XVI FLORA OF TASMANIA. 
atolls or barrier reefs, as the lives, and Keeling Island, contain few species, and 
those the same as grow on the nearest continents. In the Pacific Ocean, again, the groups of islands 
most remarkable for their ascertained number of very peculiar generic types are the Sandwich group, 
Galapagos, Juan Fernandez, Loochoo and Bonin, all of which are rising, and most have active vol- 
canos : those with the least amount of peculiarity are the Society group and Fijis, both of which are 
sinking. In the present state of our knowledge it is not safe to lay much stress on these apparent 
facts, especially as the New Hebrides and New Caledonia, which lie very close together, and both, I 
believe, contain much peculiarity, are in opposite geological conditions, the Hebrides rising and Cale- 
donia sinking ; and the Friendly* and Fiji groups, equally near one another, and with, I suspect, 
very similar vegetation, are also represented as being in opposite conditions. On the other hand, 
whole of the group including the Low Archipelago and the Society Islands, extending over more than 
2000 miles, I observe but one rising spot,f namely, Elizabeth Island, a mere speck of land, but 
which is the only known habitat of one of the most remarkable genera of CompositcB.X 
25. Many of the above facts in the general distribution of species cannot be wholly accounted 
for by the supposition that natural causes have dispersed them over such existing obstacles as seas, 
deserts, and mountain-chains ; moreover, some of these facts are opposed to the theory that the 
creation of existing species has taken place subsequent to the present distribution of climates, and 
of land and water, and to that of their dispersion having been effected by the now prevailing aquatic, 
atmospheric, and animal means of transport. 
Similar climates and countries, even when altogether favourably placed for receiving colonists 
from each other, and with conditions suitable to their reciprocal exchange, do not, as a rule, inter- 
change species. Causes now in operation will not account for the fact that only 200 of the New 
Zealand Flowering Plants are common to Australia, and still less for the contrasting one that the 
very commonest, most numerous, and universally distributed Australian genera and species, as 
Camarina, Eucalyptus, Acacia, Boronia, Helichrysum, Melaleuca, etc., and all the Australian Legu- 
minoscs (including a European genus and species), are absent from New Zealand. Causes now in 
operation cannot be made to account for a large assemblage of Flowering Plants characteristic of 
the Indian peninsula being also inhabitants of tropical Australia, while not one characteristic Aus- 
tralian genus has ever been found in the peninsula of India. Still less will these causes account for 
the presence of Antarctic and European species in the Alps of Tasmania and Victoria, or for the 
reappearance of Tasmanian genera on the isolated lofty mountain of Kina-Balou, in Borneo. 
These and a multitude of analogous facts have led to the study of two classes of agents, 
both of which may be reasonably supposed to have had a powerful effect in determining the distribu- 
tion of plants; these are changes of climates, and changes in the relative positions and elevations 
of land. 
26. Of these, that most easy of direct application is the effect of humidity in extending the 
* I find that there is a remarkable difference between the Floras of the New Hebrides and Caledonia on the 
mk han.l, and those ol tlu- Fiji islands and those to the east of them on the other. In the former, X«v, /.aland 
I he latter, almost exclusively Indian forms. The differences between the Floras of 
Fiji, Samoa, Tonga, Tahiti, and that of India, are in species and not in genera, and many species are common to all. 
f Mr. Darwin has bit Aurora Island (another of the group) uncoloured, on account of the doubtful evidence 
regarding it, whmh however is in favour of its being in the same condition as Elizabeth Island. From a list of 
* Mr. Dana, it appears to contain no peculiar plants. 
. p. 640. t. 23, 24. 
