202 
W. A. HASWELL. 
development of the correspondiDg cavity in S tratiodrilu s, 
but what evidence I have before me seems to point rather to 
a schizocoelic origin than a blastocoelic derivation. However 
this may be, the layer which forms its splanchnic boundary is 
assuredly of the nature of a nucleated membrane. The 
presence of the nuclei, though there are wide intervals 
between them, definitely proves untenable the view that this 
layer can be looked upon as a cuticle or as a secretion from 
the cells of the enteric epithelium. The somatic layer also is in 
apart of its extent quite unmistakeably a membrane of similar 
character. Forming the ventral wall or floor of the body- 
cavity is a thin membrane of material similar to that which 
forms the splanchnic layer, and with similar flattened nuclei. 
On the dorsal side this membrane does not exist, and the 
nuclei, if represented at all, are represented by nuclei which 
lie embedded in protoplasm on the inner surfaces of the 
muscular fibres. 
To avoid the misconception that in this substance lying 
betweeu the epidermis and the enteric epithelium we have to 
do with a definite epithelium, I propose to use the term 
coelen chyme introduced by Salensky, and applied later by 
de Beauchamp ( 1 ) to a very similar tissue occurring in Dino- 
philus. The ccelenchyme is to be found in what is probably 
its most significant, because its most primitive, condition in 
the tail region (fig. 15 c. and figs. 2-6 c. c.). Here, as in the 
trunk, it forms a definite splanchnic layer surrounding the 
enteric epithelium of the intestine. But there is no distinct 
somatic layer. Instead there is on either side an irregular 
strand running forwards from the corresponding rectal gland, 
and connected with its fellow of the opposite side by a 
transverse strand, which is intimately united with the dorsal 
surface of the nerve-cord. In each of the lateral strands 
runs the caudal part of the corresponding third nephridiurn. 
From each lateral strand branches are given off which 
become continuous with the longitudinal muscular fibres, 
both dorsal and ventral — a nucleus occurring in each of these 
branches close to its termination in the muscular fibre. On 
