AMPHIBIANS—GOIN 441 
the egg and a tiny froglet hatches out. This is so, for example, with 
Breviceps pentheri, of British West Africa. In this species the eggs 
are laid in holes on land and there is no free larval stage at all; the 
developing embryo lacks many of the typical tadpole structures. 
The tail is quite large and is presumably used as a respiratory struc- 
ture, as it is in the genus Hleutherodactylus. The extreme microhylid 
life history is shown by the genus Hoplophryne of East Africa. 
Hoplophryne uluguruensis lays its eggs between the leaves of wild 
bananas or within the nodes of the stems of bamboos which have 
been split sufficiently to permit the entrance of this small and ex- 
ceedingly depressed frog. Small amounts of water are retained in 
the leaves of wild bananas, but its presence has not been determined 
for the internodal chambers of the bamboos. The eggs hatch into 
tadpoles which have become specialized, as is the case in certain 
hylids, for existence in these rather barren environments. They 
have apparently taken up the habit of eating frog eggs, perhaps of 
their own species, and the tadpoles are consequently modified. Super- 
ficially these modifications remind one of those found, for example, 
in Hyla brunnea. The teeth are reduced to the point of being en- 
tirely absent, and the tail, like that of Hyla brunnea, is long, slender, 
and whiplike. These modifications are, of course, apparently sec- 
ondary and in no sense imply close relationship. 
Phrynomeridae.—Apparently the African toads deposit their eggs 
in open water. The eggs hatch out into tadpoles which later meta- 
morphose much as do most microhylids. 
DISCUSSION 
To be somewhat anthropomorphic, it is evident that the amphib- 
ians are still, today, striving toward elimination of the open-water 
habitat for their eggs and early larvae, as were the primitive forms 
that gave rise to the higher vertebrates, in which the amnion is 
always present. This leads, of course, to the basic problem of what 
were the life-history modifications that made possible the development 
of the amniotes. 
As can be seen from the foregoing survey, the problems arising 
from the deposition of eggs in open water may be avoided in part or 
in whole in several ways. There may be an acceleration of develop- 
ment so that the eggs and larvae are not left for so long a period of 
time subject to the catastrophies that may befall them in open water. 
Then, also, there is the retention of the eggs in the body of the mother. 
This, of course, is possible only when preceded by internal fertiliza- 
tion. The latter, though, has developed at least three times without 
leading to the amniote egg, for the modern caecilians, most of the 
salamanders, and some of the frogs today practice this form of fer- 
tilization. Then there can be parental care, where the parent, instead 
