. They should by no means tend to lessen the oare which ought always to be exercised to preserve the Surface layer ihtaot^ The purifying power of the main body of the sand should be considered, as a factor of safety, and as an additional guarantee of good results. As to the influence of the site of the sand, it may be stated generaUy that the " uniformity oo-effieient " should be as bw as possible. Also that the smaller the " effective sise " the more efficient is the-flltration the less liability is there to disturbing effects, and the sooner does the aand arrive at its full bacterial effioienoy. At the same time it must be operated at a lower rate, becomes clogged more easily, and thus w- 1& ■\ ■ .^ ;' ^ ./x f*. V quires more frequent sci'apmg. The Utter performaDoe, tpgeChor with (he periodio renewal of | the sand, will form the priocipal jwrt of the ezpenfte of operation. There h thus ^a minimum limit beyond which it would be uneconomical as well as unnecessary to go. Tbci best sise, taking everything into consideration, will evidently depend to a ooD- yiderabie extent on the qualiiy of the water ani o^cr local eironm- itances. The "effective siie" of the sand used in the Principal l!uropean filters varies, according to Mr. Allen Uaxen, froni .20 to .44 milli- ^ metres ; and the " uniformity cocflici(|pt " firom 1.5 to about 3.7. Asx^o the proper depth of jthe sand layer, there is even now con- siderabl^e difference of opinion among engineers. The great variations in thef a thin bed, with le European filters the^bed bus been le limit imposed by RTOuld seem to be •m 2 to 3 feet, in og effect produced ional advantage of turbed, and which, ioreaae in > ■-•■■■ '/X * nd to ooodnot the ive thickness used ill necessary, 12 or jfSor 4 layers of stipport the sand /Ground the open- bould be oarefiilly. i when the water thorou^^hly washed V ,•■■ Brieg. Bremen (old). Bremen (neio). Roftock. Altona. ■ (iueatrow. ^Stettin. Stuttgart. Luheck. M^geBmrg. 1 Srealau. Pj^ten. Berlin (tegel). •• (Straulau) SriiuMwick. Liverpool. Leicetter. . Southtoark & V* DuBlin, Chdtea.* Lambeth.* ■ Grand Junction.* We»t Middle$ez.* Homaey. Stoke Niewiftgton. York. im H. London.* New River.* , Edinburgh. .I* .M.\ J/^i^xn «u*^.i^ .)te^& i' .V ;. CNDKRDSAINS. In arranging tho upderdrainage system of a filter, which inblades the gravel bed, the object to be aimed at Is to cause the water to sink vertically through the sand, and as nearly as possible at a uniform rate in all parts of the bed. In order to effect this it is evident tha^ ^^^^^ ^j^ W-Jl^ -^ Fm. 3.— Section of Hamburg :fllt«r bed. .,... . > ■■I the resistance to horizontal motion in the ilnderdraining system must be everywhere nearly the same. Attempts have been made to calculate the proper size of the underdrains, using forkulie for tl^e flow of water through gravel and sand of various sizes. A discussion of th e matter will be found in the Report of the Mass. State Board of Health fbr 1892, and also in Allen Hazen's bo-,:^ ^ms ^— — — • • * k ■:->^M': «'.'. J^SJ?^^|" mMm'':-f^f^M^^ vc/umu-i W^^>Bm*H'MtL •=£ bed. 'draining system must been made to calculate B for th,e flow of water eussion of the matter Board of Health tor 'Filtration of Public lins, and a daily filtra- Hazen suggests the tflbrent sises should be To drain an are4 not exceeding , 290 square feet. 760 «• 16.30 « 2780 " 44u0 <• Correapooding velociiy of waT^r in drain. \ 0.30 ft. per sec. 0.36 " 0.40 " 0.46 »• 0.51 « ^dnrofthoareadramed. With the rate mentioned thU would gi v. a maximum velocity in the drain of 0.55 ft. per second ^ Fia. 4.-^S«5Uou o« Alter b»d prqpo«m in 1806 by Klrkwood for St. Louis, M<,: These underdrains arp variously constructed of open jointed channehof stone or brickwork, or of tile pipes. with perLtZ or open joints, (Sec Fig^ 3. 4, 6, 6.^ Ther« is no'idv.nt.ge in spacirthe laterab mc,^ than about 16 feet apart, as the extra qStntity of cSi«I ffavelnecese^ry would cost more than the saving in the pipe. In «,mc hltll f '" K T'V" "^ «««>'°P>«hed by means of a double bottom of open bnckwork supported o^^rches or other arrangement, of the same material. ^^ o — *» I'Le lateral drains usually rest ilpoWie bottom oiLJir< on tiK li» of the basin, but J V* l> •■- Mit!-\ „a / ■ r. v'i. y the maiD draio is oftett pitoecl Idwer. (Sec Figs, fi and 6.) If the top of the draio is higher than the qoarsest layer of grarel* that part should be < closed to preycQt the entranoe of the fine gravel. > - t ■ :»^--\ I I I I I -»- 9mN0{'. Fio. 6.— Section of Alter b«d Mew Birer Water Company (London). In several pf the old filters Tettieal TentUating pipes extend Arom the nnderdrains above the surface of the water on die bed. These are for the purpose of allowing the escape of air from below, so is not to cause disturbanoe by pasmng through the sand. They are not used in the latest fitters, as it was found that they were of no advantage, but rather a source of trouble, through the formation of ohannels between them and the sand, wKich allowed water to pass witlioi|t filtration. The baab whieh enolMes the filtering materials must of course be water-tight ; and in that respect the same oare must ly exercised in its design and construction as would be necessary in the case of any reservoir for holding water. Its depth will depend upon the thtokness . of the bed and the height to which the water is to be allowed to rise, but does not usually exceed 10 or 12 feet. The bottom ii usually tevel, or perhaps with slight depression for the lateral drains. The walls 20' ._ c "1 "W" '■^ ,Vf 6.) If the top of tho bat part should be ^ / r lmK/t«r 'C iny {haaian). KB extend ftom the I bed. Theae are )low, w ii not to ley are not used in DO advantage, but ohannelB between ioi|t .filtration. / »;■ -H>)t-.. >v ,.S» / aiut of ooorse be b l^ exeroieed in 1 the oaae of any upon the thiekneM. « allowed to rise, tm |i umully lerel, rains. The walls t ■i-^ '\ <( -»,./ 1 "M *:'•, , k / , ^'■^ i |i if' * % r?i* *'**"•' ^®'"'''*' ®"'°P'°8. «>«P«diPg upon the maUriftl used. *!