244 
LOCK : THE GROWTH OI 
It appears that the treatment had no effect snch as could 
be detected by the method employed. 
No great stress however is to be laid upon these experi¬ 
ments which were discontinued owing to the large number 
of uncontrollable factors. For example the moisture of the 
air inside the covering cylinder was not known, moreover 
the young shoots are intimately connected through the 
rhizomes with the other members of the clump to which 
they belong, and this is much too large to be subjected to 
experimental conditions. Taken together these experiments 
may be regarded as indicating that the effect of light upon 
the young halms themselves cannot cause any large propor¬ 
tion of the oscillations of growth observed. 
5 .—Effect of Changes in the Moisture of the Air. 
In the open air changes in intensity of light appeared to 
run closely, parallel with changes in the moisture of the air. 
Thus in full sunlight the temperature is higher and the air 
is proportionately further from its saturation point, whilst 
the darkest portions of many days were those during which 
rain was actually falling. It has been shown above that in 
the latter case growth is more rapid, whilst in the former it 
was slower, or even sometimes altogether ceased. Moreover 
it has already been shown that moisture has a very large 
effect upon the nature of the oscillations of growth from 
day to day. 
A series of observations were therefore made in order to 
discover whether a similar parallelism could be demonstrated 
between the oscillations of growth from hour to hour and 
changes in the moisture of the air. For this purpose hourly 
measurements were made of the growth of two halms of 
Dendrocalamus Nos. 9 and lOand of one halm of Gigantochloa 
No. 14. The respective heights of these culms at 5 P.M. on 
August 16 wereNo. 9, 337 cm., No. 10, 233 cm., and 
No. 14,161 cm. 
The measurements were carried out on August 17-19 
and the results are recorded in Table XVI. The curve of 
