GiAWF bamboos. 
225 
devise any method of recording the growth after this stage 
had been reached. 
In the later part of its growth, from Jnly 15 to August 1, 
one of the culms (No. 1) was loosely tied to the neighbour 
on which the growth was marked, in order to bring the apex 
sufficiently close. The attachment was made by a cloth band 
which slipped easily upon the smooth surface of the older 
bamboo so as not to interfere with growth. In all the other 
cases recorded the young halm grew steadily upwards parallel 
to an old trunk and at a convenient distance from it. The 
end of each series of measurements marks the point at which 
curvature set in and records could no longer be made accu¬ 
rately. A very much larger number of observations were 
begun than are here published, those halms which showed 
awkward curvatures at an early stage being then abandoned. 
None of these however showed phenomena which conflicted 
with those exhibited in the particular cases chosen to illus¬ 
trate this paper. 
X.—OBSER VA TIONS. 
1 .—Grand Period. 
Daily measurements were made for various periods of time 
of the growing halms of Dendrocalamus giganteus Munro, 
Gigantochloa aspera Kurs, and Bambusa spinosa Roxb. 
Dendrocalamus giganteus.— Halms Nos. 1,2,5, and 6 
belonged to a very extensive clump of this species on the 
west side of the pond, the oldest in the gardens and the 
parent by vegetative propagation of all the others. The 
tallest culms of this clump reached a length of from 30-35 
metres. Their circumference when fully grown was taken 
at a height of one metre above the ground ; that of No. 1 was 
57 cm., that of No. 2 was 61 cm. 
No. 3 (circumference 48*5 cm.) belonged to a large clump 
at a considerable distance from the first. This clump was at 
