27 
Decade 
Total Fall. 
Decade 
Differences 
from 
August. 
Cumulative 
Differences 
from 
August. 
1841-50. 
1851-60. 
1861-70. 
1871-80. 
1881-90. 
J uly. 
28*59 
26-51 
16 29 
30*44 
31*34 
August 
27*47 
30 82 
25-67 
25*25 
25.04 
September .. ., 
19*13 
22-66 
25-28 
33-55 
19*51 
October 
29-11 
20-66 
27-42 
27-06 
26-64 
July. 
-f 1*12 
— 4-31 
— 9-38 
4-5-19 
+ 6-30 
August 
— 
— 
— 
— 
— 
September (actual) 
— 8-34 
— 8-16 
— 0-39 
+ 8-30 
— 5*53 
,, (corrected) 
— 8 06 
— 7-89 
— 0-38 
-f 8-03 
— 5-35 
October 
4- 1-64 
— 10-16 
4-1*75 
-j- 1-81 
+ 1-60 
July. 
+ M2 
— 3*19 
— 12-57 
— 7*38 
— i *08 
August 
— 
— 
— 
— 
— 
September (actual) 
— 8*34 
— 16-50 
— 16-89 
— 8-59 
— 14*13 
,, (corrected) 
— 8-06 
— 15-95 
— 16-33 
— 8*30 
— 13-66 
October 
— 1-64 
— 8-52 
— 6*77 
— 4-96 
— 336 
Here we see: (1) Only in the Forties was the difference 
between the three wettest months less than after fifty years. 
(2) In the Fifties October actually fell over ten inches behind. 
(3) In the Sixties July fell more than nine (4) In the 
Seventies September exceeded August by over eight inches, 
as it had been about 8 behind in the forties and fifties; 
these are the “ corrected ” values. (5) August was not so nearly 
approached by July and October at the end of any decade after 
the first (6) Judged by single decades the wettest months 
were, in the respective decades, October, August, October, 
September, July. (7) Though August was ahead upon its 
cumulative fall at the close of each decade but the first, it was 
behind both July and October during the first 16 years, and 
again surpassed by both in the last five years, only regaining 
the first place over July in 1889. (8) Whereas the ten-year 
falls for August range only 5 78 inches, from 25*04 to 80*82, 
July has a range of 14*15, September of 14*42, and Ootober of 
8*45 inches. It is plain that the August rain-fall is the most 
certain. 
The above values show how mistaken it is to base average 
monthly results upon ten, twenty or even thirty years; fifty 
would hardly appear to be enough. The following table shows 
that similar caution is required even for the annual fall. 
