290 REV. R. BARON ON THE 
been subjected to enormous modification. Such a very large 
amount of specifie differentiation seems to me to point in the 
clearest manner to long isolation. The antiquity of the island 
is also abundantly evidenced by the remarkable character of its 
fauna, a subject, however, whieh need not here be discussed. 
At what period the island was connected with the adjacent con- 
tinent it is impossible to state with certainty, but as Nummu- 
litie limestone occurs on a great part of the west coast of Mada- 
gascar, there seems to have been probably no land connection in 
Eocene times; and as the inroad of the higher forms of mammals 
into South Africa from the Euro-Asiatic continent took place, as 
Mr. Wallace shows, probably in later Miocene or early Pliocene 
times, Madagascar must have been cut off from the mainland at 
least not subsequent to the later Pliocene period, as the absence 
of such mammals in the island proves. This would allow time for 
the migration of the mammals to South Africa, which would not 
unlikely keep pace with the gradual lowering of the temperature 
going on in the northern hemisphere. This also would explain 
the existence of the * comparatively cold period" succeeded by 
“a warm period," during both of which, or some part of wbich, 
as Mr. Baker points out in one of the propositions given below, 
Madagasear must have been joined to the mainland. For it is 
now well kuown that in the northern hemisphere in Tertiary time 
there was a gradual lowering of the temperature from that of a 
tropical to a temperate or evén a cold climate. This being of . 
course reversed in the Southern hemisphere, we should have a 
cold period followed by a warm one. It seems probable, there- 
fore, that Madagascar was joined to the African continent during 
some part or parts or the whole of the Miocene (including Oligo- 
cene) and early Pliocene periods. 
In summing up the character of the flora of Madagascar, Mr, 
Baker lays down the following propositions :— 
1. “ The flora of the tropical zone throughout the world is 
remarkably homogeneous in its general character, and to this 
general rule Madagascar furnishes no marked exception. There 
is no well-marked plant-type largely developed in the island 
which is not found elsewhere, and none absent that one might 
à priori expect. 
2. “ About one in nine * of the genera are endemic ; but they 
* More correctly about one in siz.—R. B. 