*% into aoooantthoneoesBily of uniformity in the fllJr»tion rate ^fnl points ofthe bed, vertical sides are probably preferabre to Local oiroamsUnces mill as a rule determine the best method ^-..jotion and the material to be used. The latter may include , 10 masonry, brick, concrete, earth ombankmontii, pud O ,>s z 4 $ '••'<■? K y -'"^ <. ^m . O' ■''^py !i! ■>:%* M K (f- ■•!■*'' u> f-.v V.v . H -i ^:f'-^'' u. a • i t. ' ■ : Ul »i •<-,';. c w^- Ul > K:'*Jif5 O (> ^■■''••'i.-i 1 ji;»y ;■'.■■ 00 a HIttMriv u. ^^Kv^a' KIs^' l^j'^^ iiffi^% iff Via. a— Setitlon of Alter bed for Albany, N.Y: Concriete is a very satisfactory, and iu most cases an eoouomical ma- terial to use for any part of the structure. ^ Ifa roof is required it should consist of groined arches, supported on pillars, preferably of brick. (See Pigs. 8,. 9 and 10.) Care is ne- cessary tOobtain a i>o\}d foundation for the latter, as the form of roof will not admit of much unequal settlement. A good plan is to form the bottom of flatJiBjrcrted arches which will giire a firm and even sup- port for all the pillars ; and the lateral underdraidk will then l|e along the hollows midway between the rows of piers. (See Figs, ti and 7.) With a roof of this kind vertical side walls will be moire economioal^than sloping ones. But the plane surface between the i^lLand the bed must be broken by projeotions, in order to jireventtfie liability of un- ' filtered water passing alpag the jonotion ; which romark also appHes to the piers. It is to prevent this same contingency that the graVScl' layer is only carried to within 2 or 3 feet of the walls, its place beiiL filled by tba sand which hero oompoeies the whole depth of the beX " 21 r^ ^. 4:^ '■^ll^'. M T: Tf- h^ .:"t i. 1 -h.. .!:'■, 'y^. (See Figs. 6 aod 7.) Around/ the inlet and oatlet ohambera there>^oald b« DO graVbl withio 5 or 6 fejbt of the walls. li'on Htrough Piers. Fio. 7.— sioUoB of filter b«d (or Allway, N.T. !" , Manlioles must be constructed in the roof for the admission of light and air. Also a " run r for entering and removing the sand scrapings, etc. With piers spaced 14 or 16 feet on centres a light and strong * roof can be built of concrete at a very moderate cost. When the roof , ia-^nTshcd it & covered to a depth of two or three feet with earth \-f ' surmounted by a lajor of loam, which may be seeded down or laid out - in flow^ bods*, etc. 1 For open filters thW sides may be of earth embankments, made water 'tight by o layer of paddle or concrete., If of the former, a paving of brick is necessary, which must be of sufficient strength to withstand the action of the ice where it is exposed. . OPERATION. ' , Before proceeding with the methods of operating a filtration plant, reference will be ma|le to Fig. 11 which shows diagrammatioally a filter bed with inlet, outlet, underdrains, eto^ With a given flow of water through the bed, j the vertical distance H represents the head re- . quired to foroci thiil quantity through the surfooo film, the sand, gravel and underdrains. 'It is variously termed " loas of head," "head on the filler,'* ''filteripg head." ■^m' KtNLtT --fl-T */«. ...J., o^4tsAm »«« 04f4ti 'ipi Fio. n.-Diagram showing THTloM parU of » fllierbed. The depth of water ia the majority of Earopeao filter beds is usaallj from 3 to 4 feet, with the full depth of sand. In some of theae filters, it was allowed to rise and fall aceordios: to fluctnations in the removal of the effluent. Such variations in depth, however, are found to have an injurious effect upon the surface layer, aiid on the efficiency of the filtering process ; in the newer plants, therefore, they are provided against by an apparatus on the mouth of the idlet pipe by which the water when it reaches a carUin lieight automatically doses the inlet. These consist usually of some form of baknced valve worked by a float. In connection with an open filter such an arrangement must be pro- tected ftiom ifro8t.jr- —^ The inlet openf^into a small chamber at the side of the bed from which it is separated by a wall. The water flows over the wall on to the bed, and is prevented from disturbing the surAuM of th« sand by paving it for a short dbtanoe from the chamber. Sometimes the water enters by overflowing an open masonry channel extending lacroHSthe surfuoe ffthebed.. (See Fig. 4.) a3 -♦^ J.-, JL "i^ 'tJll^^l }\ J The loss of head, oorresponding to a given rate of flow of water through the filter, will depeodi|pon the extent to which the surface film has formed, and the friction io the sand, gravel and uoderdrains; bfit under any given conditions it varies (within practical limits) directly os the rate. In some of the old filters, the outlet was connected directly to a clear water basin, or pump well ; and the difference in level between the surface of the wirter in the filter and of that iU4he well was, of course, equal to the bss of head. Fluctuations' in the draft upon the well produced corresponding fluctuations in the fllteridg head, and there- fore in the rate which was thus automatically adjusted to the demand • In others, however, some sort of apparatus was placed between the filter and the clear water bamn by which the rate could be kept constant. This is now considered to be of the greatest importance for the reason that bacterial tests of the effluents have shown that marked de(erior4v tion invariably follows fluctuation in t^e rales. This is probably caused by the mechanical disturbances/produced in the sand bed and surface film. For details of such t^, see Beport Mass. State Board of Health for 1894. In the newest plants, therefori^ iame kind of an arrangement is always provided tor the rcguli^n of the flow, examples of which are -shown in Figs. 12, 13 and 14: /'Since the rate varies directly as the loss of head, it is immate^al^ which is regulated. Some of these devices, therefore, r^ulat^the flow directly, while others accomplish the same result by the uMireoi method of r^ulating the loss of head. In oi;dto that the fomrir may be kept the same from one day to ' another, the latter miul^be gradually increased as the period of service