May 1901.] BOTANICAL SOCIETY OF EDINBURGH 71 



Effects of Weather on Tree Growth. By C. E. Hall. 



(Read 9th May 1901.) 



In " More Notes on Tree Measurement," published in 

 the " Transactions of the Botanical Society of Edinburgh " 

 in November 1899, an eftbrt was made to trace the effect 

 of variation in meteorological conditions on tree growth, 

 and the conclusion arrived at appeared to be that the most 

 essential factor in the growth of trees is to be found in the 

 rainfall. The present paper is an attempt to analyse the 

 figures published in the communication above referred to 

 in a more detailed manner, and along three different lines. 

 As two of these lines seem to have been based on false 

 foundations, they may be briefly dismissed ; the main 

 error of both being that the monthly tree growth taken 

 for comparison with weather factors was the plus or minus 

 growth over or under the average monthly 'percentage of 

 growth. Moreover, as tree-measuring day is the 12th of 

 each month, very rarely the 13th, and as therefore all tree- 

 growth records run from 12th to 12th, and as all meteoro- 

 logical records run from 1st to last day of each month, 

 and considering that possibly weather does not instantly 

 take effect and manifest itself in the girth of a tree, it was 

 determined, in the first line followed, to ignore the lack of 

 correspondence between the growth and weather periods. 



The unsatisfactory results of this first line impelled me 

 to work out weather tables from the 12th of one month to 

 the 11th of the succeeding month, and a comparison 

 of growth and weather on this second line was a great 

 improvement on the first line ; but when an examination 

 of some peculiar looking cases was instituted, some dis- 

 crepancies seemed to be quite unaccountable, and then it 

 was that I discovered the error in the tree-growth figures ; 

 and for the third attempt, which I shall now further 

 describe, I took for tree growth the plus or minus figures 

 over or under the average measured growth for each month. 

 I should say " half the measured growth," as the figures for 

 growth of evergreens and deciduous trees are made up of 

 the mean growths of three pairs of trees of each class. 



The basis then of this third attempt to ascertain the 

 effects of weather on tree growth is the accompanying 



