64 
ANALYSIS OF POTABLE WATERS. 
fiwLi'f ' im f ion °f Nitrites.—The presence of the alkaline nitrites in the well waters 
inr.lnv^l v, ! ^ Un r S ^ • The reactlon of the n . itri .tes upon iodide of potassium and starch, 
kririsif i\Z •+’•+ ' ? nce , as a bes .^ ^ or the iodides, may be used conversely to detect 
urnint •i nte fi m i Wat ^ S ’ which xt wil1 reveal evea when the amount of nitrite 
d l ^ r n ab y le f than / grain per gallon - Xt is merel >' necessary to add to a 
iinlc f+u 1 ! 1 a f v W , d , rops of a weak sointion of iodide of potassium, and then a few 
Xrltinn nf e r f with dilute sulphuric acid ; an instantaneous blue 
i the starch by liberated iodine will occur if the water contains any sensible 
. f The quantity of nitrites may be approximately estimated by the 
iTv O Ol^ radua t e ^ A solu t lon of permanganate which a given measure of the water 
is nmnlovJrl tni’ A " J - C ;i } W1 bleach. When the solution of permanganate of potash 
nerZuStt tlTTZ ^ a ^ oaat of organic matter, if nitrites be present, part of the 
mshntS fJ 11 be destroyed by their action: the bleaching, though not absolutely 
Si P l m ll keS P i ace m f/ e T sec °nds; and, therefore, it can be at once distin- 
matter d wMph^f n T re ^ separately from the reducing effect occasioned by the organic 
manaa n r+p f ffT ve £ graduab The decoloration of 1 grain of a solution of per- 
Snf NO ? Strengtb as 1S described in paragraph 8 , would represent 0-00237 
g / 19 f 01 . 1 c -e- would be equivalent to 0-237 milligramme of NO,. 
^ stl ™ atlon of Nitrates .—The method which may be conveniently employed for 
the foDoww'nrPPP , • 0ntnved by Dr. Pugh (Chem. Soc. Q. J., xii. p. 35), whfchf when 
from whiVb thp fn tl0nS f re ? bserYed ’ “ \ ery tr nstworthy and manageable. The liquid 
the mTxc?wfv Tu J T n 8 , XpelIed ’ as directed in paragraph 10, is filtered from 
about Wlnfl sulphate and carbonate, and is then concentrated by evaporation to 
wi l U;t°t“ Ab °f 30 grainS cx -) of this li( l uid are then mixed 
centrated IS H" • f a ,f roil f solution of ferrous sulphate in a test-tube, and con- 
and 18 alI r V6d sI °wly to flow down the inclined side of the tube, 
thp’pnrnnnraf • l bro ^ n coimution at the line of junction between the two liquids, 
on Z P T nChneS V“ mt . rates f is roughly estimated. The quantitative determina^ 
contaTn y advanWn OC f ede i d Tm The standard solution of protochloride of tin may 
about one fiiirrl n Id per cent, of the salt dissolved in a liquid containing 
of'hard bU - ° f ° rdl 1 na 1 ry by drochIoric acid of sp. gr. 1*150. A strong tube 
8 c c is scaled cp? i C ’ W f h a n i Sea ed Up ’ of . C0ntainin s at least 130 grains of water, or 
into it 1 ^ven hnlt ly J l0 * Ver extremit y> drawn off to a strong capillary neck, and 
SeLuredTv gram measures, or 3 c.c, of the tin solution is accurately 
An eaual bulk of l em ’ not exce eding l-10th inch in diameter. 
smalWi-To-mont of uceMratedwater is then measured in with the same precision, a 
neck is f dAwn offend ,pi 6 4 mtroduced and after effervescence has ceased, the capillary 
necic is diawn oft and securely sealed. The tube is next to be exposed in an aft-bath for 
the t fbeTe 11 w°im t R ^ 17 °° aad alIowed to coob The corSs of 
ounce o wat ^ d beake J ™ th 50 grains of hydrochloric acid and an 
few dron7nf / « l 3 r hydrochloric acid and 30 c.c. water), a little starch paste, and a 
mate^r unLh if ^ i° f « ° f P otassmm * To this mixture a solution of bichro- 
off the solutiofonpn^M u lAl th e starch just indicates free iodine. In measuring 
oft the solution, one of Mohrs burettes, graduated to divisions each eaual to ten water 
ST'S!,? E " S *4 kaTCS notbin S to b? desired in vdumetric 
innn y '* If tbe solution of bichromate be made to contain 7-277 grains of the salt in 
a grata o7n hZ nrif T-* A, the Solution wil1 ,)e equivalent to one-thousandth of 
solved^in . mt of w».i 3 m ’L kC n,an . ner > if 7 ' 277 grammes of the salt be dis- 
acid (NO l An eon 1 1 ’ every CUbl ( C ';™ l "" c ' fre wiU represent a milligramme of nitric 
difference in the nnmhf T^* 6 ° f c J be . tm solutlon . must be similarly tested; the 
ffivethe number nf thn 1 ° ,^ ain Y bichromate solution required in the two cases will 
mitted to exver mLt Tf " N °^ pr6Sent in the fl uantity of water sub ’ 
grains and 56 trains’ or l’ rn ex f a . I f p , e ’ 1 S alloa of water be concentrated to 280 
^ , A , g . 10 l-oth of its bulk, be operated on, the amount employed renre- 
be Snltfplfed’C 25 it ‘'n,^ ° f a •» that, if the amount of RobZed 
corresnond to Z.h.v’ on r S‘' e the amount per gallon. If the burette divisions each 
ceSed water^ suhmSed t f’• ‘, he ^“? tity of N0 ^ c0 “ taine<1 in th « P»rtion of con- 
“ : U submitted to trial, will be given in milligrammes. The tubes when 
expl L $e of tl S Tout r r „ bUrS a in ‘“ s experiment when a sufficient space Ir ihe 
f ~uM, tt X islound^iie a“ 1 '' th ° f th ° of the tllbe be lrft 
