EARLY ONTOGENETIC PHENOMENA IN MAMMALS. 83 
and premaxillary, the dentale, the angulare and the articulare, 
reveals itself by identical bony parts higher up in the scale of 
vertebrates. The general homology is even so close that we 
find no difficulty in comparing the elements of the skull and 
visceral arches of those bony fishes even in detail with the 
higher mammals and with man. 
It will at all events be necessary to consider most carefully 
whether we had not better drop the primary division above 
noted of the Vertebrates into Anamnia and Amniota, as this 
subdivision has not even contributed to help us to understand 
the amnion better, and has on the contrary kept apart certain 
classes which earlier naturalists, as Linnzeus and others, 
never thought of separating so widely. I expect that 
paleontologists will also contribute willingly to efface a 
separation based on a distinctive character which could never 
be applied to the objects of their research. Whereas at the 
same time the other anatomical differences between Reptilia 
and Amphibia, for example, break down in the case of very 
numerous fossil forms of very great importance. 
Having up to now discussed the fcetal envelopes and 
appendages that are primarily of ectodermic origin, we have 
now to consider those in which the entoderm is the primary 
constituent. These are the umbilical vesicle and the allan- 
tois, of which the latter in many cases has actually assumed 
the shape of a vascularised embryonic envelope of respiratory 
or nutritive significance in Sauropsida and in many mammals. 
2. The Umbilical Vesicle. 
The umbilical vesicle in mammals may be grouped accord- 
ing to some few modifications, of which we will have to discuss 
the respective value and genesis. The first is that which we 
find in man, monkeys, and Tarsius. In these mammals the 
umbilical vesicle from the very first remains smaller than the 
trophoblast, never fillmg the whole of it. We have seen on 
p. 44 how the rest of the trophoblast comes to be lined by 
