Ridley. — On Endemism and the Mutation Theory . 563 
auratum , had narrow, the other — I think L. croceum — slightly broader leaves. 
Rain falling on the latter was retained round the bud by the broader leaves, 
which formed a kind of cup, the buds were destroyed, and the plant failed to 
flower. In the other species, the narrow leaves did not meet at the edges, 
and through the space between the narrow leaves the rain-drop fell and the 
buds were uninjured, and the plants flowered. 
A most amusing passage occurs in one article, showing, I think, a want 
of careful thought by the author : c Can it for instance be supposed that the 
hereditary fasciation of the Cockscomb is of any use to that form ? ’ To this 
I reply it certainly can, for specimens that are not fasciated (which do not 
rarely appear) are worthless from a gardener’s point of view, and quickly 
find their way to the garden bonfire. 
A very common variation occurs in the form of variegated leaves, 
which are blotched or streaked with white. Now every gardener knows 
that this mutation is often very unstable, and the plant readily reverts 
to the original green-leaved form. There is a variegated form of Arundo 
Donax very commonly cultivated with leaves edged or otherwise marked 
with white ; when cultivated for some time the plant produces branches 
bearing typical green leaves. If these are not removed, in a few years 
the whole plant reverts to the original plain green colour. Here is a case 
of reversion of a mutation which Dr. Willis states does not exist. Let 
us compare this with the Aroid Aglaonema costatum. This plant has 
three forms — one with blackish green leaves mottled with white, one with 
dark green leaves with a central white midrib, and one with light green 
leaves with white spots. These three forms grow in limestone districts 
north of the Malay Peninsula. Unlike the Arundo they do not revert, 
but each variety reproduces the same form. No variety with plain green 
leaves has been seen. Both of the first two grow in the same area, the 
third farther north in southern Siam. Here are variations which keep 
true under any form of cultivation, and would be cited by Dr. Willis as 
proofs of his theory. But if so, it would be incumbent on him to show 
why the Aglaonema comes true in all its three forms, and why the Arundo 
does not. Here his theory completely breaks down. 
Now it is well known that it is in limestone districts that one always 
finds the largest number of plants with white variegated leaves. Near the 
limestone districts in Sarawak I went one day for a stroll and collected 
a single plant of everything I could find with variegated leaves. In an hour 
I had my arms quite full of variegated plants of many different orders. 
The reason for this is quite obscure to me, but it is clear that some advantage 
must be gained to the plant by this variegation, and that it is of so much 
importance to the plant that it is a permanently fixed mutation. Why are 
other plants similarly variegated not equally permanently fixed as a variation, 
as they should be by Dr. Willis’s theory? Another instance. The Tahan 
