PAOFIC EXPLORATION: D. H. CAMPBELL 
435 
northern end of the group. The next and largest island, Hawaii, is still 
in process of formation. 
Hillebrand noted the extremely interesting fact that Kauai, although 
much smaller than the southern islands, Maui and Hawaii, and with 
much less range of elevation, is botanically much richer than either of 
those islands, and contains a much larger proportion of endemic species. 
It would seem then, that a study of these endemic plants and their 
distribution in the islands ought to throw much light upon the processes 
of species formation. 
Some of the factors concerned in this remarkable degree of endemism 
are evident — notably the extreme isolation of the islands; but the 
agencies that have acted in the islands themselves are not all so clear. 
The very large percentage of endemic species which, according to Hille- 
brand, comprises about 80% of the higher plants, suggested to him the 
possibility of the islands being a centre of special creative energy. 
It might be asked perhaps, whether, in newly formed areas, especially 
those of volcanic origin, there is some factor present which tends to 
produce variation in plants introduced to them. This is a topic which, 
as far as I know, has never received attention. It is notorious that 
volcanic soils are extremely fertile, and some of the richest floras of the 
world are found in volcanic regions. May it not be possible (this sug- 
gestion is offered with some difi&dence), that new elements in recent vol- 
canic soils may have an effect, stimulating or otherwise, upon newly 
introduced plants, causing changes which, fixed by time, result in new 
species. This is a topic which the numerous volcanic islands of the 
Pacific should help to elucidate, and one which it seems would be quite 
worth while investigating. 
This question was suggested by observations which I made a few 
years ago while collecting liverworts in the Malayan region. In Java, 
which is entirely volcanic in its formation, the liverworts, as well as 
other types of vegetation reach an astonishing development, both in 
number and variety, and a comparison of other regions where these 
plants abound, shows that most of them are volcanic — e.g., Hawaii, New 
Zealand, the Philippines, Sumatra, and Japan. On the other hand, 
the granitic formations of the Malay Peninsula and North Borneo, 
climatically quite similar to Japan and Sumatra, are noticeably deficient 
in liverworts. 
Third, the agencies by which plants reach such remote regions as the 
Hawaiian Islands, for example, have been repeatedly studied, but very 
much can still be done, and apparently the amount of accurate experi- 
mental work in this direction is not large. Hillebrand thinks a most 