of the bed extendsv^o as to conespond with the increasing resistance of the surface lifyer. This is effected automatically in the device shown in Fig. 12, which was ddiigned by Lindley for the filters at ' Warsaw. The apparatus is oootained in a wate^tight chamber, oon- neeted on Aao side with the Alier, and on. the other with the olear the water can pass from one Co the-., to which the slits in the sliding pipe extenn' beneath the surface of «hf water. This is adjusted by weights at toother end of the chain whi( h passes over a pulley. Thus the rate ratabe kept constant ; and as the resistance of the bed inereases the yfcvel of the water in t|ie chamber will automatically adjust itself to pro- / duoe the necessary diffiBrenoes in level or loss of head. / 24 ''',■' water basi^. The rate at which other Spends upon the depth f^ I. ' ' *. ^ III u < -I 0. z. o z < z < a Ul (0 oc Ul I- •9 ul oc Ul g o 1. V Fie. 12.- Itegulator designed by Undley for the WarMw alters. ' Fig. 13 shows tlie method ofF^alation devised by Giil for the Tcge Works of the Berlin Water sapply. The outlet from 'the middle chamber is through a weir; and the depth of water on its crest, and, therefore, the discharge, is indicated by the height of the float read on the scale a. This is kept constant by means of the gate. The corres ponding loss of head is shown by the diffisrence of the readings on scf le h. Keeping the water in the filter always at the same level, a constant rate can only be maintained by the gradual falling of the level in the right hand chamber and a eonsequent wider opening of the gate. Fig. ll-shows the prihciple used in the new Hamburg filters. A nmili^;r method was recommended by Kirkwood for St. Louis. The ' ^ .::.'1 2& w fl: '". /■ I' if / 11:/ .■ / 7 /■ /? "■ / -■■J; ■::-l " Fi^o. 18/-R«galator nsedrin the Berlin (Tegel) wotka. scale on the right reads downwards, and its lerooorresponds to the leviel of the surface of the water on the filter, which must of course be ketot constant.' In the. first chamber is a float with a pointer attached. Tho readinsr of this pr/inter on the right scale evidently gives the loasof heM- The reading of the same pointehH>n the other scale gives the cdr^ ponding rate. ' QMiis is accomplished in the following way : The Out- let of the first chamber is through the weir, which is movable in a vertical direction. The smaller scale is fixed .to this weir as showii in Fig. 14, so that the distance between the .crest of the weir and tlie/ieR; «f the scale is the same as that between the pointer and the trate^ line of ihe float. Both loss of head and the rat^ may therefore be jiega- lated by lowering or raising the weir. ^ [ As to the limit beyond whiob the loss of head sboald dot be allowed to go, the general opinion seems to be that it sboald kot be greater than the depth of water on the bed, though the IiaimQee «. perimebts ha^e not shown any bad eflects from exceedii^; thijl limit. As a general thing it maybe stjited Ibat, everything else being equal, 26 ?L -j- ^S«i»M«ttt<«HnthenewH«rabBrf flltew. the higher the rate the poorer the quality of the filtrate. But with fairly low rates this deterioration is slight, so that eniircly siitisfactorj results can be obtaiaed with rates up to H or 3^ mUlion gallons per acre ,^er day. Probably higher rates could U safely employed if very great care were ctercised in the operation of the filter. The relative efieots of high and low rates from a financial point of view are diacusscd further oji: ' / I V , SCRAPING THE PILTBB: / When the clogging of the filter bed has Become such as to require * loss of head greater than the prescribed limit, the inlet is closed and- the #ater allowed to drain away until i>«MhiJk some distanoo below the suriaoo of the sand. 'When this has beorfne saffiofentlyllrm- worknien enter the bed, with planks, wheelb/rrows and broad fiat ^: ■/■,". 27 Jli 3S . J-^'^ JS^dia^» m m 4'. . tf. ■ ti shovels. With these they carefully Temove the garface layer and pile it up in little heaps, which they afterwards remoTe with the barrows. The depth removed varies from ^ to I inch, and averages, about ^g of an inch. ..The surface of the sand is then raked to loosen up the packing caused by the boots of the workmen, . and after smoothiag down any irrf^gularitiea the planks are removed and the filter is ready for another period of service. The refilling begins from belilw by admitting throi^;h the undcr- drains filtered water from another bed in action. The object of this is to drive out the air from the pores of the sand, where its presence in the for ji of bubbles would, cause considerable unneisessary friction. When the water has risen a few inches above the surface of the sand, the lower connection is .shut off and the refilling is completed by means of the surface inlet. .^ ■ "^ , Before fiitration proper begins the Water should be allowed to stand on the bed for several hours ;^or the first million gallons or .ao should be wasted. The amount wasted can be redilccd by beginnin;; the filtration at a tow rate, and gradually increasing i^^ the tnaximum. / When the scrapings have reduced the sAoS^bed to the minimum allowable thickness, the t,otal amount removed, whioh has in the meantime been thoroughly washed, is replaced at one time. Before doing so the surface of the permanent layer which is never Amoved, should be loosened up by being spaded over to a depth of six inches or so. If this is not done, there is a liability of sub-snrfuce dogging at its junction with the clean saiid. '".,,, When the filter 18 started Again, it is, except for the permanent layer, in the condition of a new filter, and so requires extra care in operating it, and the filtrate should be wasted for a much longer time than is required after the scrapings. Piefk^ of the/Berlin Water Works places this period at six days. Considering^ the labor neoescary and the time the bed is out of use, this replacing of the sand is an expensive operation, and should not occur oftener than can be avoided. In most plants the usual period is about once a year. ' \ ■■ \ SAND WASHINQ.J Sometimes it is possible to obtain new clean sand at less cost than U necessaily to wash the old.