Gatty Marine Lahoraton/, St. Andrews. 41 



according to the Hue of section. Thereafter, the tissue 

 gr;idually merges into the h()ok-])ad witli its superficially 

 arranged glands, and so on thruuglnjut the region, the inner 

 or secoud;iry ridge appearing and disappearing in eacli 

 segment. 



The supporting tissue in the anterior region of Pomato- 

 cerus triqueter differs from that in the Sabellids. Just as 

 the nerve-cords leave the cephalic ganglia, and whilst still 

 connected by a long and strong commissure, no special 

 supporting tissue is visible. The long, narrow, hypodermic 

 cells of the dorsal w;dl (PI. IV. fig. 20, hpe.) are indeed of 

 great depth, especially in the middle line, so that when torn 

 tliey resemble fibres, Avhilst within the basement-membrane 

 aie only the thin circular muscular fibres and the dorsal 

 longitudinal muscles — as yet little developed. As tbe 

 opercular stalk leaves the body-wall of the region (PI. IV. 

 fig. 20, op.) its central areolar mass joins the other tissues 

 and may stiffen the parts, for as yet the fibres of the dorsal 

 longitudinal muscles are few\ Through this mass a bifid 

 nerve-trunk from the cephalic ganglia passes. The remark- 

 able thickness and the appearances of the hypocierm of the 

 region in this species would suggest the view that it may 

 more or less be connected with the function of the special 

 chordoid skeleton of other forms. In this respect the dorsal 

 differs essentially from the ventral hypoderm of the region, 

 which is richly glandular.- The muscular tissue at the base 

 of the stalk is reticulated in longitudinal section, as if the 

 sarcolemma formed a network ; indeed, reticulation of the 

 muscular fibres themselves would appear to occur, though 

 the trend of most at the base of the stalk is longitudinal. 



The projection of the opercular stalk causes asymmetry 

 of the body-wall and of the incipient dorsal longitudinal 

 muscles, for the muscle of the same side considerably in- 

 creases in size, probably in relation to the movements of the 

 stalk. The body-wall remains asymmetrical after the stalk 

 separates,' that side being less than the opposite one, in 

 which, moreover, the slits separating the branchiae first 

 appear. This asynypaetry subsequently disappears in front 

 when the filaments approach separation, but it is a marked 

 feature. Connective-tissue cells fill up the lateral space 

 wnthin the body-wall beyond the region of the .cephalic 

 ganglia, but these do not show sj)ecial chordoid structure. 

 Deeply stained nerve-cells surround the cords and the 

 transverse fibres between them. The enlarged base of each 

 ventral fiap of the thoracic jacket has connective-tissue cells 

 similar to those in the lateral region of the body, the flap 



