NOVEMBER. 269 



protected than the other. These varied last spring by 

 just two weeks in the growth of the foliage ; and diifered 

 by the same length of time in October ; one being golden 

 while the other was still green, and when leafless the 

 other was yet clothed in maroon-tinted leaves of great 

 beauty. The fruit of the two ripened at the same time. 



The effect of a drought, whether early or late, is also 

 to be considered. While I am not aware that any pro- 

 tracted period of dry weather prevented the leaf-bud 

 from maturing, it is true that the size of the leaf is 

 affected with many trees, and the differences in this re- 

 spect between a dry April and one that has had an abun- 

 dance of rain, is a matter of from one eighth to one fourth 

 in the size of the leaves. This is particularly noticeable 

 among oaks and chestnuts, unless they are grown in 

 permanently wet situations, as near springs or in low 

 meadows. 



We see, too, the effect of an early drought during the 

 following autumn, for the leaves fall earlier in the season 

 if they were checked in April by want of moisture ; but 

 a long drought, as is now so common in August or Sep- 

 tember, does not affect the leaves injuriously as they 

 freshen when rain does come or to any noticeable extent 

 so far as their falling is concerned. We had a test case 

 in 1886, when there occurred a protracted late summer 

 drought, yet the leaves remained upon the trees longer 

 than usual a fact not to be ascribed to absence of frost, 

 but to the vigor they received from a superabundance 

 of rain in April and early May, a vigor in nowise checked 

 by the low temperature of August 29, when frost formed 

 in damp situations. And well I remember the parched 

 and dusty summer of 1874. From May until September 

 scarcely a drop of rain fell. In August the leaves began 

 to fall, and the woods were bare by October 1. There 

 were no colored leaves in the forest save of such trees as 



