52 Dr David Christison on the [sees. lt. 



substantial increase in girth did take place in deciduous 

 trees, as a whole, previous to the month of June, I resolved 

 to carry the investigation still further back, and to ascertain 

 whether an appreciable increase might not be traced even in 

 April. Four years' observations have proved that such an 

 increase does take place. The amount indeed is very small, 

 and to establish the fact it is essential to measure a consider- 

 able number of trees, because, even in a favourable season, 

 some do not increase in April at all, while in others a slight 

 decrease has been observed to occur, the aggregate decrease, 

 however, being always exceeded by the aggregate increase. 

 Thus, in April 1884 the aggregate increase of twenty-eight 

 deciduous trees, due to nineteen of their number, and after de- 

 ducting the apparent slight decrease in a few of the others, was 

 1'30 inch, or not much less than the May growth (1'85 inch); 

 similarly, in April 1885, the increase, due to fifteen of the 

 twenty-eight trees, was 1'35, actually greater than that for 

 May (1"05); on the other hand, in April 1886, when the 

 spring was unusually late, the increase, due to only seven of 

 the twenty-eight trees, was but 0'40 inch, while that of May 

 sprung up to 2'30 inches; lastly, in 1887, when the spring 

 again was very backward, the April increase, due to twelve 

 trees, was 0"65 inch, that of May being 2"90. It thus 

 appears that in an advanced spring the increase of girth of 

 deciduous trees in April may be equal, or even superior, to 

 that in May, while in a backward spring the April increase 

 may be very inferior to that of May. The greatest April 

 growth in any one tree was a quarter of an inch, in a 

 Hungary oak in 1885. 



The question now arose whether any increase could be 

 detected earlier than the month of April. I had already 

 ascertained by experiment on forty-eight deciduous and ever- 

 green trees that there was no appreciable difference between 

 measurements of their girth taken in October 1882 and again 

 in March 1883. This proved pretty conclusively that no 

 permanent increase in girth takes place earlier than April, but 

 bearing in mind the activity shown by many trees even as 

 early as February, as indicated by swelling of their buds and 

 expansion of their flowers, it seemed not improbable that 

 temporary increments might occur thus early in the season. 

 With the view of testing this point, therefore, I this year 



