ON THE RAINFALL IN THE BRITISH ISLES. 443 
* Secondly.—That (with some exceptions) the time of year when the max- 
imum monthly amount occurs, changes with the variations in the mean annual 
rainfall ; stations where little falls haying their maximum in summer, but 
those where the fall is large having theirs in the winter-months. 
« Thirdly.—That, so far as the present investigation has proceeded, these 
laws hold good in all three districts into which we have supposed England and 
Wales to be divided. 
“ Fourthly.—That at stations where the rainfall is very large, the minimum 
monthly fall occurs about two months later in the year, and the maximum 
about six months later than at those of small amount; and that in the former 
case, these periods of extremes are more clearly defined than in the latter.” 
In Scotland a summer maximum is not generally found, but an autumn 
preponderance is noticed at the drier stations, and a very decided winter’ 
maximum at the wet stations. The month of minimum does not appear to 
vary much. 
In Ireland the returns are not sufficiently numerous to give very reliable 
results. 
It will be seen from the foregoing remarks, that not at all the stations are 
these rules observable ; were it so, a small number of returns would suffice 
to show them as well as a large. But the exceptions are few, and may be 
often arranged in groups—thus giving hope of discovering the causes of the 
changes we have been speaking of. This, however, is a part of the work 
which at present has scarcely been entered on, and it would be premature 
to attempt assigning any reasons for the phenomena we are noticing until 
they have been more thoroughly and minutely examined. That there are 
more causes than one seems certain. 
An attempt has been made to discover whether the same relative monthly 
values are found at the same station in all decennial periods; but very little 
has yet been done in this direction, though it is hoped it will be completed in 
the forthcoming year. It is likely that averages of a larger number of years 
than ten will give more satisfactory results than the present decennial ar- 
rangement ; but this, too, must be left for the future to prove. A correction 
will be necessary for the height of gauges above ground when the work of 
more minute examination commences, as that to some extent affects the 
monthly values. 
The whole investigation must be considered as still in its infancy, and some 
modification of the views here stated may be requisite as we advance. It 
seems, however, scarcely right to withhold what is so apparent, because 
we are not acquainted with every particular we wish for. Rather, should the 
publication of what is here offered, with a caution as to its being too freely 
depended on, either excite wholesome criticism, or in any way stimulate the 
work of collecting and properly examining rainfall records, much good will 
have been done to this branch of Meteorology. Meanwhile no pains must be 
spared in conducting, if possible, to a successful issue, the present investiga- 
tion into the monthly distribution of rain over these Islands. 
Some preliminary steps have been taken towards approximately determin- 
ing the correction applicable to such of the returns previously referred to as 
have been obtained from gauges at any height above the ground. It is 
known to many persons that the deficient amount of rain collected in gauges 
elevated above the ground varies with the time of year. Of course if the 
deficiency was constant, the monthly falls would bear the same relation to 
one another as if the gauge had been placed on the ground; but this is 
not the case ; the deficiency is not uniformly distributed throughout the year, 
2H 2 
