s
doubtful. The Devonian lung-fish has characters which do not seem to
lead on to the Amphibia. The same general cause probably led many groups
to leave the water, or adapt themselves to living on land as well as in
water, and the abundant Dipoi or Dipneusts ("double-breathers") of the
Devonian lakes are one of the chief of these groups, which have
luckily left descendants to our time. The ancestors of the Amphibia
are generally sought amongst the Crossopterygii, a very large group of
fishes in Devonian times, with very few representatives to-day.

It is more profitable to investigate the process itself than to make a
precarious search for the actual fish, and, fortunately, this inquiry
is more hopeful. The remains that we find make it probable that the fish
left the water about the beginning of the Devonian or the end of the
Silurian. Now this period coincides with two circumstances which throw a
complete light on the step; one is the great rise of the land, catching
myriads of fishes in enclosed inland seas, and the other is the
appearance of formidable carnivores in the waters. As the seas
evaporated [*] and the great carnage proceeded, the land, which was
already covered with plants and inhabited by insects, offered a safe
retreat for such as could adopt it. Emigration to the land had been
going on for ages, as we shall see. Curious as it must seem to the
inexpert, the fishes, or some of them, were better prepared than most
other animals to leave the water. The chief requirement was a lung, or
interior bag, by which the air could be brought into close contact with
the absorbing blood vessels. Such a bag, broadly speaking, most of the
fishes possess in their floating-bladder: a bag of gas, by compressing
or expanding which they alter their specific gravity in the water. In
some fishes it is double; in some it is supplied with blood-vessels; in
some it is connected by a tube with the gullet, and therefore with the
atmosphere.

* It is now usually thought that the inland seas were the
theatre of the passage to land. I must point out, however,
that the wide distribution of our Dipneusts, in Australia,
tropical Africa, and South America, suggests that they were
marine though they now live in fresh water. But we shall see
that a continent united the three regions at one time, and
it may afford some explanation.

Thus we get very clear suggestions of the transition from water to land