150 P. HERBERT CARPENTER. 
Ophiocoma le canal aquifére vient aboutir directement a 
Vextérieur, par un pore unique sans canal. Le canal du pore 
suit directement le canal aquifére, dont il n’est qu’un pro- 
longement qui se rencontre chez les individus adultes. Chez 
les Ophiothrix, les choses ont lieu comme chez lOphio- 
coma nigra.” But then he goes on to tell us how in 
Ophiothrix rosula it is easy to expose the heart or piriform 
gland by slitting up the envelope which surrounds it and the 
stone-canal ; and it is then ‘ impossible de ne pas voir la com- 
munication directe de la glande piriforme avec l’extérieur.” 
Farther on again he says, ‘Son conduit est extrémement 
apparent, et ainsi aucune erreur ne peut exister sur ce point.” 
But if the single pore-canal of the madreporite is the direct 
continuation of the stone-canal, how can the duct of the 
piriform gland open externally? Apostolidés is silent upon 
this point; and his figures! (like that given by Perrier of the 
corresponding parts in an Urchin’) merely show the termina- 
tion of both “ duct” and stone-canal on the inner side of the 
madreporite. There is no indication whatever that the former 
opens externally through the pore, although Apostolidés’ injec- 
tions have caused him to have no doubt whatever concerning 
the existence of this communication. He tells us that he has 
confirmed the results of his injections by the section method. 
But he does not figure a single section showing how the duct 
of the piriform gland opens externally ; and I much doubt 
whether he or any one else will ever do so. 
I feel therefore that we are fairly entitled to use our present 
knowledge of the vascular system in the Brachiate Hchi- 
noderms as a basis for criticising the descriptions of its 
peculiarities in the Urchins which have been published by 
Perrier and Koehler. The researches of these authors have 
been principally carried on by means of injections, and not by 
the section-method which has been chiefly employed in the 
investigation of the Brachiate Echinoderms. Lach plan has 
its advantages ; but it will be generally admitted that the 
1 ¢ Arch. de Zool. exp. et Gén.,’ vol. x, pl. viii, fig. 6, and Pl. ix, fig. 8. 
2 Thid., vol. iv, pl. xxiii, fig. 1. 
