Groom. — On Dischidia rafflesiana (Wall). 233 
the pitchers of D. rafflesiana than those of any other species 
not possessing pitchers — that is they are at a stage of evolu- 
tion immediately preceding the pitcher-stage — and ants build 
nests under these leaves, bringing thither fragments of leaves, 
&c. : (3) Ants are constantly found in the pitchers in all the 
widely distant spots, in the Malayan region, at which the 
plant has been observed 1 : (4) The pitcher-form ensures that 
seclusion from light which ants love when nesting; (5) it has 
already been mentioned that probably the greater part of the 
solid matter found in the pitchers is conveyed thither by ants. 
The one objection to this view is that, as rain-water can get 
into some of the pitchers, the existence of the ants is im- 
perilled. But the truth is that ants constantly nest in much 
more dangerous positions as far as the risk of inundation is 
concerned. Wallich certainly found large black ants drowned 
in the pitchers, but observations of other observers show that 
this is an exceptional occurrence. It is no argument against 
the view to say that, as ants are so abundant in the tropics, it 
would be curious if they were absent from the pitchers. The 
very fact of the prevalence of ants in tropical regions, makes 
it possible for them to be an important factor in the evolution 
of the surrounding flora. 
It has been pointed out that the lips of the pitcher are 
sharply incurved. If a pitcher be filled with water and then 
reversed, only a little water escapes : in fact the pitcher is 
roughly constructed on the principle of the common ink-pot 
from which ink cannot be spilt. And this may be the 
rationale of the use of the incurved margins. But the latter 
would also provide a platform for the ants when the mouth is 
directed downwards, and a shade when the mouth points up- 
wards. However ants prefer the shade when nesting, and 
hence would possibly show a preference for pitchers the 
mouths of which were not directed upwards, especially as in 
such pitchers there is less danger of their being swamped. 
1 Singapore, Malacca, Borneo, Java, Billiton. It would be an extremely 
valuable piece of information if it could be ascertained if the same species of ant 
is always present. 
