IN CEYLON. 
437 
tree was observed at Gangaruwa, Peradeniya, in flower before 
leaf-fail had commenced, a condition also noticed in the 
Passara district in 1902. 
At Henaratgoda on 15th February, 1902, the majority of 
the trees had shed all leaves and flowers. 
In the Gampola district on the 5th February, 1902, many 
trees had finished flowering and the fruits were setting. 
In the Passara district, where one gets less south-west 
monsoon rain and where from April to August and Septem¬ 
ber it is dry and hot and from September to March the 
rains are heavy, the trees behave differently. 
Between Passara and Lunugalla at an elevation of 2,000 to 
3,000 feet, on the 25th May, 1901, some of the trees were 
quite leafless, and others in partial new foliage ; very few 
had a good output of leaf on that date. In several specimens 
the flowers were produced long after the trees had regained 
complete new foliage. 
Variation in Ceylon and Java . 
Now we can consider the variation which occurs in 
climates so dissimilar as Ceylon and Java—two tropical 
islands near to but on opposite sides of the equator. I am 
indebted to Dr. Treub for much information respecting the 
periodicities of certain plants in Java, and the comparison 
obtained is interesting in many ways, particularly as details 
of some of the periodicities have been obtained at Buitenzorg 
and Peradeniya. 
Peradeniya and Buitenzorg . 
The best example is perhaps to be seen in trees of Bombax 
malabaricum. At Peradeniya the fall of old leaf is very 
conspicuous during December; the leafless phase extends 
from December to April or May, the flowers and fruits 
appear in January and February, and the burst of new leaf 
occurs in April and May. Plants of this species rarely 
possess any old leaf during January and very few put on new 
8 (,9)05 (11) 
