84 
FETCH : 
a^vailable. It has, for example, been recorded that if a 
” ligature ” be tied tightly round the stem in the early morning, 
it will be found to be quite slack at midday, and the expla- 
nation has been given that this is due to the shrinkage of the 
stem owing to the loss of water during the day by transpiration. 
Actual figures have been given, e.g., that a tree 36 inches in girth 
will decrease by half an inch between sunrise and midday (2). 
The kind of ligature employed has not been recorded, but it 
may be pointed out that if a piece of string were tied round the 
stem in the early morning when everything is damp it would 
naturally become slack, through drying, by midday on a 
dry day. 
It has further been stated that Hevea of the same age as 
those of the present account and at the same elevation, 20 
inches in girth, will show an increase of a quarter of an inch 
in girth after a heavy shower, as determined by ordinary 
estate measurements. 
Both these effects have been sought for unsuccessfully in 
the case of the trees whose measurements are recorded here. 
Measurements have been taken in the early morning, at mid- 
day, and in the evening, but they did not show any appreciable 
difference. Again, measurements have been taken after heavy 
rain which followed several days of fine weather, when rain 
fell on the day, or the next day after, the usual measurements 
had been recorded, but in no case was any increase observable. 
The average increase in girth at Peradeniya during the 
maximum period of growth is about 2 millimetres per week. 
A sudden increase of 6 millimetres would, therefore, equal 
three weeks’ growth. Moreover, if such increases really 
occurred, one might expect to find corresponding decreases 
during the dry weather, but the greatest decrease recorded 
only amounts to 3 millimetres in nine weeks, and it was 
never greater than 1 millimetre per week. In ordinary 
estate measurements, in which the trees are not measured rd 
a fixed line,, but only approximately at a height of 3 feet, it 
has been found that the possible error is fully half an inch. 
The initial girths of the trees and the increments for each year 
are given in the following table. The increments are tabulated 
separately for Januar y-March, the dry season, April-October, 
