33G s. GOTO : 



Madrcporite — The maclrcporito is tolerably large and some- 

 what rhomboidal in my specimen (PL XIII, fig. 202) ; it is situat- 

 ed near the centre of the disk, its outer margin being removed 

 from the margin of the disk more than four times as much as 

 its inner margin is from the centre. In tlie alcoholic specimen 

 the furrows are deep and conspicuous along the margin of the 

 plate but shallow and indistinct in the central area. They are 

 mostly radiating. 



Locality. — Misaki ; depth 560 m. 



Specimen in S.C. 



Eemarlxs. — This species appears to stand very close to 

 Pentagonaster puhlnus Alcock. In my species the armature of the 

 adambulacral plates consists for the greater part of three rows, 

 and only in the distal part of the arms are there four rows : a 

 furrow comb, two rows of granules on the actinal surface of the 

 plate, and a row of short thick spines between the furrow comb 

 and the rows of granules — a condition characteristic of P. puhinus. 

 Alcock, however, does not give the number of granules in the 

 two outer rows, which is of some importance for specific deter- 

 mination in such cases as the present, where two very closely 

 alhed forms come in question. I also find that nowhere in my 

 specimen is the adcentral spine of the second row of the adam- 

 bulacral armature replaced by a pediccllaria. The paxilloo are 

 closely set, but their granules are well spaced. What Alcock 

 says about the granulation of the marginal plates, the basal in- 

 terradial plates and the madreporite also do not apply to the 

 present species. It may also be mentioned in passing that P. 

 pidctniis is an intermediate form between the typical Tosia and 

 MecUastcr. In view of the close relationship of the present 



