GIANT BAMBOOS. 
251 
little to do -witb. the changes of growth.* On the Qther hand 
observations taken hourly in conjunction with readings of the 
psychrometer gave evidence of amarked parallelism between 
the two curves, whilst the curve of temperature had an almost 
exactly opposite character. 
In this connection it is to be noted that in the early part of 
the period of growth when these observations were made, 
drops of water were excreted from the tips of the sheathing 
leaves in considerable quantities during the whole night, 
the night being regularly the period of most rapid growth. 
This rapid excretion of water has been noticed by Molisch.t 
This observer found however that it was impossible to make 
records of the pressures normally existing in the Btems owing 
to the remarkable disturbing influence exerted when a 
manometer was introduced. 
During the hotter and drier parts of the day a slight reduc¬ 
tion in length was sometimes observable. This was always 
small in amount, and usually only just to be detected by the 
method of measurement employed. It never amounted to 
more than 2 or 3 mm. per hour. 
It isthus clearly established that changes of moisture are the 
chief cause of the observed oscillations in the rate of growth of 
the culms of Dendrocalamus. The temperature in the shade 
beneath the clump, where the young growing culms are 
protected during the early part of their growth, shows only 
slight and slow changes. On the other hand when the sun 
is brightly shining the older members of the clump are 
exposed to a much higher temperature, and this, in 
pect the periodic changes in illumination must co-operate with the 
parallel changes of moisture in inducing corresponding changes in t 
of transpiration of the full grown calms, and may thus e 
able indirect influence upon the rate of growth of tl 
i. Jard. Bait., 2nd Sapplei 
p. 30 et. seq. ■ 
