75 
average of the surrounding 15 years. I found the method 
very efficient. But of course such a process eliminates 
every trace of the longer periods, be they accidental or 
in the climate. The present study does not therefore 
give any contribution to the determination of slower 
changes in the climate. Such a determination by means 
of growth-rings must not be impossible, but it will require 
a great mass of material and the utmost care in the 
discussion. 
8th, I am sorry to say that many of the tree-sections 
used for the present investigation are no longer in my 
possession. What is still in existence is now safely 
deposited in the Bofanical Laboratory of Groningen. 
Ladies and gentlemen. I am going to talk to you to 
night on the growth-rings of trees. How they may be 
considered to a large extent as natures own registers of 
at least some of the main elements of the weather in long 
past years and how they thus offer an occasion of in- 
vestigating whether or not some regularity in the recur- 
rence of these elements of the weather can be discovered. 
Could such a regularity really be discovered, the impor- 
tance of the fact could hardly be overestimated, for it is 
evident how then a basis would have been obtained for 
forecasting, in general traits and to a certain extent, the 
weather in coming years. 
My own investigations in the matter were made about 
the vears 1880—81. Though I spent quite a considerable 
time on them, Ï could not bring myself to publish my 
results, because I sorely felt how incomplete they were 
and because Ï hoped that, perhaps after some years, 
occasion would offer to obtain more satisfactory materials, 
which would lead to more reliable results. Î[ am now 
sorry that IÎ took this course. Had I then published my 
results, it would possibly have encouraged other men, in 
