Mar. 1892.] THE BOTANICAL SOCIETY OF EDINBURGH. 315 



34 far July. In the three Hungary Oaks, the most vigorous 

 growers in early spring of all my Deciduous trees, the same 

 tendency is shown, but at an earlier stage and in a considerably 

 less degree. In one the May growth slightly exceeds the 

 April growth, in another they are equal, in the third April 

 slightly exceeds May, and the general proportions are 8'7 

 and 9-5. Of the five British Oaks, only two show an actual 

 inferiority in June compared with May, but in N"o. 72, the 

 most reliable of all, the proportions are nearly equal ; that 

 of May is double its monthly average, and that of June 

 nearly a third below its monthly average. The general average 

 of the five British Oaks is Iv'T for May and 15-3 for June, 

 in strong contrast with the proportions for the thirty trees 

 in mass, which are 12 for May and 31 for June. 



In the single Q. 23al7(sf.r{sJm\e has only 16 p.c, while May 

 has 18 p.c, but it is not a very reliable tree. In the single 

 Q. rubra there is no actual inferiority, yet the tendency to 

 it is probably shown by May being one-third above average, 

 and June one-third below it. The tendency to halt in the 

 Oaks is emphasized by the sudden jump from the month of 

 depression to a very high proportion in the following one 

 which characterizes them all. The only tree of another 

 species which shows this tendency is the single Hawthorn, 

 in which the July proportion (22) is below that of June 

 (28) and of August (24-5) ; but it should be remembered that 

 in the trees which show no actual retardation of growth 

 in the Table, and particularly in sucli as have a nearly 

 equal increase in neighbouring months, a tendency to halt 

 might nevertheless come out if the observations were taken 

 more frequently than monthly. 



e. Proportion of Girth increase in the First and Second Half- 

 Seasons in Deciduous Trees, 



The division of the growing season into two periods of 

 three months each is the only practicable one, with 

 observations at monthly intervals ; but the periods of actual 

 growth are not necessarily equal in these two divisions. 

 Probably the period of growth is shorter in April tlrnn 

 September, at least in young trees, as the percentage in my 

 young group was only 1'5 in April, and was 50 in 