- But this is rarely tlie case ; hence an im- portant part of the equipment of a fair-sized filtratiob plant is the apparatoB for the washing of the sand. The simplest of the methods ' 28 / .-■*^' J^ emf loyed for this purpose oonsista of a broad 'shallow box, which is set in an inolincd position. The dirty sand is thrown into this box, and a jet of water played upon it from a hose. The water overflows from tjbe Wer end of the box and carries the dirt with it. This is oon- tihued until the water runs off clean. The more elaborate methods employ mechanical means to force the water throt^gh the sand. l)rnm-washcni, operated by horse or steam power, are largely used in Grcrmany. They are set in an in- clined position, ane, uot being at all superior to those of continuous filters, while the meth9d of« operation is not suited to cold climates, and either requires a greater area^tif bed or a higher rate of filtration. A description oF the Law renoe filter may be found in Tram. Am. iSe principal item of expense copi^ted wit)^ the dpcntJQn of thii plant is that for Rcroping ; and it is'^Mriid that the amount soraped^fnr any given quantity of. water filtered is independent of the rat^. Alio, the allowance' for the area out of use wilt not vary with the rate to any extent. Hence an increase in the rate wiU not by any means prodnoe a proportionate . reduction in the cost of filtration. A rate of H,000,- 000 gallons per acre of bed in use will give results entirely satisfaotc^ from the standpoint of efficiency, and. at a cost which is usually by no- means excessive. The sixe of the individual beds will depend in.part upon the extent of the total area, the smaller plants having necessarily to use smaller beds. ' A large bed costs less per-unit of area th^n a small one, on aooount of the proportionately greater length of wall in the latter case. With a large b^d it is, however, probably mW difficult to obtain ar uniform ute of filtra^on over the whole^area. ^During ilieirintcr of cold climates the oost of maintenance is consi* derably incrtesed by the expense of removing the ice which forms in the bed. \lt is also difficult to avoid injuriously disturbing the surface of the Ban«Ik Beside this when the water is dr»wn down, the surface sometimes Jffcezes before it can be scraped. On account of such disadvantages as these filter beds should be covered in all cold oli- mates. The l|est method of ooostrncting these roofs has already^ been referred to. \ ^ • The proper number, shape, and ar^of the bedii of a system otnonly be determined for any particular case -by careful study of the local coii- ditionsj and by mfking cottiparative estimates of the different items of- cost of construction, maintenaoco, etc. There will be opportunities for the exercise of considerable ingenuity in the general laying out of the system, the relative placing of its parts, the arrangement of the piping, drains, etc., in order that convenience and economy may be happily combined. ' : , -. ' THE COST or. CONSTRUCTION. <■'■,',: /» Thi^ will of course depend On the local oircumstanoeaand the kind of materials used. As in all hydraplic work, great care is required itvthe oonstructioti, and the best quality of materials must be used.; In the main, it is thd same class of woric as is required in the building of distributing reservoirs. ' r ■mL Wt y \ # ft fl \ '.■ * • #; \ ", 'M ■ .>:?k^^''-'fei*- 4«%«#,'4 > '. t I; i. ; ♦ ■» FiB. 10.-FILTEH BEDS AND CLEAR WATER BASiN AT ILION, N.Y. pe ■^ . ■ $-: u Be Hi W Zu Ni Hi Hi Aa So Po I. 1! 18 OOD pie in( not ] / .1 cov N wa be ■ bao waf Fig kk tLLEPTICAL URUINbU AHUHtS USED IN R06P 6f WELLE8LEV , (mass.) covered RESERVOI^ w it *', The following table gives the cost of construotion for seveN^i^aro- pean and American iBlters : < Place. >••••• !•• •••••« London Berlin (1884-87) .... Hamburg (1893)' ..;... Warsaw USSS) . .'. . .v '.'.. Zurich (1886) Nantuckipt, Masn. (1892) '...'. Hudson.'NiY. (1874-88) IIion(1893) Aahland, Wis., (1897) ......'.".'..! Someraworth, N.H. (1898) Poughkeepsie, N.Y. (1872) do do (1896) Coat per Acre. Covered. $66,000— $70,000 78,000 86,000 80,000' 64.000 Open. $24,000— $40,000 30,500 45,50%, - 73,000 96,700 90,000 41,000 Lindley gives the general cost of <»6tinental filters as $45,000 per acjre for open, and $68,000 fon covered. (The following figures, giving in detail the bids received Feb. 15th, 1898, for constructing tho water filtration plant now in process of construction at Albany, N,t., will be of more interest. When com- pleted it will be by far the largest plant yet built in Amerioa; and in general design and completeness of cquipment.it will be" second to none. It wil consist of a settling basin of 16,000^000 gallons capacity, eight covered filter beds, each with yV of an acre of sand surface, and & clear water reservoir with a capacity of 600,000 gallons. There wUl also be provided an office building containing fully equipped chemical and bacteriological laboratories. The price for the sand and gravef included the necessary soroehing, washing and putting in place. ^ The sand washer is of the ejector typo. Other detaik are shown in Figs. The bids are as follows : .\ 8§ VOIR. '■^i ' •■i|j IM M& •! ,!■ i CM 11 t Material. Shale Rock Excavation Earth excavation (above .grade)..... Earth excavation (he low grade) Boiled clay and gravel embankments Silt and loam filling... General filling (rolled) . Puddle in place Broken etone or gravel for lining Sodding Seeding Gravel in road war rolled Vitrified brick laid as &• .*•..•••••■ pavmg Stone curbing^ Concrete in fUHtrs ..,. . Concrete in vaulting . . . Other concrete Brick u>6rk In)ip'd E^rtland Cemeni AmericaiJI do do Rosenda^: Cement. . .-. . Furnishing ani placing 2r'dra.\n pipe in piers. 