430 
ALEEED GIBBS BOURNE. 
investigation has proved that this orifice leads into a sac which 
a complete series of sections have shown to be entirely closed. 
Fig. 4 shows the sac as seen when the pericardium is opened 
and one wall of the branchial vein removed. It is impossible to 
remove the whole of the branchial vein since one wall is closely 
adherent to the pericardial wall, with which is also fused one 
wall of the sac : the whole forms a very thin membrane (fig. 3, w, 
and fig. 5, w). The sac is lined inside by an epithelium. 
This does not form an even surface but is very irregular, 
dipping down into branched crypts ; the section chosen for 
fig. 5 passing through the orifice of the sac does not show 
these crypts. 
The epithelium is of very different thicknesses in different 
regions, every here and there occur patches of a much thickened 
epithelium, and at other places the cells become very small 
indeed. A large number of the cells, more especially in the 
thickened patches, present glandular contents which stain 
deeply. I am inclined to think that the greater number of 
the cells in the thickened patches, perhaps all of them, are 
glandular, but those only show which happen to present con- 
tents. The whole surface appears to be ciliated, even the 
gland-cells. This may be, however, due to imperfect pre- 
servation, that is to say there may be smaller cells between 
the glandular cells which are richly ciliated and these may 
cause the whole membrane to appear ciliated. The structure 
of a portion of a thickened patch is shown in fig. 6, where a is a 
cell with glandular contents, m is the refringent margin which 
the cells present, p the slight amount of pigment which exists 
among the bases of the cells. 
It is easy to understand how any injection might have 
ruptured the thin membrane dividing the lumen of the sac 
from that of the branchial vein, and so how Lacaze-Duthiers 
was led to the belief that here was a special communication 
between the exterior and the blood-vascular system. There 
is, however, no doubt that such does not exist. 
What is this sac? Does it represent any structure found 
in other members of the group of Ctenidiobranchiate Palliate 
