52 



PROCEEDINGS OP THE NATIOISTAL MUSEUM 



VOL. 81 



reaching the firm substratum as if they were used for supporting 

 the entire sponge. I have found many cases in which the diverticula 

 are arranged in more or less longitudinal rows usually nine in num- 

 ber, which in inferior parts of the entire tube are arranged in a 

 cruciate manner. Above, this arrangement becomes indistinct and 

 irregular. Where the upper terminal opening with its natural mar- 

 gin is present and uninjured, it is closed by a transversely stretched 

 narrow-meshed latticelike plate. The latter is somewhat concavely 

 incurved, or occasionally not curved, and united to the honeycomblike 

 lateral wall. In a few cases this latticelike transverse partition occurs 

 in the interior of the tube, as already reported by Schmidt and Mar- 

 shall. I agree Avith Schulze's opinion regarding its formation and 

 significance, as mentioned in his Challenger report. There is some 

 doubt in regard to the onychaster of specimens described by other 

 authors. In the specimens of the Challenger expedition, Schulze does 

 not describe the onychaster, while in the specimen from Andamans in 

 the Indian Ocean, he reports it for this species. On the other hand, 

 the discoctaster was mentioned by him in the former specimen and 

 not in the latter. At any rate the spicules in question are present in 

 all the specimens before me. 



APHROCALLISTES INTERMEDIA, new speciea 



Three fairly large, macerated fragments and several well-pre- 

 served fragments (cotypes, U.S.N.M, No. 22121) of this species were 

 collected from two stations (Table 15). 



Table 15. — Record of specimens of Aphrocallistes intermedia 



I have some doubt as to whether inter^nedia should be placed under 

 Aphrocallistes or whether a distinct genus should be erected for it. 

 The characters distinguishing it from Aphrocallistes are found in 

 the presence of scopulae on the gastral regions and in the existence 

 of a peculiar gastral oxyhexaster. The species is generically associ- 

 ated with Aphrocallistes rather than with Ghonelasma^ because it 

 has diactins in the subgastral regions. 



Splculation. — The dictyonal net forms a regular and nearly uni- 

 form hexagonal honeycomb. The hexactinic dictyonalia are joined 

 in a regular manner to form a net or latticework with nearly quad- 

 rangular meshes. Slender, faintly tuberculous, short, cylindrical 



