562 Ridley -—On Endemism and the Mutation Theory. 
and species are largely characters of the floral organs ; the struggle for 
existence is almost entirely among the seedlings and young plants in which 
these organs are not present. To take an example, is it conceivable that in 
Dillenia it can make any difference whether these leaves are acute or obtuse, 
or the petiole one inch or one and a half inches long ? These are the only 
characters that can show till the plants are at least ten years old, by which all 
that are going to die out will have done so.’ But the struggle for existence 
does not cease in any plant when it has developed out of the seedling stage. 
It is surely well known that the struggle continues throughout the whole life- 
history of the species. I need not, however, go into this, as plenty has been 
written to show this both by Darwin and later botanists. 
The lengthening of the petiole and acute or blunt apex of the leaf may 
have the utmost importance to the Dillenia in the relations of the leaf 
to sun and rain. But I will give one or two examples of ‘ infinitesimal 
variations ’ which have the greatest importance to the life and propagation 
of a plant. 
Metroxylon Sagus , Rottb., and M. Rumphii , Mart., the two Sago Palms, 
are very closely allied plants, but the latter is armed with spines especially 
on the leaf sheaths, the former is unarmed. In parts of Borneo it is 
impossible to cultivate successfully M. Sagus , because the wild pigs ( Sus 
barbatus ) attack and devour the young shoots as fast as they come up. 
M. Rumphii , guarded by its spines, is immune from the attacks of this 
animal. Here the development of spines protects the spiny Sago, which 
would otherwise be exterminated as the smooth Sago is. Another instance 
of a totally different character is shown in the case of the Macarangas, 
as described in my paper on Ants and Plants in the * Annals of Botany ’, 
vol. xxiv, p. 471, where it is shown that in certain species of the genus the 
young leaves are liable to so serious an attack by caterpillars that the plant 
may be severely injured, if not killed ; the very slight variation of the longer 
persistence of the bud sheaths for some days, and the development from the 
glands (common in most species of the genus) of food bodies attractive to 
ants, by inducing ants to take up their abode in the hollow stem, protect the 
plant from caterpillar attacks. 
Castilloa elastica in the Malay Peninsula is attacked by a beetle, 
Epepseotes luscus , the larva of which tunnels the stem, causing the destruc- 
tion of the tree. The insect can only escape through the scar of a fallen 
leaf, as it is the only part of the trunk not guarded by lacticiferous vessels 
through which it cannot pass. A comparatively slight thickening or harden- 
ing of the texture of this point would effectually stop the beetle from 
escaping, and render the tree invulnerable ; and from some such slight 
variation I found trees immune from the attack of this beetle, while others 
standing close by were destroyed. 
I cultivated in Singapore two kinds of Lilies, of which one, Lilium 
