THE STAEFISH SOLASTEE ENDECA. 31 



The Stone-Canal. — The stone-canal takes origin as a groove beginning on the hydro- 

 coele, opposite the origin of radial pouch II and passing to the right on the posterior 

 wall of the axial ccelora. The outer end of this groove next seems, so far as I could 

 make out, to leave the surface, and burrowing a way underneath the epithelium in the 

 same direction emerges again in the axial ccelom close to the inner opening of the 

 hydroporic canal. The final conversion of the original groove into a canal takes place 

 about the same time as the conversion into a canal of the dorsal end of the hydroccele 

 groove itself, i. e., a day or two after fixation. By this time also the opening of 

 the stone-canal into the hydroccele has become shifted to its permanent position in the 

 interspace between radial pouches I and II of the hydroccele. 



There is still divergence of opinion as to whether in adult Starfish the stone-canal 

 and the axial sinus communicate with one another or are completely separated. How- 

 ever, the weight of opinion and evidence is strongly gathering in favour of the latter 

 view (Macbride, 15, p. 351 ; 16, p. 441 ; Durham, 6, p. 330 ; Goto, 9, p. 273 ; Perrier, 

 20, p. 1146 ; Hamann, 14). I have given elsewhere in detail the results of an exami- 

 nation of a series of sections through the madreporic interradius of a large adult 

 S. endeca (8 a). The pore-canals number about four hundred and may roughly be 

 divided into a set occupying the middle area of the madreporite and a set around the 

 margin, the two sets being approximately equal in total number. The canals of the 

 middle set join in small groups to form collecting-spaces opening into a central 

 chamber within the madreporite, which I have called the ampullary part of the axial 

 sinus and which is continued directly into the stone-canal. The marginal set also join 

 a system of collecting-spaces which, on the one hand, lead into the aboral end of the 

 axial sinus and, on the other, are connected by a number of apertures with the central 

 chamber mentioned above. Taking into account both development and adult structure, 

 none of the pore-canals should, strictly speaking, be described as opening into the stone- 

 canal, and the term ampullary part of axial sinus should be substituted for ampulla of 

 stone-canal. 



Perihcemal Pouch IXjI. — The last derivative of the anterior ccelom that falls to be 

 referred to here is the pouch forming that segment of the external oral circular sinus 

 which belongs to interradius IX/I. The origin of the segment in question is difficult 

 to follow, but I believe I have traced it to an outgrowth from the axial coelom 

 immediately in front of the first radial pouch of the hydroccele occurring towards the 

 end of the free-swimming stage. Much later, in one instance, a small outgrowth 

 from near the tip of the (larval) ventral horn of the posterior coelom came to open into 

 the cavity of this perihsemal segment, and thus presumably to contribute a little to 

 its extension. I could not find that this occurred in other cases, but if it does it 

 might go some way towards reconciling Solaster on the one hand, and Asterina and 

 Cribrella on the other, in regard to the mode of origin of their oral perihsemal 

 pouches. See page 35. 



