90 
But just such a combination it might be interesting to 
find out. 
Suppose for instance that in English India, in the region 
where famines occur somewhat periodically, we could find 
out some tree or other, the growth of which depends 
approximately on the same conditions of the weather as 
those on which the success of the rice-crop depends. It 
does not seem to me so improbable that such a thing 
might be found. Well then, if it so happens that this 
tree reaches a considerable age, we would have the 
means of studying whether there is a regular recurrence 
in the return of the bad conditions for the rice-crop. 
The importance of such a regularity caused, years ago, 
several investigations to be made into the existence or 
non existence of such regularity. 
No conclusive results were reached. 
The very long and authentic records of trees might 
perhaps have furnished the means of attaining an end that 
could not be attained by the more direct but very imper- 
fect records of the past. 
Notes. 
I. The following table rests on data more extensive 
than those mainly used for the lecture. The cold winters 
were borrowed from Küppen, Zeitschr. Oestenr. Ges. 
für Meteor. 1881, Vol. 16, p. 186. 
The mild winters, cool and hot summers were borrowed 
from a work by Hahn on Sun spots. (See page 91). 
The numbers in brackets represent number of winters 
or summers. În computing the means, the weights have 
been taken proportional to these numbers. 
The only positive indication furnished by this table, is 
a slightly injurious effect of a cold winter, preceding the 
growth-season. 
