254 PERCY SLADEN TRUST EXPEDITION 



but numerous smaller ones occur and also numerous forms connecting with the larger 

 oxea (l). 



(3) Oxyasters (fig. 2 c) ; with no distinct centrum, or quite a small one, and slender 

 rays, usually about eight in number; total diameter of spicule about 0"016 mm. 



Topsent has pointed out the variability in the size of the spicules in this species, and 

 also mentions that the smaller oxea are frequently centrotylote. I have therefore no 

 hesitation in identifying the "Sealark" specimen with the European form. I have 

 thought it desirable, however, to give measurements and figures of the spicules of the 

 "Sealark" specimen as a contribution to our knowledge of the range of variation. Topsent 

 [1898 b] has also expressed the opinion that Lendenfeld's Asteropus incrustans is identical 

 with Jaspis johnstonii, an opinion with which I agree. It appears to me very probable 

 that Row's Coppatias albescens from the Red Sea [1911] may belong to the same species, 

 and possibly also Lindgren's Dorypleres hiangulata from Java [1898], also recorded by 

 Thiele [1903 b], under the name Jaspis hiangulata, from Ternate. 



Previously known Distribution. Adriatic (Schmidt, Lendenfeld) ; Mediterranean 

 coast of France (Topsent) ; St lago, Porto Praya, Cape Verde Is. (SoUas) ; Azores 

 (Topsent); IRed Sea (Row); ? Java (Lindgren) ; ^Tei-nate (Thiele). 



Register No., Locality, &c. lxxviii. 11, Cargados Carajos, 28.3.05. B. 2, 

 30 fathoms. 



Family G-eodiidse. 



Astrotetraxonida with trisene megascleres, a cortical layer of sterrasters, and various 

 forms of euasters, to which microrhabds may be added. 



The question of the probable origin of this family from the genus Aurora has 

 already been discussed. The classification of the Geodiidse is an extremely difficult 

 problem and one which cannot be properly tackled without a comprehensive re-investiga- 

 tion of the group. I may state, however, that I have the gravest doubts as to the value 

 of the character and arrangement of the inhalant and exhalant openings for the dis- 

 crimination of genera, upon which so much stress has been laid by SoUas and subsequent 

 writers. I therefore use the old genus Geodia in a far less restricted sense than is 

 customary, but, laying the greatest stress as usual upon skeletal characters, I have framed 

 the diagnosis thereof so as to exclude Pachymatisma, Caminus and Geodinella, which may 

 conveniently be kept distinct. 



I have followed Lendenfeld [1910 b] in referring the genus Erylus to a separate family. 



Genus Geodia Lamarck [1815 a]. 



Geodiidse with well- developed trisenes arranged radially at or near the surface ; 

 without microrhabds and without spherules. 



15. Geodia auroristella n. sp. 



(Plate 47, fig. 3.) 

 The single specimen, in its present condition, has the form of a thin crust, which has 

 evidently been sliced off either from a larger specimen of the sponge or from some foreign 

 object which it was encrusting. As there is no larger specimen of the sponge in the 



