Oil the Rain and Drainage -Waters at Rothamsted. 277 
percolating through it would bo loss than would otherwise be the 
case. It must also bo borne in mind that the total amounts of 
water retained by the different thicknesses of soil would be very 
<lifferent, and would vary in relation to one another at different 
times. It is to be regretted that our data do not enable us to 
estimate accurately the influence of these various circumstances. 
Unfortunately, too, the results obtained by the experimental 
drain-gauges of different depths are somewhat anomalous, and 
such as do not seem capable of satisfactory explanation by the 
aid of the various considerations above referred to. 
On looking at Table XIX. it will be seen that up to the 
end of 1874, the drainage from the 40-inch gauge was on the 
whole somewhat less than that from the 20-inch gauge, and that 
from the 60-inch gauge was distinctly less than that from the 
40-inch gauge. In November and December 1874 the walls sur- 
rounding the experimental soils were, as already mentioned, 
covered externally with cement, and then thickened by an addi- 
tional half-brick. Since this time the relative proportions of the 
drainage from the three gauges have been different. The 40-inch 
gauge has given on an average considerably more drainage-water 
than the 20-inch gauge, while the 60-inch gauge has given much 
less than the 40-inch, and nearly the same quantity as the 
20-inch gauge. The exact state of matters during the two 
periods of the experiment will be seen from the following 
Table (XX.) :— 
Table XX. — The Average Annual Eainfall and Drainage through 
20, 40, and 60 Inches of Soil, during the Periods 1871-74, 1875-80, 
and 1871-80. 
Rainfall. 
Inches. 
Drainage in Inches. 
Drainaj 
56 for 100 Rainfall. 
20-Inch 
Gauge. 
40-Inch 
Gauge. 
60-Inch 
Gauge. 
20-Inch 
Gauge- 
40-Inch 
Gauge. 
60-Inch 
Gauge. 
Four Years, 
1871- 
-74 
27-344 
9-683 
9-476 
7-753 
35-4 
34-7 
28-4 
Six 
1875- 
80 
34-189 
16-944 
18-544 
16-899 
49-G 
54-2 
49-4 
Tea 
1871- 
-80 
31-451 
14-040 
14-916 
13-241 
44-6 
47-4 
42-1 
On looking at the column for rainfall it will be observed that 
the last six years have been much wetter than the preceding 
four. We expect therefore to find a greater divergence be- 
tween the amounts of drainage from the three gauges during 
the first four years than during the last six. The considerable 
increase of rainfall in the latter period thus helps to explain 