2 in. agricultural drain pipe.... ., 6 int. drain pipe open joints : Fur. ani laying all vit. pipe cement joints . . . Placing i aU gates, etc I\tr. by board Fur. and placing all castironpipe and spe eials IronAlttr covers .,.,,* Sand washing appara tus ...... Sand run fixtures Eegulator houses Omce a)»d laboratory ouilding Filter gravel in place . . lilter sand in place. . . . Split atone lining Rough Stone paving, . . Fasteners fur. and placed in concrete vaulting . . Iron liBnce .. OoAneotion with pump well and closing .old • • •••••••• - Total .,..., Quantities. 5,000 cu. yds 60,000 « 3,000 " 21,000 2.%000 18,000 13,000 1,900 " 3,000 sq.yds. 8 acres 800 cu. yds 120 M. ^ 800.1in. ft 11,000 cu. yds. 7,000 " 3,000 « 4,500 " , 500 bbls. 14,000 " 1,500 " 2,000 lin. ft. 16,000 " 672 each 2 sets 8 each 8 each 7,000 ou. yds 36,000 " 2,0^ B(j. yds. 3,000 860 lin. ft. Prices of Success- ful Bidder. $ 1.20 .27 .30 .52 .15 1.18 .71 J L/15 2£rl.00 .60 '20.00 .60 2.31 3,85 2.13 812} 3J121 2.|14} .m 525.00 .04 .11 5,337.00 1,140.00 20,701.26 4.40 393.00 407.50 862.24 4i881.00 1.05 1.00 .82 .82 2(k).00 a.00 Other Bidders. f 0.30 to .13i" .20 •• .45 " .15 " .07 " -.67 " .99 " .15 " 12. 6P " 1.00 " 18.00 " .75 " 2.34 " 3.90 *• 2.60 " 7.00 " 2.35 " 1.90 " .86 " 300.00 •• .05 " .10 " 3,850i00 « 700.00 " $ 1.50 .476 .60 .90 .50 .40 1.07 1.40 .60 70.00 , 1.60 30.00 1.50 3.50 7.50 4.30 10.00 :f 00 * 2.21 1.95 1,318.00 .10 •m 5,933.00 2,000.00 14,760.00 <* 20,000.00 5.00 " 6.60 250;00 " 1,000.00 100.00 '< 511.00 1751,00 " 900.00 2,700.00 " 1.00 " .90 " 1.0« " .93 " 160.00 " .1.00 " 10,200.00 2.00 1.78 3.60 2.50 .225.00 2.00 Tjoilr c « •a.i $ 0.80 .20 .40 .50 .20 .20 1.00 1.26 I .16 60.00 1.00 20.00 .50 3.00 5.00 3.00 6.00 2.75 2.15 1.00 700.00 .05 .10 6,000.00 470.00. 15,000.00 5.00 800.00 200.00 500.00 3,000.00 1.50 1.26 2.00 .80 100.00 1.00 » .t09.8«6 5i4 1 -tjm- ^i^sopr I322..358 •' . S.^87.346 •322.440 ." • m Sg •aJ a ^ W:2 50 $ 0.80 476 .20 60 .40 90 .50 50 .20 40 .20 07 1.00 1.26 I .16 50.00 J.00 20.00 .50 3.00 5.00 3.00 6.00 2.75 2.15 1.00 700.00 .05 .10 6,000.00 470.00* 15,000.00 5.00 800.00 200.00 500.00 3,000.00 1.50 1.26 2.00 .m 100.00 1.00 The items of special interest ar| given in italics. ; It will thus be seen that a oovered filter plant of this area (5.6 acres) with settling basin, dear water ba«ln, and all other appurtenances com- plete can be bialt for less than^t£^,000 per acre. For further details and infom^tion concerning this plant see En- gineering News, Feb. 10th and (fcti 20th, 1898.. . • ■ MAINfBtrjlMlE. The total cost of roaintenanoepf a filtration plant is made up of the ope^tiog expenses, and the interest and sinking fund charges. Tie former— the operating ^penses— odmprise :— (a) The cost of superiotendi^nce, and of attendants to look after the relation, etc. ,..."*""^ (6) The co« ^of'Steraping and removing the sand. — (c) The cttst of washing the sand. {d) The cost of replacing the washed sand when renewal of the bed becomes necessary. , It is onlj in very large plants that a special superintendent is re- quired, so that the expense for that purpose would not form a very laigc part of the total cost. The proper handling of the gates, and the , running of the plant in general requires a degree of intelligence con- siderably above that of the ordinary laborer. The wages of the gate- men therefore will be from $2.00 to $3.00 per day. Scrapin? and removing the sand by wheelbarrows sterns io cost, under ordinary circumstances, between $40 and $50 per acre, depend- ^ ing upon the wages paid. At Mount Vernon, N. T., tlie s'hoveUers and barrow wheelers arc paid $1.10 per day, and the i^crapers $1.25. Mr. Chas. Fowler, for many years in charge of the filters in Pough- keepsie, N.Y., says that one man working 1 hwrt- is required for every 160 square foot of surface cleaned and removed. This would cost at $1 . 50 per day, about $43.50 per acre. Lindley gives 30 days at 10 hours each for every acre, which at the samo rate is $45.00 per aon; In the small plants at Ilion. N.Y., and Ashland, Wisconsin, the cost 18 at the rate of about $50 per acre.. \ l ^ The cost of sand Washing variei with the uLthbd empbyed. In Poughkeepsie, when they used a sinple inolinedl trough and water jet it cost as high as $1.50 per yard. By Improving their methods they reduced thU cost, till to^y it is only! 27 ots. pjL yard. la fiadsoo the cost is 20 cents per yard, and in Ilion 18. In Germany it variM from 14 to 20 cents per yard \ ~ ^;50o;:m" $322,440 r MM^^hiJ:^^:. fi 41. h I I •♦ • ' The periodical replaoiDg of "the sand in th« bed must be done care- fally of course, but should no|i cost, including wliatever is neo^ssaiy to be doAe to the permanent liyer, more than 40 cts. per cd. yard. The cost of these varidTks operations will of course depend upon the scale on which they are carried on. It will obvioulsly be easier to ^eep the price low with a large plant than a small one. In the'pase of the- former a force of gatemen and laborers can be pcrmanci)tly employed. In the smaller plants the operations of scraping and sand washing onl^ take place at intervals, and arib performed K>y laborers hired temporarily fortbe purpose, or by employcias ftom, other parts of the water system. lir using the aboVe data to make an estimate of the total operating ^ixpenses, \re shall employ as a unit the ciost per milliojn gallons of wa>erfiltered. ' . ^ ^ Asstttiriifg an average yield of 50 million gallons per acre between scrapings, the total cost would be as follows : * Sdrapiug and removing"^45| per ao $0.90 per mill. gals, filtered. WashinglOOcu. yds. 8an0.26 '* .*!» *;*\ tvm ■■■■f.'.VAf X • s. . { ■• " .. care- ififlary ■ 1 ' • ■ n the v ■ ■ Icee^ . ' *i ■ .. ■ >fthe- S " • -^ . '; .". oyed. ' .1 . " ' ■: only ' * ' ^ rarily '* Item. * . atiog , • : ; " tns of ■ ■ - ^ iween ■a^ m ^ teats test^ fUly fthe ' the ' o ^ 2! a O as .J 1 < 1 X CO -S O en 3' -4 H < s- S < ^ f^ O D S * T >; :S o I— 1 !t •J a •J 5 S OS -, w cu fe ► - . o 1— 1 • H f ■ ■< fc""*^^ -^ « • ^-s )-l .■ b, < :,' 04 o H 00 'O O =»= """T h^*: ■•?■ ,■ j«t:i r»i,' T 'i4\ .