CIRCULATORY APPARATUS OF THE NEMERTEA. 63 
broadest in the tip of the snout, is still uninterrupted in the 
cerebral region, but becomes further backwards ever narrower, 
more so ventrally than dorsally, so that about slice 110 the 
two longitudinal muscular layers ventrally unite. In section 
160 the same takes place on the sides and dorsally, so that 
further backwards Amphiporus hastutus is in the main 
not to be distinguished from any other Hoplonemertean. 
In the layer of gelatinous stroma before the brain and 
between the two layers of vesicular cells are seen isolated 
single or groups of the same cells. These are, when isolated, 
round or ellipsoidal, but when in groups, like soap-bubbles, 
more polygonal; their size varied from fourteen to thirty 
micmm. Their contents are transparent, however, with a 
fine granulation, which apparently is protoplasm, and which 
sometimes, as in vegetable cells, lay round the nucleus, and 
sent threads to the walis of the cells, thus forming irregular 
star-shaped figures. I also saw cells in which the protoplasm 
was distinctly arranged against the inner side of the cell 
membrane. Then the nucleus also lay against the wall. These 
groups of cells diminished backwards, so that in the cerebral 
region still a few groups are visible, and still further backwards 
none occur. The more these groups disappear the more very 
small vessels become visible, which, occurring even in the tip 
of the snout, bear a highly primitive character. Sometimes, 
but rarely, they appear to be simple lumina in the gelatinous 
stroma; their wall is merely a modified hyaline gelatinous 
tissue, which absorbs more colouring matter than the sur- 
rounding. Sometimes, outside the hyaline basal layer, a few 
large nuclei are situated, exactly resembling those of the vesi- 
cular cells. Most frequently, however, these lumina are limited 
by some, often by a closed ring of vesicular cells (see fig. 70). 
Such a vessel seen in profile has the appearance of fig. 71. 
I was not able to trace these vessels. They seem to pass into 
one another, forming a reticulum. I have illustrated it in 
fig. 8. A transverse section, however, through the head has a 
totally different appearance from one through the head of 
Valencinia longirostris (see figs. 13 and 53). 
