388 Mr. W. K. Fisher on 



b'. Superambulacral plates absent, no con- 

 spicuous buttress extending from the 

 upper enlarged end of the first two pairs 

 of anibulacral plates to the body-wall at 

 the interradial angle. 

 c'. All adambulacral plates carinate on the 

 furrow face. Genotype, Proynaster 



griinaldii, Perrier Pi'ognaster *, Perrler. 



c^. Adambulacral plates alternately cari- 

 nate and non-cariuate. 

 d' . Eays long, slender ; disk small ; 

 abactinal, marginal, and actinolateral 

 plates arranged in regular longi- 

 tudinal lines along ray, a series of 

 adradial platesbeiug always present; 

 all but the median radial or carinal, 

 which are larger, form also trans- 

 verse series ; plates are covered with 

 email, papilliform, skin - covered 

 spinelets, and most of them bear 

 an eularged spine ; papular areas 

 generally very small but sometimes 

 nearly as large as plates ; fortici- 

 form pedicellarifB present ; actino- 

 lateral plates in three to live series, 

 the upper subequal to the iufero- 

 margiual plates ; superomarginal 

 plates not conspicuously larger than 

 the inferomargiual plates. Geno- 

 type, Zoroaster ful(/ens, Thomson . . Zoroaster, Thomson. 

 <f . Plates of ray arranged in regular 

 longitudinal series, the carinal 

 plates the largest; abactinal and 

 marginal plates armed with fairly 

 large, skin-covered scales which 

 mask all plates except some of the 

 disk-plates and the carinal series 

 along ray ; no pedicellaria3 ; actino- 

 lateral plates, in two or three series, 

 much smaller than inferomarginals ; 

 superomarginal plates much larger 

 than inferomargiual plates. Geno- 

 type, Pholidaster sqiuniiatus, Sladeu. Pholiduster, Sladen. 

 h^. Superambulacral plates present ; a con- 

 spicuous buttress, the specialized first 

 superambulacral plate connects the 

 upper end of the first two ^mbulacral 

 plates with the body-wall at interradial 

 angle. 



* Proynaster, Perrier, * Comptes rendus,' cxii. no. 21, May 5, 1891, 

 p. 1226. Type, P. grimaldii. Also ' Resultats des campagnes scienti- 

 tiques du Prince de Monaco,' fasc. xi. 1896, p. 22, pi. ii. tigs. 1, 1 u-b. In 

 the meantime, Perrier described Prognaster as a new genus with P. longi- 

 cauda, new species, ostensibly as type (Exped. scientif. du ' Travailleur ' et 

 du ' Talisman,' 1894, p. 119). This species does not seem to be con- 

 generic Avith grimaldi, which, of course, is the genuine genotype. 



