THE DEVELOPMENT OF OPH.IOTHRIX FEAGILIS. 579 



increase in length and become finger-shaped ; they are at 

 first arranged in a straight line antero-posteriorly, and will 

 be numbered 1 to 5^ commencing with the most anterior. It 

 is a question of considerable interest where the stone-canal 

 enters the hydrocoele. The latter organ, as it grows, extends 

 forward beneath the anterior coelom and eventually reaches 

 beyond it, although the anterior coelom was originally 

 situated entirely in front of the hydrocoele. 



The stone-canal was thought by Bury (6) to open iuto the 

 hydrocoele between lobes 4 and 5. Bury regards the opening 

 of the stone-canal as a fixed point when comparing one class 

 of Echinoclerms with another, and imagines that " the inter- 

 radius in which the riog-canal closes" is the point which 

 varies. I confess that I cannot follow this reasoning. In 

 dealing with an organ like the hydrocoele, which at one 

 period of development has an identical form in all groups 

 of Echinoderms, viz. an open hoop with five projecting lobes, it 

 seems to me impossible to evade the conclusion that the 

 front and hind ends of the structure correspond throughout 

 the phylum, and hence the point where they meet so as to 

 "close the ring" must be identical throughout the group. 

 The stone-canal opens on the inner surface of the hoop or 

 ring, whilst the lobes project from its outer surface, and 

 hence it is quite conceivable that the position of the opening 

 of the stone-canal could shift without involving any such 

 moi'phological impossibility as jumping a lobe. But before 

 we have recourse to this modest hypothesis it would be well 

 to satisfy ourselves that as a matter of fact the point where 

 the stone-canal enters the hydrocoele is different in the 

 different groups. In Asterina gibbosa this point is situated 

 between lobes 1 and 2, and I have found that this is also true 

 of Ophiothrix fragilis, as the sagittal sections shown in 

 figs. 42 a and 42 6 clearly demonstrate. The first figure shows 

 the opening of the stone-canal and its oblique course back- 

 wards to join the anterior ccelom ; the second shows the pore- 

 canal opening to the exterior. This second figure shows how 

 fury's error arose, for he did not use sections, but based his 



