EARLY ONTOGENETIC PHENOMENA IN MAMMALS. 15 
balance in favour of choosing the first eventuality. Principally 
because only in case these have been the actual phylogenetic 
steps can we conceive that the amnion on its very first 
appearance was of immediate significance to the embryo, as a 
sort of water-cushion, shielding the embryonic rudiment 
—already at its very earliest appearance—from external 
pressure or mechanical insult. We have now seen that not 
only does the amnion appear as a closed sac from the 
very earliest in very numerous cases in different orders of 
mammals (to the lst already given the monkeys and man 
should yet be added), but that in this case an explanation of 
its earliest origin is not far to seek. We have indeed 
admitted that the trophoblast is an early larval envelope by 
the presence of which a chorion or serous membrane is pre- 
destined to make its appearance sooner or later. From this 
larval envelope 
(Fig. 129)—the embryonic ectoderm has to become segregated 
also in the case of Nemertea and Gephyrea 
in one way or another, as also the amnion is being originated 
in different ways. We know from the examples we possess 
amongst Nemertea of the Pilidium larva and the Desor’s larva 
that at one time this segregation takes the form of a splitting 
process (Desor’s larva) when (as is the case in the dorsal 
plate of that larva which I have formerly [’85, Figs. 53a, 95] 
described) the plate of future embryonic ectoderm provisionally 
remains attached to the larval envelope along a circular line of 
attachment much in the same way as we see the embryonic ecto- 
derm of the hedgehog attached to the trophoblast (Fig. 37) 
with the closed amnion-cavity above it; whereas at another 
time (the lateral plates of the Desor larva or the embryonic 
plates of the Pilidium) the separation of the definite ecto- 
derm from the larval layer takes place by a process of 
invagination, during which that portion that is destined to 
become the outer wall of the embryo sinks away from the 
level of the larval surface into the cavity enclosed by that 
surface and develops further in this more sheltered position. 
In this latter case a circular fold ensures the continuity 
between larval and definite ectoderm, and only when the 
