34 DR. J. F. GEMMILL ON THE DEVELOPMENT OE 



diverticula of the enteron grow out, pockets from the epigastric ccelom extend outwards 

 along their aboral aspect, and these pockets bifurcate as the diverticula themselves 

 become divided into two. The adjacent walls of the epigastric and hypogastric 

 cavities now form a circular sheet with paired radial extensions, which, as it stretches 

 between the enteron and the aboral body-vs^all, may be called the epigastric mesentery. 

 Later, the circular part of this mesentery disappears as a continuous structure, being 

 broken up into very numerous strands of fibrous tissue which serve as dorsal suspensory 

 ligaments to the walls of the stomach, and leave passages, chiefly interradial in 

 position, through which the epigastric ccBlom communicates with the rest of the 

 perivisceral cavity. The so-called interradial cseca grow out among these fiibrous 

 strands, but do not extend beyond them. On the other hand, the paired radial 

 extensions of the mesentery above named do not become broken iip into fibres, but 

 remain as the boundary-walls of the epigastric ccelomic pockets, which in the adult 

 lie above the paired radial diverticula of the stomach (p. 40). 



On the left side in the young larva, as w^as stated above, the wall of the posterior 

 ccelom meets that of the hydroccele groove to form a mesentery, which, to begin with, 

 is crescent-shaped and extends dorso-ventrally, with the convexity directed backwards. 

 This mesentery is somewhat thickened, and it receives the radial pouches of the 

 hydroccele as they first grow out. These pouches, in further growth, keep as close as 

 they can to the left body-wall, and, accordingly, an angular space is left between the 

 hydroccele and the enteron, which is occupied by a leftward extension of the posterior 

 ccelom. This is the first indication of the tendency, afterwards very fully carried out, 

 for the oral (left larval) aspect of the enteron to be covered by a lining derived from 

 the posterior ccelom of the larva. 



The hydroccele (p. 28) becomes nipped ofl: from the bottom of the left lateral 

 diverticulum. The left mesentery then separates the posterior ccelom from the 

 remainder of this diverticulum and from the anterior ccelom. The mesentery in 

 question has a broad attachment to the wall of the enteron, and later this area of 

 attachment becomes greatly increased through secondary obliteration of part of the 

 left lateral diverticulum (p. 30). 



From a comparatively early stage (three or four days before fixation of the larva) 

 the tissue in question begins to be excavated by pockets from the posterior ccelom, 

 which, passing for a short distance towards the middle of the left side of the 

 larva, then elongate in the larval sagittal plane concentrically with the curve of the 

 posterior ccelom and its horns, but naturally forming parts of a much smaller circle. 

 So far as I can make out, the constant pockets are two in number and originate one 

 from the dorsal and one from the ventral horn of the posterior ccelom. That from 

 the dorsal horn comes off quite close to the tip of the horn in the neighbourhood 

 of the second radial pouch, while that from the ventral horn arises near the tip at a 

 time when the latter is just beginning to extend dorsalwards in front of the enteron 



