124 FOSSIL ASTEROIDBA. 



of this fossil. The dimensions of the marginalia, compared with the minor radius 

 and the ornament, are precisely the same as in C. Smithiae, from which the 

 specimen only differs in the possession of numerous valvate pedicellarias. In view 

 of the somewhat freakish way in which pedicellarias occur in starfishes, it does not 

 appear to me that this character alone entitles us to make a specific or generic 

 distinction on behalf of this specimen. 



There is another specimen also in the collection of the British Museum (E. Ill (>) 

 which has similarly numerous valvate ^edicellarise, but which shows the abactinal 

 aspect. In all other respects the specimen cannot be distinguished from a 

 specimen of G. Smithias. 



Genus METOPASTER. 



In the key-table I have only distinguished four species of Metopaster, namely, 

 M. Parkinson!, M. uncatus, M. quadratics, and M. cornntus, the latter being a 

 doubtful species. If one examines collections of Cretaceous Asteroids, one finds 

 that practically all the specimens have been rightly assigned to these species. 

 Specimens which could be assigned to the species M. Mantelli, M. BowerlanJci, 

 M. zonatus, M. sublunatus, M. cingulatus (see pp. 38-55), are very rarely met with. 

 The very considerable variation which occurs in undoubted specimens of 



M. Parkinsoni in the number of the supero-mar- 

 ginalia, their form, amount of ornament, and the 

 shape of the ultimate plates of this series, makes 

 specific characters founded solely upon these charac- 

 ters of doubtful validity, particularly as such varia- 

 tions occur even in an individual specimen, and it 

 is upon a rather extreme variation of these charac- 



TEXT-FIG. 33. Marginal of Meto- 



arkinsoni, showing a more t e rs occurring in very few specimens that this large 



scattered type of ornament than 



that xisuaiiy met with. mimber of species have been described. On the 



other hand, the presence or absence of spine-pits on the raised central area of the 

 plate is a constant character in species of Metopaster. Two specimens figured in 

 the Monograph appear to belie this statement. The specimen figured on PL X, 

 fig. 4 a, shows no spine-pits on its infero-marginalia, but is figured as M. Parkinsoni. 

 I have isolated a dorsal ossicle, which shows the specimen undoubtedly to belong 

 to M. uncatus. The specimen figured on PI. XI, fig. 3 a, as M. uncatus, shows 

 spine-pits on its supero-marginalia in one inter-radius only. After very careful 

 examination of this specimen, I have come to the conclusion that this inter-radius 

 the right-hand upper inter-radius of the figure has been added by a dealer 

 from a collection of ossicles of M. Parkinsoni to an imperfect specimen of M. uncatus. 