^4 \'\ ■=/ ^- tJ 4 ■ , To get the total cost of miintcnaacc, wo must inoliide with the operating expenses the charges for interest jtnd sinliug fonds. ^Ms will of course depend upon the cost of construction ; and the latter will vary with the mazinMim rate of filtration adopted, and the proportion of the total area to be out of use while being cleaned. These being decided unon, it wijl then be easy to calculate the first cOsf per million gallons of daily yield. For exaniple, if,"^ith the rajte chosen, the daily," yield of the plant will be 2 n^illioa gallons per acre of the t4)tal ariea of beds, the first cpst per million gallons will be h^lf the cost of gon- straotion per acre, 'and so on. The diagram gives the eodt per million gallons filtered, corresponding to different construction costs, which will pay the interest and sinking fund cjiarge^ necessary to cancel the whole first cost with interest at the end of 40 yenrsL For example, v^ith a first cost of $60,000 per a^r^, and ^ net yield of 2 million gallons per acre of tot^l area, the cosd per million gallons > with interest at 4 per cent, would, bo S4.15. ^^, '** The amounts taken from the diagrams added to jtlie estimated iRt of op'oration wUl give the total cost of maintenance per million gallotiB of water filtered. With interest at 4) per cent, and a first cost o^ $80,000 per aoj^e, this w(^ald amount to about $6^' J^di to tfiis $2.60, for the expense of operation and we should have $8.50 as the total cost of filtering 1 million gallons of water, or 1,000 gallons Tor less than tV of a cent. With open filters, or more favourable local conditions, this charge wonld be considerably reduced. " " Having now discussed the method and dost of Sand filtration, the cext and last question to be considered is the nature' of the results which this process can be depended upon to produce. There can be no question as to its efficiency from nn seathetic point of view. The complete removal of even the most minute particles in suspension, to- gether with a large partof tbe dissolved organic matter, veosures the entire elimination of any oharaoteristics the water may possess which would be disagreeable to sight, taste- or smell. Yet it is because of the effectiveness of the purification from a sanitary standpoint that this system is especially noted. This is due, as we have seen, to its destructive effect upon the baoterio, whiclp is almost sufficient to cause their entire disappearance daring the paange of the water through the filter. The average reduction 'tnTa well designed and well managed plant will be as great as 98 or 99 per cent., as shown by comparing the number t>f germs in' thie effluent with that in the applied water. But in reality it is even grelnter than this. For I - Q K -4 «> A4 ■ ■» Off of^m. is^it.t.$ot^' «^Li.ot^m. oiAQRANf aivma that portion, of the cost of filling a million gallon* of water which It NECESSARY JO PROVIDE FOR INTEREST. AND 40 YEARS SINKINd FUND. ^ i, , ■■Ml " ■'■:. 'if. ■-,„ '^:■■> ■■'■ it has be^ showo that of the fettr baotena which "{ii'o present in the efflaent, a 'CjBrUin nnmlier oome from the undcrdrains, and have therefore, .pot pasKd throagh the fiU«r at, all. ! These belong to som^ .of 0ie species of water bactcrin, and, oon^^^luBntly, will be quite harmless! '¥rom tests made on the experiniealal filters at, Lawrence with %n easily rccogniied and haiidy species, the actual reduction was found to be from 99.9. to 1(>0 pet- cent.'' Now, when it'ip considered the filter is capat^le of producing suol^' effects upon bacteria wliich exist normally 4bu water, it will be evident that the 4fl^t upon the patb* ogeni^ ort^diisease germs which iare out of their tiatnriil habitat and in a decidedly unfavourable envJironmeDt will bq»;^ach greater. Thus what may be called the "'hygienic efficiency V of^this system must be reiQarkably high. The process is comparable-ib nature's method of purifying the sutfaqe water which furnbhes the under- ground supplies; aqd if properly' carried out, the v^atcr produced is probably of almost equal whole.somcnesa. The continued experience of places where sand filtration plants,,have beon in operatioir for some time only gdi^ to strengthen this otnclusron. In America the method is only just beginning to be employed. Up to the yeur 1892 there were but two plants of this description in the country, vie, those at Hudson and Poaghkeepsie, which hfive been already referred to. Since that ^tioie '14 new ones have been com- pleted and three others are under oonstrdction, t^e latter including the lai^ plant at Alban v. The experience to b% derived from these plants is too limited to be of much value for some time to oome ; but the officials connected with the majority of them have invariably expressed their entire satisfaction with the method of working and with the results obtained. In England and the continent, however, the experience of many years is available, this method, as we have seen, having been used long before the ratiomtU of the process was understood. In England particularly, many of the supplier are from surface water;) made avail- able by means of large storage reservoirs ; and in nearly every instance the stored water b filtered before being supplied to the consumers. The new supply for Liverpool, which was put into operation in 1892, comes from an artificial lake formed by damming the Yyrnwy River in Wales. This lake is 68 miles from Liverpool, and lies in a sparsely inhabited district remote from railways or towns. ITet the water from tliis source, safe as it may appear,- is also made to pass through asand filter before being albwed to enter the distrUKltion pipes. In Germany the use of any sarfaoe water without miration is prohibited by law. ft * lil »w But in addition to the high esteem in which the method is held whereyer it has been used, there are certain health and mortality Statutics which are perhaps of even greater significance. The typhoid fever death rate IS now considered to be a pretty good index of the punty of a city s water supply. Keeping this in mind tiie foUowiog * tables prepared by John W. Hill. M.Am.8oc.C.E., will be found to tumish interesting information in this connection. Tablb I.— Ttphoid Pkv«r Death Bate. Class I.