l.M THE AMEEICAN MONTHLY [May, 



Moreover, different series of results compared with a varying 

 standard could only be compared among themselves by applying 

 vexatious and complicated corrections. The conventional stand- 

 ard was therefore fixed upon as the most convenient, all things 

 considered, that could be obtained. 



In estimating the previous contamination in terms of this stand- 

 ard. Dr. Frankland points out that it is necessary to bear in 

 mind that rain-water contains nitrogen in the form of ammonia, 

 nitrites, and nitrates. According to the analysis of twenty-one 

 samples collected at Rothamsted, the amount of nitrogen in these 

 forms was 0.047 P'^'t^s in 100,000 parts of water. The number 

 0.032 as obtained from a more limited number of analytical de- 

 terminations was, however, fixed upon for reasons wliich are 

 made clear in the Sixth Report of the Rivers Pollution Commis- 

 sion, but which need not be specially considered here. 



In arriving at the amount of the previous sewage or animal 

 contamination in any given sample, the amount of nitrogen in 

 the form of ammonia, nitrites, and nitrates is first ascertained by 

 the ordinary processes, and from their sum the number 0.032 — 

 representing the nitrogen of the rainfall in these forms — taken; 

 the remainder represents the nitrogen which has been derived 

 from any oxidized animal matters with which the water has been 

 in contact. For instance, a water containing 0.326 parts of nitro- 

 gen in the form of ammonia, nitrites, and nitrates has obtained 

 (0.326 — 0.032) = 0.294 parts from animal matters. According 

 to the assumed standard of comparison, this amount of combined 

 nitrogen is contained in 2.940 parts of average London sewage ; 

 hence such a sample of water is said to exhibit 2,940 parts of 

 previous sewage or animal contamination in 100,000 parts ; or 

 100,000 pounds of the water have contained at some time an 

 amount of organic animal matter equal to that found in 2,940 

 pounds of the assumed London sewage. 



With regard to the other principal determinations of the com- 

 bustion method, the weight of opinion at the present time ap- 

 parently is that the attempt to distinguish between the organic 

 carbon and the organic nitrogen is only partially successful. The 

 determination of the total organic nitrogen is, however, by itself a 

 valuable one, which would without doubt be generally used in 

 place of the albuminoid ammonia determination of the Wanklyn 

 process, if it were not for the difficulty of working the method. 



The detail of the more important phases of Dr. Frankland's 

 system of analysis is given in the foregoing somewhat at length, 

 because, all things considered, it may be taken as the most elabo- 

 rate attempt to exhibit the natural history of water by chemical 

 means only that has ever been made. That it has failed to do all 

 that was hoped for it is only due to the complexity of the subject, 

 which does not admit of a complete understanding from any one 

 point of view, but which must be approached from at least four 

 sides — namely, the environmental, physical, chemical, and micro- 



