On the Rain and Drainaye - Waters at Rothamsted. 257 
On the 11th of May, 1870, rain being anticipated, the rain 
ffausre was washed with distilled water at 11.30 A.M. : a collec- 
tion of rain was then made at 3 P.M., and a second at 4.30 P.M. 
These two samples of water were found to contain per million 
as follows : 
Total 
Solid 
Mutter. 
Carbon 
in 
Organic 
Matter. 
Nitrogen as 
Chlorine. 
Organic 
Matter. 
Am- 
monia. 
Nitrates 
and 
Nitrites. 
Total 
Nitrogen. 
Collection at 3 p.m. . . 
, , 4.30 P.M. 
40-8 
29-4 
0-93 
0-62 
0-18 
0-19 
1-07 
0-37 
0-18 
013 
1-43 
0-69 
1-0 
0-8 
The second collection of rain-water is seen to be far purer 
than the first, the earlier rain having removed from the atmo- 
sphere much of the ammonia, chlorides, and oi'ganic dust which 
it previously contained. 
In considering the results obtained by Dr. Frankland, it will 
be convenient to look in the first place at the general composi- 
tion of the rain-water, and afterwards at that of the dew and 
hoar-frost as shown by his analyses ; and then at the variations 
in the composition under various circumstances. 
Table XI. — The Maximum, Minimum, and Mean Amounts of certain 
Constituents in G9 Samples of Eain-Watee, in parts per Million.* 
Total 
Solid 
Matter. 
Carbon 
Nitrogen as 
in 
Organic 
Matter. 
Organic 
Matter. 
Am- 
monia. 
Nitrates 
and 
Nitrites. 
Total 
Nitrogen. 
Chlorine. 
Hardnif.t. 
Highest proportion 
Lowest proportion 
Mean, C9 samples 
85-8 
6-2 
33-1 
3-72 
0-21 
0-90 
0-66 
0-03 
0-19 
1-28 
0-04 
0-37 
0-44 
0-01 
0-14t 
1-94 
0-13 
0-70 
16-5 
0-0 
3-1 
160 
0-0 
4-7 
* It should he borne in mind that the whole of these numbers are simply 
arithmetical means; all the figures are therefore somewhat higher than they 
would have been if, as before, the quantity of raiutall which each analysis 
represents had been taken into account (see note, page 251). The latter plan was 
impossible in the present case, the samples collected during the course of a shower 
representing no definite amount of water. 
t This figure is the mean of 34 analyses. In 35 of the analyses of rain-water 
given by Dr. Frankland, the presence of nitric and nitrous acid is not mentioned. 
These acids were probably, however, in no case really absent ; they appear indeed 
to rank among the constituents invariably present in rain-water, though often 
VOL. XVII. — S. S. S ° 