-jq or less per 100,000 of Population. CITY Tbefiague.. Botterdam .. Ghristiania.. Dresden Vienna anich Copenhagen . Berlin "SOCBOB OP WATER SUPPLY. Filtered from ^and dunes. . . . . tiUered from River Mease.. .. Filter Uallerj bv River Elbe..! Springs in theSchneebere..!. Spring -Water from Mangfall Val'ley!'." ^iiVer;d7romL:TVg:r;;dRi;;;8n;;;. 1R90 3 6 12 9 9 8 9 9 I8il 12 4. 9 8 6 7 8 10 1892 Breslau . . . . , Amsterdam Stockholm . Brisbane..., London.. . . . , Class 11.-10-20 per 100,000 of Population. 1893 2 5 6 4.S 7 15 9 9 1894 3.4 4.8 3 8.2 5 2.6 6.7 4 ATc::- < K \ for .4.9 6.2 6.8 6.9 7.0 ,7.1 7.9 8.0 .1 5 3 C 11.6 13.9 14.3 14.3 5 14.6 5 9 6 15.8 .]5' 20.4 21.4 21.4 21.6 21.8 22.4 2:1.8 26.2 26.4 27.6 X ^enioe...,. Milwaakee. " Table I— Continued. Class IV.— 30-40 per 100,000 of Population. Rome . Boeton...... Detroit 2vK>n, .. Turin ...>.. Liyerpool ... Buflalo. .... Providence . Covington .. Tjake Michigan «.....'.*!*.' ".* Pontanadi Trevi, Aqua Felice A PtoiV Lake Gochituate and Sudburr River.', Detroit River Driven Wells Lake Vyrnwy J.,' 24 Niagara River. . . Pawtuxet River . Oliio River...... • ••••• I 44 33 30 26 33 33 31 37 35 36 26 34 43 33 29 .10 18 13 51 61 20 ^ 46 32 44 64 41 44 29 24 25 25 63 • • • • 50 34 37 2ft -47 39 M 4.3 45 40 27 18 26 30 28 26 20 24 58 .36 47 42 I 30.2 ! 32.0 I 32.2 32 6 33.8 : 36.0 i 36.8 ! 37.0 [ 39.2 :t9.2 "^9.4 9. Francisco. Prague . . . . . Minneapolis. Baltimore. . . Newark .... St. Louis. . . . .Newport, Ky Philadelphia Denver^.,, Cleveland.... Class V.— 40 50 i«r 100,000 of Populatio^. Impounded from Mountain Streams . . 1 59 .33 Mississippi River ...',..'.'.' ' 81 Lftke Roland, Gunpowder River. . .... nifj^unded from Pequannock' River since April, -18^2..... Mississippi River...: , Ohio Ri vttr Schuylkill and Delaw^reRiver." .'.['.'.'. South Pjaite River...... .. ...... .:.. Lake Erie -. 57 60 34 64 » • • • 66 41 37 45 34 81 30 34 I 42 53 36 36 60 42 47 45 ! 28 64 52* 37 40 63 103 58 41 57 64 I 47 I 27 35 57 45 49 15 31 37 32 35 40.2 43.2 15.1 45.8 45.8 47.0 47.5 48.2 48.3 49.2 Class VI.-50 60 per 190,000 of Population. St. Petersburg. Cincinnati Moscow. ...... ^Toronto Quincy, III.... Dublin filtered from tti<%r Neva. Ohio River,,. Springs, Ponds, Moscor and Yanxa Rivers..... i........ Lake Ontario '.'.'.'.'. Filtered from Mississippi River. . .*. Filtered from River Vartry . . . 57 |..... 67 I 62 73 93 76 62 75 94 28 68 40 68 43 .48 39 51 43 40 42 68 87 49 ; 52.3 60 , 52.4 29 17 79 48 57. 57.8 68. 58.8 Knojtville Milan, luly... Jersey City.... WasbinstODi.. Looisvine, Ky. Chattano<^ .. Chicago.. Pittsburgh.... Lowell, Mass.. Atlanta^ Qa... Lawrend«,Mase Alexandrt«iEg'i Class VIL— Over 60 per 100,000 of Population, li'iliered from Tranessee Rivtr . Passaic River .^....'.. Potomac River Ohio River , Tennessee River ..!...! Lake Michigan.. Allcigheny River." ...!]..!!!! ^iven Wells, Merrimac River. . * . . Filtered from Chatlahootchie River FUteired fh>m Merrin(ac River. . . .. BiverNile Kiver ^»*:i^;^ri'^'2ji'^j0i.j,s.^.^i "41 I 101.5 91 83 88 145 83 46 168 149 123 208 160 96 83 81 66 160 100 98 119 116 348 233 36 62 63 70 72 55 104 100 90 87 102 77 168 67 62 60 66 84 86 42 111 61 66 93 79 154 59 61.9 • • • • 62. 76 75. 71 76.6 72 79.4 48 80. 31 84. 66 91.7 65 92.4 43 92.8 48 .96.2 100 162.4. 136 189.4 / x*i-m'3.LStftii f^WSPW/ Table II — -Typhoid Fever Deat^ Rate. Death Rate per 100,000 of Population, arranged upon Basis of Death Bates for 1894. J': I * Glass 1.— Less than 10. Munich 2.5 ChrJRtiania 3.0 The Hague 3.4 Bei-lin 4.0 Rotterdam 4.8 Vienna 5.0 Haniliiirg 6.0 Brenlau 6.1 Copenhagen....... 6.7 D'.esden 8.2 Stockholm ... .... 8.3 Ahi8terdam . ...... 8.5 Brisbane. 9.6 Class IV.--30-40. Rome/.... 30 St. Lbuis > 31 Chi.Ago... 31 Philadelphia 32 Denver M San Franci.sco . . . . . . .^5 BuflTa'^o 36 Newport, K}-^ 37 Class II.— 1020. ' Brussels.... Buda Pesth Londqn ..... Kjinburgh. Brooklyn .. Newark ... , New York . Toronto Manchester Venice Trieste 14 14 15 151 15 15 17 17 18 18 19 Class V.— 40-50. Covington, Ky v 42^ *AtIanta, ja.. ' 43 Providence, R.1 47 Lawrence, Mass .... 48 ^^Chattanooga. Tenn.. 48 Dublin, Ireland '48 Baltimore 49 St. Petersbure ...... 49 Class III.-20-30. Dayton, O..... 20 Turin ... 24 Olaagow 24 Milwaukee 26 *D9yenport 26 Detroit 26 Cleveland 27 Boston 28 New Orleans 28 MoRCOw 29 Sydney; N.S.W 29 Paris. I.n.... 29 Cuss VI.— 50-60. Gincinnatik. 60 Lowelli Mass . ... ... 53 Pittsburgh.... 66 Prague.. '..'.... 67 *Enoxville -69 Class VII.— Over 60. Milan. Washington City LouisviTle Jersey City *Qnincy, III. Alexan Iria, Bgypt Cairo, Egypt • • ■ • • • ■ •••••*< 62 71 72 76 79 IpO 136 -f^ ..■■r \ ^Mechanical filters. 1^. ^hese figures show that the lowest rates are fur those cities denying their supply either from springs or from aarfaoe waters which hftve bfeen subjected to sand filtration carried out in acoordanoe with the ' , ^ strictest modern requirements. They also indicate the general inferiority _ of -the watet supplied to American cities when compared with European supplies. Chicago and Berlin have about the same popuU' tion, yet the typhoid death rate of the former is more than ten times, that of the latter ; or, in other words, the chances of oontrMtiiig ' . typhoid fever in Chicago are ten times as great is in Berlin. ^, Out of the many' instance!^ showing the beneficial effect o^ filtration with regard to the prevalence of certain infectious diseases, two of the , most noted will beoitcd, one in Europe Vnd the other in America. «... *2 , I ■:. .\ 1894. - . ). i . 20 . 24 • . 24 . 26 . 26 . 26 • . 27 , • 1 . 28 . • . 28 .,29 . 29 «■ . 2]) ^ «i" f. .60 . 