246 Proceedings of the Royal Irish Academy. 



other circumstances influencing the presence of those salts in ■\7ater 

 containing animal matters, will at once be evident ; and their absence, 

 unless accompanied by other indications of purity, cannot be relied on 

 as a proof of the freeclom from such contamination. 



Before I conclude, I wish to call attention to another fact, which 

 I have noticed in connexion with this subject, viz., the rapidity with 

 which nitrites are sometimes formed in waters contaminated with 

 sewage impurities. This is a subject of considerable importance in 

 an analytical point of view, as I shall endeavour briefly to explain. 



It is well known by those who have analysed potable waters, that 

 the method which chemists now principally employ to ascertain their 

 purity or otherwise is to determine the C[uantity of ammonia a given 

 amount of the water will yield on distillation, both before and after 

 the addition of a strongly alkaline solution of permanganate of potash. 

 The first obtained is termed the free, and the second the albuminoid 

 ammonia. The fonner is regarded as the representative of the nitro- 

 genous organic matters previously existing in the water, which have 

 undergone more or less decomposition, whilst the latter is produced 

 by the action of the alkaline permanganate on those substances still 

 present in the water. Consequently, the less of each that is furnished 

 by a sample of water when so treated, the purer organically is it re- 

 garded, and the safer, other circumstances being similar, would it be 

 for potable pui'poses. When lately analysing a sample of water that 

 had been contaminated with sewage, to ascertain the amount of such 

 pollution, which was afterwards the subject of an important legal in- 

 quiry, in my fir'st trial I found that the water yielded a quantity of 

 free ammonia which was ec[uivalent to 0.970 parts of a grain per 

 gallon, but, on repeating the detennination a few days afterwards, 

 it was discovered that it had fallen to 0.186 parts of a grain 

 for the same quantity of water, or to less than one-fiith of the 

 former amount ; whereas the quantity of albuminoid ammonia 

 yielded had slightly increased. This result as to the great de- 

 crease of free ammonia, which at first rather surprised me, I 

 ascertained was due to the formation of nitrites, which had been 

 developed to a large extent, in so short a time, at the expense of 

 the free ammonia. Such being the case, if the water had not been 

 examined till the date of the second analysis, and if the niti'ites had 

 not been taken into account, this water would have been regarded as 

 containing much less fi'ee ammonia than it did, and consequently 

 that the previous sewage contamination was less than it really was ; 

 this point is therefore one of some analytical importance. 



It is right for me to observe, in connexion with this latter fact, of 

 the decrease of fi-ee ammonia in waters by keeping, that long after I 

 had made that observation I met with, in the Chemical News for 

 March 2nd, 1877, a letter wiitten by Professor Pattison Muir of 

 Owen's College, in which he calls the attention of chemists to some 

 observations his brother had just made in the laboratory of the Uni- 

 versity at Sydney, in which he had noticed that the amount of fi-ee 



