286 
in that region: in fact the highest point of the coelome is here not 
directly under the base of the somite, but nearer the median line. 
This is the condition seen in all the posterior segments, and suggests 
that here the coelome never had and never will have the connection, 
for its line of growth appears to be directed laterally and on too low 
a level to meet the somite. 
3% Reconstruction in wax by Born’s method shows that the course 
of the anterior cardinal vein is not straight, but that, after running 
close behind the medullary canal in the region of the pharynx, it 
inclines slightly outwards until opposite the anterior end of the first 
somite. Here, by a sudden bend ventrally and laterally, it becomes 
rather a vessel of the somatopleure, and situated on a level between 
the base of the somite and top of the coelome. While in this posi- 
tion it interferes with the communication between the somite and 
coelome: Figs. 2 and 3_suggest that by its presence and growth it 
has pressed the sides of the communication together and so choked 
off the connecting cavity, thus causing the bend in the tissue which 
forms the base of the somite. This hypothesis is strengthened by the 
fact, that in my specimens the open communication is found most 
clearly in those in which the veins have been least developed. This 
relationship of the base of the somite, cardinal veins and coelome 
appears in the region of the first three (in one chick four) somites in 
all the sections I have examined. It seems to me that the condition 
is, however, a very transient one. In the region of the anterior end 
of the third somite, the largest part of the vein bends suddenly and 
runs down the somatopleure to join the Ductus Cuvieri; but behind 
this point the posterior cardinal vein continues for a short distance 
in the same position, and, although much smaller, seems to bear the 
same relation to the communication of the third somite, that the an- 
terior cardinal does to that of the first and second. In the region 
of the fourth somite it has become much smaller and altered its po- 
sition (Fig. 4), but here the communication probably seldom takes 
place. 
This suggestion, that the cardinal vein has much to do with the 
closing of the communication, can only be considered hypothetical, for, 
as is well known, the relative development of organs, especially in 
the early stages, is extremely variable and irregular, and thus many 
exceptions might arise. But that the coelome communicates originally 
with the cavity of the first, second, third and occasionally the fourth 
somites in the chick, may be considered well established. And fur- 
thermore the condition seen in the Amniota (reptiles excepted) seems 
