HAS THE DEEP OCEAN EVER BEEN LAND? 4iI 
The present geographical distribution of plants and animals 
has a very important bearing on the former distribution of land 
and sea ; for the land is a barrier to the migration of marine 
animals, and the ocean is a barrier to the migration of terres- 
trial animals and plants. Mr. Wallace has shown that the pre- 
sent distribution of the higher animals, which came into exis- 
tence in the tertiary era, can generally be well explained by the 
present distribution of land, but he allows that this will not hold 
good for the older groups. “The general conclusion,” he says, 
“we arrive at is that the causes that have led to the existing 
distribution of the genera and larger groups of terrestrial mol- 
lusca are so complex, and have acted through such long periods, 
that most of the barriers which limit the range of other terres- 
trial animals do not apply to them, although the species are, in 
most cases, strictly limited by them.”* And Mr. Starkie Gard- 
ner has pointed out that the eocene palms of Europe could hardly 
have reached tropical America by the present northern route, as 
they are not found there. 
In the distribution of reptiles, and of some birds, in Polyne- 
sia, we have evidence of the existence of a former continent. The 
Brush-turkeys, or Megapodes, are birds that are unable to fly, and 
yetthey arefound in Borneo, Celebes, the Philippine Islands, Austra- 
lia, New Guinea, New Caledonia, the Marian Islands, the Samoan Is- 
lands,and others inthe Pacific. Reptiles arewidelyspread through- 
out the islands of Polynesia, and we can only account for it by sup- 
posing a former land communication. Mr. Wallace, in his ‘Island 
Life,’ attempts to explain the fact by suggesting that reptiles 
have some unknown and exceptional powers of dispersal. But, 
if so, why is the phenomenon limited to Polynesia? And why 
should Mr. Wallace himself explain the small number of reptiles 
in Great Britain and Ireland by the supposition that they are 
unable to cross the English and Irish Channels? 
To give one more case. Dr. Gunther has drawn attention to 
the remarkable affinity between the fresh water fishes of Africa 
and those of South America. Two well-marked families, the 
Chromides and the Characinide, are found only in these two re- 
gions (with the exception of the genus Etroplus, of India), but 
the species and genera in each region are different, shewing long 
isolation. Dr. Gunther believes that all these fishes are the de- 
scendants of a common stock which long ago had its home ina 
region now submerged under the Atlantic Ocean. Many more 
instances could be given, but these are sufficient to show that 
the present distribution of animals is not altogether favourable 
to the hypothesis of the permanence of continental areas from 
very remote periods. 
If now we sum up the argument, we and that the geographi- 
cal distribution of the higher animals combines with the existence 
of tertiary rocks on oceanic volcanic islands to prove, with but 
little doubt, that the present continental areas have been perma- 
——— a — = ———- = = 
* Geographical Distribution of Animals, ii, p. 528 
