nSTTERNAL TEMPERATURE OF LEAVES. 



281 



Number of Rainy Days. 



The average rainfall is given again for comparison, 1884- 

 1906— 



The truth of the reasoning of the above section can, however, 

 only be tested by further observation. At present it is only 

 put forward tentatively as a suggestion. 



It seems more certain from the evidence so far obtained that 

 there is a rough correlation between small rainfall and number 

 of growing species. This is seen in the very large number in 

 young leaf in February, the driest month of the year, and in 

 the almost complete absence of growth in October, the wettest 

 month. 



The matter may be put in another light by chosing a few 

 species in which the phenomenon is most marked and stating 

 the months in which they grow. The species are here chosen 

 for the striking way in which they exhibit this phenomenon. 

 They are those in which after a period of rest of varying dura- 

 tion the whole tree quickly, almost suddenly, becomes covered 

 with a new growth bearing leaves strikingly different in colour 

 from the mature leaves. In the following table the attention 

 will therefore be confined to a small number of species exhibit- 

 ing this periodicity phenomenon in its most marked form. 



A complication arises here from the fact that while young 

 and in vigorous growth a tree usually shows new growth much 

 more frequently than it does later in its life. It takes a few 

 years for it to settle down into the periodicity characteristic 

 of its species. Thus, in the following table there is growth 

 noted for certain months for a certain species which record 

 would not occur but for the fact that the irregular growth of a 



