54 SIR ROBERT CHRISTISON AND DR CHRISTISON ON THE 



in May, a month earlier than leaf-shedding trees. This conclusion is amply con- 

 tinued by the measurements of the two succeeding years. In 1881, indeed, the 

 proof is not positive, as the first measurements did not take place till the end of 

 June ; but as 51 per cent, of the whole annual growth was accomplished by that 

 date, it is fair to conclude that a considerable proportion of the increase must 

 have taken place in May. In 1882 there is no room for doubt. The increment 

 till the end of that month actually exceeded the increment of any other month, 

 and the only question is whether a portion of that remarkable growth was 

 not due to April. Unfortunately, as no measurements were taken at the end of 

 that month, this point must remain doubtful. 



But the reliability of results obtained from so limited a number of trees and 

 species may justly be questioned. At all events, it may be held that, although 

 true of these species, they may not be true of evergreens in general. Fortunately, 

 however, these results are amply corroborated by observations on the larger 

 number of evergreen trees, first measured for monthly comparison in 1882. 

 The proportion of annual increment in these eighteen trees due to May was 34 

 per cent., almost identical with that of the selected six, which was 35 per cent. 



The limit of the growing season in evergreen trees is better ascertained at 

 the end than at the beginning. Of the six selected trees only one — the African 

 cedar — showed the slightest trace of increase in September, and that only in 

 one of the two years in which observations are available. The increment 

 recorded, moreover, was so slight as to come within the limit of probable 

 error. 



In August the proportionate growth seems to be much less in evergreen 

 than in deciduous trees. In August 1880 the increment of the six selected 

 evergreen trees was only 9 per cent, of the annual increase, while in the 

 deciduous group it was 27 per cent. In 1881 there was a greater equality, the 

 respective percentages being 31 and 34. But in 1882 that of the evergreens 

 again fell to 13, while the deciduous percentage reached 25. The results for 

 the latter year were confirmed by the observations on the larger number of 

 eighteen evergreen trees, whose proportionate growth for August was only 15 

 per cent, of the annual increase. 



On the whole, the conclusions to be drawn from all these observations are — 

 First, that in ordinary seasons the growth of wood in deciduous trees is mainly 

 confined to June, July, and August. In September it is scarcely appreciable. 

 In May however a small growth does take place, which in favourable seasons 

 may become of no insignificant amount. The Hungary oak not only grows 

 with exceptional vigour in May, but probably in favourable seasons makes a 

 start in April. Secondly, that evergreen trees as a class begin to grow probably 

 a month earlier than the deciduous group. They make substantial progress 

 in May, and some of them perhaps make a start in April. On the other 



