( 1082 ) 
cally insisted upon by Hint and others: the derivation of viviparous 
mammals from oviparous ancestors, which were provided with an 
egoshell and liad a free allantois. 
The Prototheria (Ornithodelphia), to the embryology of which Hur. 
has so largely contributed, are thus entirely out of the direct line 
of descent of all the other mammals. It is only natural that Hir 
should be a bit prejudiced on this particular point. He admits in his 
above-cited paper (p. 109 last paragraph) that the unilaminar blas- 
toderm of the Prototheria is unmistakeably the trophoblast, but denies 
that the cells situated internally to that in the region of the white 
yolk bed are the mother-cells of the embryonic knob. L feel inclined 
to believe that the footnote on p. 1079 and the misunderstanding there 
explained away, will make him reconsider this denial and agree 
with me that also here we ought to suspend our judgment till more 
material is available. 
I must finally point out that if my contentions on these last few 
pages concerning the phylogeny of the allantois may to some appear 
to be too provisional or unreserved I have just in hands and partly 
already in the press a description of the very early stages of Galeopi- 
thecus. In this mammal — the full description of which I hope will 
be published before the autumn — we find a representative of an 
order which is undeniably primitive, in the possession of develop- 
mental features that force us to conclude that in the very early 
stages it possesses a connective stalk (Haft- or Bauchstiel) between 
embryonal shield and trophoblast, whereas later on this stalk disap- 
pears in consequence of the development of the coelom and is 
gradually replaced in about the same situation by a free allantois, 
the origin of which may be traced to processes in the very matrix 
of the original connective stalk. This would be a direct argument 
from ontogeny favourable to my speculations and not explainable in 
any other way. L cannot however here do more than hint at it. 
This paper might perhaps yet have harboured some amount of 
refutation of certain objections to my theoretical views that were 
advanced by Mac Bripp in a paper on Aniphioxus in No. 215 of 
the Q. J. (vol. 54). I refrain from doing so because in that case 
there are no positive facts upon which to base a rejoinder, as in 
the case of Prof. Hir’s attack. Moreover, since it is patent that 
Mac Bripr (I. e. p. 332) has failed to understand my own views 
about the phylogeny of the allantois to such an extent that he can 
present it as follows: “along the stalk of connection between embryo 
