312 s. goto: 



authors and the second group represented by P. arcuatus and 

 others. Under these circumstances the limits between Tosia 

 and Mediaster must be more or less indefinite and arbitrary, and 

 I have therefore preferred for the time being to describe the 

 species already known under their old names. There is no doubt 

 of the close affinity of P. arcuatus to Mediader, but the question 

 is where to draw the line. The name Mediaster is retained 

 because the typical species so far described appear to me suf- 

 ficiently well marked to be placed in a genus by themselves. I 

 look upon Pentagonaster as used here as a provisional genus 

 which will undergo further modifications mainly in the way of 

 curtailment. It may also be mentioned that the genus Tosia 

 merges into Hippasteria through such a form as T. aurata Gray, 

 specimens of which I have recently been able to examine. 



FisHEE [ : 11, p. 204] has recently raised Verrill's subgenus 

 Ceramaster to the rank of a genus, and includes in it the following 

 species : C.japonicus, C. leptoceramus. C. patagonicus, C. clarki and 

 C. arcticus. There is much to be said for this genus, and under any 

 system of classification these species ought to belong together. The 

 materials at my disposal are however not extensive enough to 

 enable me to form a definite opinion on this point, inasmuch as in 

 my opinion examination of intermediate forms is especially needed. 

 Fisher [ : 11, p. 166] proposes to restrict the names Pentagonaster and 

 Tosia to the Australian species, and more recently Ludwig [ : 12] 

 has expressed himself substantially to the same effect. He refers 

 pulchellus Gray, ahnormalis Gray, crassimanus Möbius and diibeni 

 Gray to Pentagonaster, while australis Gray, oimata Müller & 

 Troschel, rubra Gray, nobilis Müller & Troschel, aurata Gray 

 and magnifica Müller & Troscel are referred to Tosia. 



