26 DE. J. F. GEMMILL ON THE DEVELOPMENT OF 



(1) Anterior Coelom and its Derivatives. 



As the anterior coelom expands, its lining epithelium for the most part decreases 

 in heio-ht in order to flatten out over the larger surface. In front the cavity of the 

 coelom extends into the three larval arms, and posteriorly it grows backward on either 

 side of the middle chamber in the form of left and right lateral diverticula, which 

 ' become evident a considerable time before the ccelom itself is separated from the 

 enteron. The anterior coelom, as a whole, now rests saddlewise on the middle 

 chamber, and in the course of time the parts forming the flaps of the saddle, 

 ^. e. the two diverticula, extend backwards so far that they completely overlap the 

 enteron on either side. 



An opening to the exterior — the hydropore — is acquired by the anterior coelom, on 

 the right side of the larva in the position already described (p. 14), at about the tenth 

 day of development. A short funnel-like invagination of the ectoderm meets and 

 fuses with a similar but longer pouch from the anterior coelom, rupture of the fused 

 walls afterwards taking place. The anterior coelom contributes much more than the 

 ectoderm to the length of the hydroporic canal. The hydropore is situated from the 

 first on the right side of the body, but the internal opening of its canal is, to begin 

 with, a little to the left of the mid-dorsal line. So far as I can make out, the dorsal 

 sac, or madreporic vesicle (right hydroccele of Macbride), is budded off' from the anterior 

 ccelom dorsally a little to the right of the middle line almost in the angle formed in that 

 region between the right lateral diverticulum and the coelom itself. It separates from 

 the anterior coelom shortly after the latter has become finally closed off" from the gut. 

 Its development is diflacult to follow, since it remains for only a very brief period in 

 open communication with its parent cavity. By this time the junction of the preoral 

 lobe with the body of the larva is indicated by the transverse dorsal furrow referred 

 to on p. 15. 



The following parts may now be distinguished as derivatives of the anterior coelom 

 of the larva: (1) the fore part of the ccelom forming the cavity of the preoral lobe 

 (preoral coelom) ; (2) the central part (axial coelom) receiving the internal opening of 

 the hydroporic canal and communicating behind with the two following ; (3) the right 

 lateral diverticulum, and (4) the left lateral diverticulum ; (5) the dorsal sac, or madre- 

 poric vesicle. The further development of each of these will now be described. 



Preoral Coelom. — The general history of the preoral coelom resembles that of the 

 corresponding coelom in Asferina (15) and Cribrella (18). It reaches its greatest 

 development in the later free-swimming stages, though it has begun to diminish a day 

 or two before fixation actually takes place. During the torsion and flexion of the 

 preoral lobe the cavity is constricted at the neck (pp. 16, 19) and becomes gradually 

 separated from the central part of the anterior coelom, which, being included within 

 the body of the Starfish at metamorphosis, remains as the axial sinus. The preoral 