65 - . S6 .57 «-, . -59 * #*- ■ m 'Hamburg and Altona, though under separate governmentd, prao- tically form one oontinaoua oity irith a joint population of ahont 800,000. They both draw their water supphes from the sewage polluted Elbe River upttn which they are situated. Altona is the nearest to the mouth of the river, and its i^ter intake being three or four miles further down is below the outfall of both its own isewers nnd those of Hamburg. The Hambutg intake is about two miles above ^the bity. In 1892 an epidemic ' of «holera occurred, during Which Hambvlrg, Vith a population^^f 622,530, had 17,976 oases with 7,611 deaths, while Altona, with 143,000 population, had during the same tim^ 562 cases and 328 deaths, and. in mattj;'of the oases credited to Altona the disease was contracted in Hamburg. Wandsbeisk (20,000)^ just across the river from Hamburg, enjoyed the same immunity as Altona. Both of these places purify their water supplies by sand filtration, ^ whilo in Hamburg the only attempt in that direction was the employ- mAit of settling basins. A filtration plont was, however, in process of construction at that time, and was put ,in operation in Hay, 1B93, since when the typhoid rate Ims diminished from about 30 to 6 per 100,000. • ' -4 The other case is that of Lawrence, Rlassachuset^, which is situated on the Merrimac Bivcr a few mitSssbelow Lowell. In spite of the great dilution, the water of this river wmch supplies both cities is seriously polluted by the sewage draining inlb it from the towns built along its banks; and as a consequence the typhoid fever rates in both places were unusually high. In Sept., 1893, a system of intetliiittent sand filtration was eoii- pleted. The effect upon the health^ of the citizens is shown in the following figures : ' i Deaths from Typhoid . If'ear. Fever per 100,000 Population. 1888; ...i..... 114 - 1889...., .......J:./^. 127 ^ 1890.... ...^..,.,... 134 1898 ..^...i:^;......; .U..... 106 , 1893 ..;!....... k 80 1894- : ..\ 47 (23) • 1896 .....;..,. 31 (17) 1897. ....: ...........1. \.i... 16 ■ . 43 ■' ,. ^ / .. - ■-• I 'i •• '- ^ ..4|£-:.: ■ V.-. :■•,•' ■ . ■ 1 ■ \ :^^--'l *i-^^!^PS«3:^i^teSW^; * / For the last four years the figures in the brackets represent .the number of deaths among people (prin^pally mUl operatives) who were aooustomed to drink oanal water withoqt fiUratioQ ; so that the actual reduction iq the death rate which should be credited to the filterq is much greater even than the figures indicate. Many attemptei have been made duriog the last 12 or 15 years to improve upon the sand filter, but so far without success. Some of the methods invented involve less expense for oonstmotion or give better ohomical results ; but the thoroughness of the bacterial purification effected by the sand bed has not been eqtialled by any. The only one of these methods which has been used to any consider- able extent is (he American cr Mechanical System of Filtration, which is employed in several placarHiN^e United States. Sand is the "ma- terial used in this process also, bdt it is contained in cylindrical iron &nks. The water efitcrs the tanks, usually under pressure, and is driven through the sand at a rate 50 timei' as great ai the maximum .allovfed. in the case of the filter bed. For this reason the latter method, is sometimes referred to as Slow Sand Filteation. With such a high rate the surface film must be formed artificially; and this is doqe by adding a solution of alum to the water, which forms with the . oaxPQoates present awbite floooulont preoipiute. Such a filter will, oi course, soon become clogged ; but it can be quickly and easily <^leaned without removing any of the sand from the tank. The results pro- duced, under test conditions at least, are undoubtedly good ; and it has the advantage over the filter bed of somewhat lower first coat. On the other hand, the charges necessary for repairs and depreciation wil be considerably higher' on accoitnt of the less permanent character of .^its construction ; so that if proper allowaitoo could be made for this, the difference in the actual cost of filtering a given quantity of water by cither procesf would not be very great. In. any case it will pro- bably not involve an addition of more than 10 per cent, to the ordinary cost of the water. Indeed, calculations have been made showing that - if a proper valuation is put upon the lives saved by its use, the con- struction^^ a filtration plant is often in the long run a source of economy, Such considerations, however, seem quite unnecessary. A city's' water supply should be pure,* wholesome, and attractive in ap- pearanoe, just as the streets should be clean and well paved, and the publio buildings architecturally beautiful. Resides, having gone to the - expense of obteiningapublio water'supj^, indifference with regard to ^ it! piurity or unwillingness to provide for it would seem to bo utterly . %'- ■'♦ ■» 44 "M . -J-.-* %'- unreasonable; especially when it can be sccared for such a compait- tisrely small increase in the total cost as sand filtration inyolves. IC would be mncli cheaper, of course^ to purify only that part of the supply which is used for purely domestic purposes. But that ^ould require a double set of distribution pipes ; and we should also lose tlie satisfaction of knowing that the whole of the public supply could be used for any purpose with perfect impunity so far as health is con< cerned. And when we consider that the expense of filtering the whole supply only amounts to two or three cents per mointh for each con-> snmer, it will scarcely be considered excessive. In conclusion, the author wishes to acknowledge his obligation to Messrs. Cbas. R. Bettcs, Oh. Engr.. Quetan's Co. Y^ater Co., Far Bbckaway, TS. Y. ; A, N. Russell, Secretary Water Commissioners, Ilion, N. Y. ; G. £. Hoflhiaster, Mgr. Suburban Water Co., Mount Vernon, K Y.; Sani. Wheeler, Superintendent, Ashland, Wisconsin, for their courtesy in furnishing him -with valuable data and infolrma- tion; also, to W.^. Bryan, M.I.CE., Ch. Engr. of the East London Water Co., ibr titer opporttanlCy to make personal observations of the working of the large plant under hjs control. ■c>. . "i# J), f ' ;>^'* c^^- t IP^^ 'W^S .#» • -' ^■■' > .; ■■' ',; - ■ •/ f !- -■■ ^ ♦ ■ - ,* ^ . '■, . * ■ IT.-;--;:...' , ■"■•■■...-:-.^ ■ ■ • * — -TT'' ■* * ■• "1 ■ r - ' 1 ■ r 4' # i. ▼ « -^-*-^ / 1 i' ■ 9 __ ■ * • » 1 ■■■;■- ;■:;:: .:i ^: ■i .■■ • I ^ * * ' 1 ' y^ • _ ■■.! ■ 1 • f- ,4_ « 4. «. ■/ t *