68 
however, to estimate in an easy and accurate manner the 
total amount of organic nitrogen, and also to make the 
determination of the organic carbon simpler and less compli- 
cated. There is no doubt at all in my mind, in the case of 
a sewage-polluted stream or well, that the determination of 
the carbonic acid gas given off (after driving off the free 
dissolved carbonic acid gas at 94° C.) on heating the water 
to 100° C. is a nost valuable portion of water examination, 
because that 2 ^>articular carbonic acid gas has been formed 
entirely from organic matters in a ^partially decomposed 
or gyutrescent condition. This organic matter is entirely 
lost in the course of evaporation as required by the “ com- 
bustion process,” and it certainly constitutes a large pro- 
portion of the total organic matter present in a sewage- 
polluted water, a fact which has been proved by my analyses 
of the Irwell water, and experiments made with it. I find 
that Professor J. W. Mallett, F.R.S. of the University of 
Virginia, mentions (in his report to the National Board of 
Health, Washington, 1880) the source of error; he saj^s : 
“ As regards the combustion process we find distinct con- 
firmation of the existence of the two forms of constant 
error which have been pointed out as affecting the Corpo- 
ration. The weaker the solution — or in other words the 
larger the quantity of water to be evaporated for a given 
amount of organic matter — the less is the amount of organic 
carbon contained, indicating relatively greater loss of this 
element. On the contrary, the weaker the solution, or the 
greater the quantity of water to be evaporated, the larger 
is the figure for organic nitrogen, indicating relatively 
greater gain of this element from the atmosphere. 
I conclude, therefore, that in many cases it will be neces- 
sary to determine the carbon corresponding to the organic 
matter volatile at 100° G. in the manner I point out in my 
previous paper on the subject. 
As one proof out of many which I could lay before you, I 
