116 
1873. 
Rainfall in 
inches in 
5in. square 
receiver, 
3ft. from 
the ground. 
1873. 
Rainfall in 
inches in 
5in. square 
receiver, 
34ft. from 
the ground. 
1873. 
From 186 
Average 
fall of rain 
in inches 
for 6 yrs. in 
Sin. square 
receiver, 
3ft. from 
the groimd. 
8 to 1873. 
Average 
fall of rain 
in inches 
for 6 yrs. in 
Sin. square 
receiver, 
34ft. from 
the ground. 
January 
3-823 
2-799 
2-976 
2-130 
February 
0-543 
0-349 
2-209 
1-656 
March 
2-660 
2-142 
2-347 
1-846 
April 
0-641 
0-570 
2-162 
1-858 
May 
2-000 
1-750 
1-871 
1-583 
June 
2-880 
2-633 
2-557 
2-288 
July 
4-359 
3^793 
2-890 
2-566 
August 
4-176 
3-378 
1 2-797 
2-378 
September . . . 
2-188 
1-894 
1 3-868 
3-322 
October 
4-630 
3-804 
j 5-097 
4-228 
November ... 
2-207 
1-764 
2-851 
2-177 
December 
0-713 
0-525 
1 3-291 
2-760 
30-820 
25-401 
34-916 
i 
28-792 
I thought it might be interesting if the ratios were given 
of the excesses of rainfall measured at 3 feet from the 
ground over the amount measured at 34 feet above the 
ground. These ratios appear in the following table. It is 
curious to observe that the ratios of the total amounts that 
fell in these gauges, both for 1873 and for an average of 
6 years, are identical, the ratio of the difference between 
the fall in the upper gauge and that in the lower being 
0'824. And even in the heavy rainfall of 1872 it will be 
found that the ratio of the deficiency between the fall in 
the two gauges was nearly similar, being 0’855. But when 
the monthly ratios are examined, either for the year 1873 
or for the average rainfall in the gauges for 6 years, it will 
be seen that the greatest difference of fall happens in Janu- 
ary and the least in J une. The oscillations, too, of the dif- 
ferences between the fall of 1873 and the average fall of 
6 years is very marked. On the theory that the excess of 
rainfall in the lower gauge is due to the particles of invisible 
vapour in the air between it and the higher gauge coalescing 
with the falling rain drops, the table seems to show that in 
