192 



CHEMICAL COMPOSITION OF RAIN. 



The results in the foregoing tallies are manifestly incom- 

 plete, and in other respects imperfect, so that, under ordinary 

 circumstances, one would prefer postponing their collation until 

 they could be made more complete ; but the circumstances are not 

 ordinary, and, as the inevstigation has nijw been discontinued, 

 with but little prospect of its resumption for some years to come, 

 I have thought it best to bring the results together in their 

 present condition, notwithstanding the obvious blemislies, so that 

 they may at least serve as a point of departure and as an indica- 

 tion of the needful precautions, should the work happen to be 

 resumed on some future occasion. Meanwhile, the very incom- 

 pleteness of the records precludes much discussion of t!ie figures 

 at the present stage. 



^^dlere possible, the rain water analyses have been sum- 

 marised into annual aggregates, and the summaries thus ob- 

 tained are as follows : — 



In four cases it has been found possible to make a conv 

 parison between summer and winter rains, in respect of their 

 nitrogen content, and these are tabulated below : — 



* Not including August and December, 191 1, aiidMay, igi2. 



t Not including December. 1911, and May and August, 1912. 



t Excluding one-fourth of an inch of rain during September, 1912. 



H Not including March. 1912. 



II Excluding one-fifth f>f an inch of rain during June, 191 1. 



§ Excluding one-tenth of an inch of rain during June, 1912. 



