82 DISCOVERY REPORTS 



and testes in series (subgenus Parastropecten, above). In Trophodisciis the gonads are as 

 in Leptychaster, sensu stricto, and, moreover, the species is paedophoric. 



It must be freely admitted that there is a wide difference in appearance between L. 

 kergiielenensis and Trophodisciis almiis, which has broad marginals superficially covered 

 with granuliform spinelets. These mask the deep fascicles between the transverse 

 elevated ridges of the plates. One can scarcely believe in the existence of these deep 

 grooves until the plates are cleaned. The first adambulacral plate is of a different form 

 from the others, being compressed, as in Parastropecten and Leptychaster s.s. In the 

 latter all the plates are normally so short that the slight disparity in the first plate is not 

 so apparent as in Parastropecten. 



Remarks on Priamaster. A detailed comparison of " Priamaster" magnificiis and 

 Leptychaster accrescens reveals so many points in common that one is forced to conclude 

 that they are closely related species. If accrescens is a Leptychaster, then magnificiis falls 

 in the same genus. 



Priamaster, of which magnificiis is the only known species, is eliminated, as is also 

 Priamasterinae Fisher,^ a subfamily of the Radiasteridae for Priamaster and Gephyreaster . 

 This subfamily was based upon a misconception of Priamaster, especially as regards the 

 structure of the tube feet. The distribution of the gonads was not given in the original 

 description. The podia of Gephyreaster have well-developed flattish sucking disks, 

 broader than the column of the podium. In Priamaster, as in other species of Lepty- 

 chaster, the terminal knob is of a low acorn shape or hemispherical form without a 

 sucking disk, and much narrower than the column of the podium. 



In the foregoing synopsis of Leptychaster I have indicated two divisions of the sub- 

 genus Leptychaster. A subgenus Priamaster might be justified for the second on the 

 basis of the differences mentioned. However, in young specimens of accrescens the 

 distribution of the papulae is interrupted on the ray by a longitudinal band midway 

 between radial line and margin. Between these two bands there is a radial area of 

 papulae. In general facies and in possessing short, angulated adambulacrals the southern 

 species form a fairly homogeneous group. But it is improbable that the accrescens- 

 magnificus group is paedophoric. 



Remarks on Mimastrella.- Sladen^ has described and figured Mimas ter cognatus, 

 while Ludwig* added a description and figures of the characteristic spinelets which he 

 compared with those of Liiidia. These 4-pronged spinelets are equally like those of 

 Lophaster, as Koehler^ pointed out, but with a vast difference in size. 



Mimastrella cognata bears a very close superficial resemblance to Leptychaster^ in 

 respect to the small paxilliform superomarginals, adambulacral armature, tube feet, 



* Ann. Mag. Nat. Hist., ser. 8, xx, 1917, p. 172. Also 1919, p. 216. 



2 Fisher, "Notes on the Systematic Position of Certain Genera and Higher Groups of Starfishes", Proc. 

 Biol. Soc. Washington, xxix, 1916, p. 5. Type Mimaster cognatus Sladen. 



3 Sladen, 1889, p. 336, pi. 47, figs. 3, 4; pi. 62, figs. 4, 5. 



* Ludwig, 1903, p. 6, pi. 2, figs. 11-17. 

 5 Koehler, 1920, p. 257. 



« Fisher, 1911, pp. 53, 175. 



