360 s. goto: 



marginals, the presence of an anus, etc. ; the points of difference 

 are the rudimentary condition of the marginal spines, the weak 

 development of the teeth, the presence of suckers on the tube- 

 feet and the pedicellariœ. The author thinks that with these 

 restrictions the genus may be referred to the Plutonasteridœ. It 

 appears to me that this genus occupies an intermediate position 

 between the Ar chaster idœ and the Pentagonaster idee, being related 

 to the former by the absence of the superambulacral plates and 

 to the latter by the character of the plates and the pedicellariœ. 

 The total absence of spines from the marginal and ventrolateral 

 plates in the species described below increases its resemblance to 

 the Pentagonaster idee. In fact this genus stands next to Nymph' 

 aster, being related to it through such a form as A^. symholicus 

 Sladen, which has been placed by some authors (Vereill, Fishee) 

 in a distinct genus, Nereidaster. 



Kœhlee gives a detailed description of one species from the 

 collection of the " Investigator " in the Indian seas [ : 09, p. 8]. 

 The species described below, though presenting some notable 

 differences from Kœhlee's, shows a close affinity to it and there 

 is no doubt that both are to be placed in the same genus. 



Johannaster differs from Nymphaster by the presence of several 

 rows of more or less irregularly arranged plates between the 

 superomarginals of the arms'^ and the papular areas extending into 

 the latter. It may also be added that it will be necessary to 

 modify the generic diagnosis by striking out the phrases bearing 

 on the spines of the inferomarginal and ventrolateral plates. 

 Fishee [:11, p. 162] has recently placed this genus in the 

 Goniasteridœ (s. Pentagonast er Idœ), 



1) As for Xersklfister bowersi Fisher it appears to me that it is more nearly related to 

 Pentajona.ster arcnatus and P. m'lsukloisis tlüin to Xympli/mter .synùolicus. ('/• Fisheb [ :13, p. G29]. 



