IX. Genetic Relations to other Echinoderms. 1915 401 



from oral to apical along the straight plane of the stone-canal. 

 Hence it did not change its interradius. 



As was shown in "What is an Echinoderm?" these movements 

 are precisely those which would naturally take place on the hypothesis 

 of the origin of Asteroids from such a Pelmatozoon as JSdrtoaster. 

 That hypothesis makes these movements very simple and inevitable, 

 but the converse conclusion does not necessarily follow : the con- 

 clusion, namely, that because these movements must have taken 

 place, therefore the ancestor of the Asteroids was an Edrioasteroid. 

 All that is claimed is that the facts thus far are fully consistent with 

 such a conclusion. 



We turn now to the next difficulty : the position of the plane 

 of closure of the hydrocoel. As already explained, this plane is 

 placed by MacBride and Gemmill between their rays I and V, i.e. in 

 the Asterid plane of Cuenot (my interradius 1V/V). 



In the primitive Pelmatozoon, so far as can be inferred from the 

 embryology of Antedon and the anatomy of early forms, the closure 

 of the hydrocoel was in what I have termed the M plane (see Treatise 

 on Zoology, 1 900, p. 20). This corresponds with MacBride and Gemmill' s 

 interradius I/II (my I/V). 



It is important to notice here that in an Asteroid larva (Bipinnaria 

 asterigera) described by Bury, the closure does still take place in the 

 M plane (1895, Quart. Journ. Micr. Sci„ vol. 38, p. 65). 



Apart from the hydrocoel and its extensions, the food-grooves of 

 Pelmatozoa (which are the initiators of the rays) are bilaterally 

 symmetrical about the M plane. Therefore we are not surprised to 

 find that in Asteroids this same plane is that in which the brachiolarian 

 notch occurs and closes, and that the rudimentary rays (apart from 

 the hydrocoel) are similarly bilaterally symmetrical about that plane 

 (Fig. 2). Other relations of this plane have already been 

 mentioned (see Gemmill, 1914, Phil. Trans., p. 277). 



The M plane, therefore, is as fundamentally important a plane for 

 Asteroidea as it is for Pelmatozoa. The Asterid plane is a plane of 

 superimposed symmetry, due to the migration (a) of the anus, (b) of 

 the hydrocoel-closure. One speaks of the "migration" of the 

 hydrocoel-closure because it cannot be imagined that the closure 

 takes place in a different part of the hydrocircus itself. If we 

 postulate any homology whatever with Pelmatozoa, we can only 

 explain the difference of position by supposing that the whole 

 hydrocoel has shifted round. The extent of the shifting would be 

 one-fifth of a circle, i.e. 72°; and the direction of the shifting, as 

 viewed from the oral face, would be contrasolar. Such shifting 

 need not involve the hydropore, with which the hydrocoel has only 

 a secondary connection. 



Now we know very well that, in those Asteroids where the 

 hydrocoel-closure is in the Asterid plane, just such a shifting or 

 torsion does take place in the early stages of metamorphosis. To 

 quote Gemmill (1914, Phil. Trans., p. 251), "the disc undergoes 

 torsion through about 75° in the (starfish) horizontal plane, counter- 

 clockwise as viewed from the sucker." It must be remembered that 

 the development of the lobes of the hydrocoel is quite independent 



