4 BULLETIN 76. UNITKD STATKS XATIDNAL MUSEUM. 



CLASSIFICATION OF THE ASTEROinEA. 



In this paper an attempt has been made to combine the best features of tiie 

 ciassiliciitions of Shxden, Perrier, and Verrill, but the results of my own work have 

 led me to modify niy former sclieme (1906) in many important details. The 

 Astropectinidae lias been enlarged and the Plutonasteridse al)andoned, a part of 

 its {icnera poing to tlie Astropectinidae, a part to the Bentiioijcctinida;, and Mimas- 

 ter to tlie Goniasteridro. It was a very unnatural group and the reasons for the 

 change are discussed under the Astroj)octinidic. The Goniasteridce has been 

 enlarged to include the old family Antlieneida? and a new subfamily for CMtovMster 

 has been instituted. Clixtaster has been removed from the Linckiidse and regarded 

 as a separate family, Cha-tasteridse Ludwig, 1897. The Acanthasterinse and Val- 

 vasterina? have been raised to family rank. 



The weakest point in every scheme of classification of the Asteroidea is between 

 the Phanerozonia and Spinulosa. Under the latter order Perrier ranges Porania 

 and it*) allies in the Poraniichr, while Dermasterias, Asterope, and Petricia are placed 

 in tlie Plianerozonia, or rather Valvata. I have merged the Poraniidae with the 

 Asteropida (Gynmasteriida? of authors) as a phanerozoniate family. Sladen regards 

 Asterina and its allies as belonging to the Plianerozonia. I have placed them in the 

 Spinulosa, following Perrier's system. Sladen classifies the Linckiidse in the Crypto- 

 zonia near Zoroaateridse; I have placed tlie family in the Plianerozonia. The 

 meaning of this is that no sharp line exists between the Plianerozonia and Spinulosa. 

 It is, however, not at all diihcult to separate the Forcipulata from the Spinulosa. 



In grouping the families I have followed as nearly as a lineal arrangement 

 will permit what seems to be the most natural order, but there are two or three 

 other combinations which are probably just as "natural." The Porcellanasteridse 

 are placed first to avoid sandwicliing them between the Astropectinida? and some 

 otiier family, and because they stand a])art from the other Phanerozonia. 



The following cltussification is used in this paj)er: 



ECHINODERMA ASTRORADIATA. 



Class ASTEROIDEA. 

 Order PHANEROZONIA. 

 Family PORCELLAXASTERID.E. 



Subfamily rORCKLi:.A.NA.STERIN"^S. 

 SubOimily CTKNODISCIN"^*:. 



Family GO.XIOPECTINID.E. 

 Family ASTROPECTINIDiE. 

 Family LITIDIID/E. 

 Family BEXTnOPECTINID^.. 

 Family ARCTIASTERID.'E. 

 Family CILETASTERTD.E. 

 Family ODONTASTERID.K. 

 Family GOXIASTERID^E. 



Sul>f:imily MIMA.SXKRII«r..iF:. 

 SulilUiiiily PSETTDAUCHA-STERIlsrvK. 

 SubOimily NECTRIINJ!*:. 



