446 Messrs Purvis and Courtauld, The limitations, etc. 
Conclusions. 
The above experiments, therefore, suggest some doubts as to 
the accuracy of the copper-zinc couple method for estimating 
nitrates and nitrites in waters heavily charged with organic 
nitrogenous compounds. The ammonia produced thereby does 
not represent the reduction of the nitrates and nitrites only, but 
also, partly, the complex nitrogenous compounds which undergo 
decomposition under the conditions of the experiment. No doubt, 
the amount of ammonia produced by the couple from the organic 
substances is small. But when it is necessary to search for small 
quantities of nitrates, and such an event has occurred in the 
authors’ experiments with sewage ( Proc . Camb. Phil. Soc. vol. xiv. 
pt. IV. pp. 354 — 360), these small differences cannot be neglected. 
We desire to thank Miss Willcock, of the Physiological 
Laboratory, who kindly gave us the soluble albumin used in these 
experiments. 
