On the Rain and Drainage - Waters at Rothamsted. 251 
ment of Science in 1854, and to tlie Report of the Association 
for that year we must refer for all details of this investigation.* 
A summary of the results will be found in Table V. 
Table V. — Niteogen existing as Ammonia in Eain "Water collected 
at Rothamsted 1853-4. 
1853. 
1854. 
Rainfall. 
Inches. 
Nitrogen as Ammonia. 
Rainfall. 
Inches. 
Nitrogen as Ammonia. 
Per Million 
of Kain. 
Per Acre, 
lbs. 
Per Million 
of Kiiiu. 
Per Acre, 
lbs. 
January .. 
2-034 
0-64 
0-30 
February . . 
0-949 
0-78 
0-17 
March 
2 '363 
i-'ig 
0-63 
0-514 
0-78 
0-09 
April 
2-999 
0-G7 
0-46 
0-498 
0-80 
0-09 
1-682 
110 
0-42 
4-384 
0-37 
0-38 
June 
3-395 
1-05 
0-80 
4-484 
0-77 
0-78 
August .. 
2-978 
0 69 
0-46 
September 
2-011 
0-01 
0-28 
October . . 
3-659 
0-57 
0-47 
November 
2-052 
0-06 
0-31 
December 
0-408 
1-3J 
0-12 
We have here during the first 12 months a total rainfall of 
29-014 inches, containing nitrogen in the form of ammonia 
equal to 5*20 lbs. per acre. In the whole 15 months over which 
the determinations range we have a rainfall of 34-41 inches, 
containing on an average 0'74 of nitrogen, as ammonia, per 
million of water.f 
An attempt was made to determine the small amount of 
nitrogen existing as nitric acid in some of the above rain-waters, 
but the methods then known did not prove sufficiently accurate 
for the purpose. 
The rain collected at Rothamsted during 1855 and 1856 was 
analysed by Professor J. T. Way ; he determined the quantity 
both of ammonia and nitric acid in mixed samples of water 
representing the rainfall of each month. The ammonia in the 
rain was determined by a method similar in principle to that 
* " On the Amounts of, and Methods of estimating Ammonia and Nitric Acid in 
Rain Water." By J. B. Lawes and Dr. J. H. Gilbert. Eeport of British Asso- 
ciation, 1854. 
t This average is not the mean of the fifteen monthly proportions of nitrogen 
per million given in the table, which would amount to 0 • 80, but is arrived at by 
dividing the total lbs. of nitrogen per acre contained in the fifteen months' rain 
by the total lbs. of rain per acre which fell in that period ; the figure thus arrived 
at truly represents the composition of the water, supposing the whole rainfall of 
the fifteen months had been mixed together. The same method will be employed 
in calculating general averages throughout this paper. 
