62 Animal Geography . [January, 
frages so exalted if it did not harmonise well with the 
phenomena of the variation and distribution of species. 
But there are undoubtedly difficulties in the way. A simul- 
taneous glaciation of the whole earth, preceded or followed 
by a peiiod like the Miocene, with a luxuriant vegetation 
penetrating at once almost to either Pole, implies a wonder- 
ful oscillation in the total available amount of heat at the 
earth’s surface. 
Another important cause has been the fluctuation in the 
distribution of land and water on the earth’s surface. It is 
true that on this subject exaggerated views have been enter- 
tained. Mr. Wallace does not consider it either justifiable 
or necessary to assume that — at any rate in recent geological 
periods — the great masses of land and of water have changed 
their respective positions ; but he holds that along the mar- 
gins of the continents great mutations may have taken 
place. Northern Asia and the Sahara are probably recent 
upheavals of shallow sea-bottoms. Borneo, Java, and 
Sumatra may probably have been united with further India, 
and again separated ; Australia may have very probably ex- 
tended to the great barrier reef, and the Antilles may have 
been part and parcel of the American continent. Nay, at a 
very early period Madagascar may have been connected 
with Africa, and subsequently with Ceylon and Southern 
India. 
Such changes would follow naturally from the conditions 
assumed in Mr. Belt’s hypothesis of the Glacial period ; but 
Mr. Wallace decidedly opposes the theory of a former direct 
communication either between Africa and South America or 
the latter and Australia. The occurrence of certain peculiar 
forms of life in countries mutually very remote, and their 
absence in intermediate regions, may, he shows, be explained 
much more naturally than by the assumption of direCt and 
special connection. Such species may have been once very 
widely diffused, but being worsted in the great struggle for 
existence they have been exterminated, save in remote 
islands and inaccessible districts. Such species it will be 
found are rarely powerfully armed. As a confirmation of 
this view, it may be noticed that the strongest and most 
formidable animals are found either in the large continents 
or in the islands recently dissevered from them, such as 
Sumatra or Java. The faunae of true oceanic islands, on the 
contrary, are characterised not merely by scantiness, but by 
the general feebleness and inoffensive character of the spe- 
cies. A very curious faCt, to which Mr. Wallace more than 
once refers, is the evident extinction of many large species 
