the bird." (Loc.cit. II. page 221.)

Darwin was clearly prepared to go further than Hooker in accounting for
the southern flora by dispersion from the north. Thus he says: "We must,
I suppose, admit that every yard of land has been successively covered
with a beech-forest between the Caucasus and Japan." ("More Letters",
II. page 9.) Hooker accounted for the dissevered condition of the
southern flora by geographical change, but this Darwin could not admit.
He suggested to Hooker that the Australian and Cape floras might have
had a point of connection through Abyssinia (Ibid. I. page 447.), an
idea which was promptly snuffed out. Similarly he remarked to Bentham
(1869): "I suppose you think that the Restiaceae, Proteaceae, etc., etc.
once extended over the whole world, leaving fragments in the south."
(Ibid. I. page 380.) Eventually he conjectured "that there must have
been a Tertiary Antarctic continent, from which various forms radiated
to the southern extremities of our present continents." ("Life and
Letters", III. page 231.) But characteristically he could not admit any
land connections and trusted to "floating ice for transporting seed."
("More Letters", I. page 116.) I am far from saying that this theory is
not deserving of serious attention, though there seems to be no positive
evidence to support it, and it immediately raises the difficulty how did
such a continent come to be stocked?

We must, however, agree with Hooker that the common origin of the
northern and southern floras must be referred to a remote past. That
Darwin had this in his mind at the time of the publication of the
"Origin" is clear from a letter to Hooker. "The view which I should have
looked at as perhaps most probable (though it hardly differs from yours)
is that the whole world during the Secondary ages was inhabited by
marsupials, araucarias (Mem.--Fossil wood of this nature in South
America), Banksia, etc.; and that these were supplanted and exterminated
in the greater area of the north, but were left alive in the south."
(Ibid. I. page 453.) Remembering that Araucaria, unlike Banksia, belongs
to the earlier Jurassic not to the angiospermous flora, this view is a
germinal idea of the widest generality.

The extraordinary congestion in species of the peninsulas of the Old
World points to the long-continued action of a migration southwards.
Each is in fact a cul-de-sac into which they have poured and from which
there is no escape. On the