Mar. 1892.] THE BOTANICAL SOCIETY OF EDINBURGH. 287 



Secondly, the two classes were by no means equally 

 affected in the same years. In 1888, when the Deciduous 

 class fell 4i inches below average, the Conifers were only 

 f inch down. On the other hand, in 1891 when the 

 latter fell 2h inches the Deciduous group all but maintained 

 their average. 



With the view of defining within narrower limits the 

 period when these depressions occurred, let us take the 

 girth-increases in half-seasonal periods instead of in years. 



In the year 1888 the Deciduous class fell 4 J inches 

 and the Coniferie only f inch below average. Taking 

 now the incidence upon the half-seasons in the same 

 year, it appears that in the first half the Deciduous 

 group fell If inch helotu its average of 10 inches, while 

 the Coniferse was f inch above their average of 11. 

 In the second half, on the other hand, both groups were 

 affected, falling 2 inches and If below their respective 

 averages of 12 and 9 inches. 



In 1891 the Coniferse fell 2| inches below their yearly 

 average ; the Deciduous trees, on the other hand, maintained 

 theirs. But on going into the matter, it comes out that in 

 the first half-season both groups suffered, falling 2h and 

 3 inches respectively hdoio average, and that while in the 

 second half-season the Coniferte only regained their average, 

 the Deciduous group actually rose 3 inches above theirs. 



The differences here seem far too marked to be accidental, 

 and unless they are due to inferior reliability in the 

 Pinaceae, owing to transplantation in some and to lack of 

 vigour in others, they indicate that the two classes may be 

 influenced in different degrees and ways by the same causes 

 of variation, and this is confirmed by my previous ob- 

 servations (Proc, 1888-89, p. 397). 



