STUDIES IN MAMMALIAN EMBRYOLOGY. 319 
proliferated connective and vasifactive tissue have set in. 
What is first noticed is a certain loosening of the tissue round 
about the spot where the blastocyst lies. Fig. 38 indicates 
this. In the right half of this figure we see that the decidual 
tissue adjacent against the blastocyst is undoubtedly less 
compact than that which occupies the extreme right of the 
figure. 
On examining the preparations more closely we find that the 
reason of this diminished compactness is most of all the change 
which has come about in the finest blood-vessels and in the 
vasifactive tissue, close to the spot where the embryo lies. In 
fig. 37 no lumen can yet be detected in those vessels ; in fig. 
38 lumina are present, not, however, in the bulk of the decidual 
tissue, but close to the embryo, thus creating that looser region 
just alluded to. Moreover, these blood-spaces have from the 
first something peculiar to themselves, and may not by any 
means be identified with ordinary capillaries in the rest of the 
organism. This peculiarity finds its expression in a special 
differentiation, which the endothelial lining of these spaces 
undergoes, often from the very moment of their origin. Instead 
of a flattened endothelium, which we might have expected from 
the state of things that is noticed in an earlier stage (fig. 37), 
the endothelium is massive, bulges out into the lumen of the 
blood-space, and gives to this space a very characteristic aspect. 
In sections it can easily be distinguished from a cut gland into 
the cavity of which, even when it is nearly obliterated, the 
boundary cells never thus protrude. Nor can we be mistaken 
and regard as blood-spaces what others would perhaps suggest 
might be those very glands in stages of retrogressive metamor- 
phosis. I have carefully examined hundreds of sections, and 
can say with full conviction that the stages of reduction of the 
glands can be all through easily distinguished from these 
vascular spaces. Sometimes even the vasifactive tissue in its 
rapid development was seen to take the place just before occu- 
pied by a gland tube, but then the remnants of the glandular 
epithelium and the newly developing vascular endothelium 
could still be distinguished with satisfactory certainty. 
