472 COLLECTIONS FEOM MELANESIA. 



4. 8. erasswula, id. ibid. p. 371. Basse Kooks, Ceylon. 



5. 8. australimsis, id. op. cit. 1883, ad. p. 350, pi. xiv. fig. 2. 

 W. Australia. ' 



6. 8. badllifera, var. rolusta, id. loo. cit. p. 351, pi. xiv, fig. 3. 

 S. Australia. 



Group 2. Without bacUlar or acerate flesh-spicule. 



7. 8. tefhyopsis, Carter, Ann. & Mag. N. H. 1880, v. p. 137, pi. 

 vi. figs. 39, 40. Gulf of Manaar, Ceylon. 



8. 8. glohostellata, id. op. cit. 1883, xi. p. 353, pi. xiv. fig. 5. 

 Galle, Ceylon. 



9. 8. hacea, Selenka, Zeitsch. wiss. Zool. xvii. p. 569, pi. xxiv. 

 figs. 14, 15. Samoa Islands. 



10. 8. jnirpurea, sp. n. N. coast of Australia. 



11. 8. clavosa, sp. n. N. coast of Australia*. 



iH^no Atlantic 8tellettce which I have seen do the minute or any 

 stellates possess capitate rays, except in a MS. species of Schmidt's 

 from Florida, which has minute drawn-out stellates (i. e_ incipient 

 spinispirular spicules) with very slight heads to the slender rays; a 

 larger stellate is, however, present in addition to these, and has not 

 heads to its rays ; the large stellate of 8. intermedia, Schmidt, 

 from Algiers, has the ends of the rays roughly tuberculated by pro- 

 minent groups of tubercles, but the spicule itself seems to be homo- 

 logous with the " balls " of Geodia, and not with the small stars of 

 8telletta, which are present as well. The Indo-Paoiflc species more 

 often have the head. In Stelletfa {Edonemia) densa. Bowk., from 

 the Piji Islands, the tuberculation of the rays is sometimes rather 

 coarser at their apices than on th'e remaining part, and in Eeionemia 

 aeervus the rays of the delicate stellate are very fine and slightly 

 capitate. Carter does not describe or figure any heads on the rays 

 of the stellates of his species from this region except 8. globostellata. 

 Selenka's species has no heads. 



The two species from Australia to be first described agree with 

 each other and with Eeionemia aeerviis in having small heads to the 

 stellates, although they differ from it, and agree with Stelletfa tefhy- 

 opsis, in the probably more important character of the absence of 

 a flesh acerate or baciLla;r spicule; the character of the apex of 

 the ray of the stellate in the latter species has not been described. 

 The Samoa^Islands species has no surface linear spicule assigned to 

 it by its describer, but it differs fundamentally from our species in 

 its large, noncapitate-rayed stellate. 



* 8. mastrvm of Carter (? Schmidt) deseribei (Ann. & Mag. Nat. Hist. 1882, 

 V. pp. 135, 136, pi. vii. figs. 41, 42) from the Gulf of Manaar and Australia, 

 includes two distinct species, of which the first at any rate is distinct from 

 Schmidt's species ; they belong to a remarkable group of forms which connect 

 Btelletta with Geodia : the surface-disk forms a character of sufficient import- 

 ance to distinguish the species which possess it from Stelletta s. str. 8. nux of 

 Selenka (iZeitsch. wiss. Zool. xvii. p. 569, pi. xxxt. figs. 11-13), from the Samoa 

 Islands, is probabljr a Tethua s. str., as its stellate agrees with the large stellate 

 of that genus, and its " forks " are rare and probably foreign to the sponge. 



