108 



TRANSACTIONS AND PROCEEDINGS OF THE [Sess. lv. 



and more irregular in the two former years than in 1890, and 

 the results are therefore not so precise and reliable, but still 

 they are quite sufficient to serve as a general test of the 

 results for the latter year. In the Table No. III. I have 

 reduced the results for the different periods to daily rates of 

 increase, expressed in ten-thousandths of an inch, so as to 

 make them comparable with each other. 



As the results for the different trees vary much, and to a 

 considerable extent neutralise each other in sum, it will be 

 better to consider each tree separately. 



Xo. 8. Abies Loioiana heha.\ed in 1888 and 1889 very much 

 as in 1890 : the rate rose to a maximum early in summer, 

 fell off" very much for a long period in midsummer, and rose to. 

 a second maximum well on in autumn. No. 92 of the same 

 species followed a quite similar course in 1888, the only year 

 in which it was under observation, and to a more marked 

 degree, as increase stopped entirely in the first fortnight of 

 August, and became very vigorous for six weeks thereafter. 



No. 91. Abies grandis behaved in 1888 much as in 1890. 

 In both years it tdok a fortnight of total rest early in summer. 

 In 1889 a sensible depression only was recorded about the 

 same date, but a total rest might have been shown by more 

 frequent observations. 



Table III. 



