Standardization of Well Water 121 



the sample was treated with fifty cubic centimeters of the alka- 

 line permanganate solution, which converts into ammonia cer- 

 tain forms of the nitrogen contained in any organic bodies which 

 may be present in the water. The amount of organic matter is 

 thus indicated, and consequently the purity of the water. Leff- 

 man gives .123 parts per million as the highest possible amount 

 of albuminoid ammonia allowable in pure water. It will be 

 noticed that samples 4, 8, 12, and 22 greatly exceed this amount ; 

 but, since both 4 and 22 are wells of exceptional depth, it is prol)- 

 able that this is not indicative of impurity in them. Deep wells 

 very often contain harmless ammonium compounds which increase 

 the apparent amount of ammonium. Samples 8 and 12, there- 

 fore, come under suspicion. It is worthy of note, also, that in 

 both 4 and 22 the oxygen-consuming power is correspondingly 

 high, showing that the organic matter was of vegetable, and there- 

 fore of harmless origin; also that in all four cases the ammonia 

 was given off slowly, indicating a slow decomposition of organic 

 matter, which is not so harmful as a more active decomposition. 

 The fact that in some cases the results obtained b}" adding the 

 permanganate at once are higher than when the results of the 

 two separate methods are combined, may be due to this — that 

 in the latter case, some of the combined nitrogen is driven off in 

 some other form before the permanganate is added. 



Closely related to ammonia, in its significance, is the presence 

 of nitrites and nitrates. Leffman, in his general standard, says 

 that uncontaminated water shows little if any indication of the 

 presence of nitrites.^ It will be noted that the results from this 

 analysis run from very faint traces to 1.23 parts per million. 

 Since every sample is assumed to be healthful, we may safely 

 conclude that well-water in the Granville vicinity may have as 

 high as 1.23 parts per million and yet be considered pure. 



There is a difference in the opposite direction with nitrates. 

 Leffman allows as high as 1.25 parts per million; none of the 

 samples analyzed, save two, gave more than .5 parts per million. 

 Many of them show but traces. Since in both the exceptions 

 (samples 2 and 10) there is no decided departure in other respects 

 from the average, it may be concluded that these excesses are 

 not suspicious. 



3 Ibid., p. 93. 



