AND ITS RELATIONS TO RECENT COMATULA. 191 



together with Glenotremites and the typical Comatulce, in the 

 single genus Antedon ; while Zittel*, using Antedon as equivalent 

 to Comatula in its older meaning, separates off certain well-marked 

 recent types as subgenera, and assigns the same position to So- 

 lanocrinus, with which he groups Comaster and Glenotremites. I 

 have shown elsewhere f, however, that our scanty knowledge of 

 the organization of Comaster (supposing Goldfuss to have been 

 accurate) is sufficient to show us that it is a very peculiar form. 

 There are many and striking differences between it and other 

 recent Crinoids, to which Solanocrinus is much more closely allied 

 than to Comaster. Schliiter % doubts whether the mere fact that 

 the embryonic basals of recent Comatulce undergo a partial re- 

 sorption and transformation into the rosette is a sufficient reason 

 for regarding them as generically distinct from Solanocrinus, in 

 which they are more or less distinctly developed on the exterior 

 of the calyx. The difference is an important one, however, from 

 a morphological point of view ; but I do not think, that it is one 

 of any practical value, on account of the difficulty of determining 

 the presence of a rosette in fossil Comatulce. So far as I know, 

 all recent Comatulce (excluding Com aster) have a rosette ; but this 

 is absent in all the fossil forms in which we are able to see the 

 base of the calyx. But even in these the primitive unmetamor- 

 phosed basals do not always reach the exterior of the calyx, being 

 sometimes invisible when the centrodorsal is in situ. Hence the 

 absence of external basals in a fossil Comatula is not a sure sign of 

 the presence of a rosette internally ; so that I do not think it 

 possible to make any generic distinction between the forms with 

 external basals and those without them. I therefore follow 

 Miiller, Schliiter, and de Loriol in uniting Solanocrinus with Co- 

 matula, which is practically the same as with Antedon ; for 1 

 cannot refer any of the known species of Solanocrinus to the type 

 of Actinometra. 



I. — The type of S. costatus, as represented by Goldfuss §, is a 

 Comatula with a centrodorsal piece in the form of a short rounded 

 pentagonal column on which there are ten vertical rows of cir- 

 rhus-sockets. These rows are separated by vertical ribs, of which 



* Handbuch der Palaontologie, I. Band, p. 396. 



t Journ. Linn. Soc. Zoology, vol. xiii. pp. 454-456. $ Op. cit. p. 36. 



| Op. cit. tab. 1. fig. 7. 



