﻿20J -ENDEAVOUK" SCIENTIFIC RESLI/IS. 



7mm. in diameter; (ii.) by the gauziness ot its texture; (iii.) 

 by the slenderness of its fibres, which do not exceed 80 ;< in 

 diameter; (iv.) by the larg^e size oi its megascleres ; (v.) by 

 the "manubriation" of the basal end of the principal styli ; and 

 (vi.) by the paucity of numbers of its stout acanthostyles 

 which, when echinating, are only slightly embedded in 

 spongin. The original description says, "surface minutely 

 honeycombed; the cells are from x-2 mm. in diameter, etc." 

 This statement appears to me entirely misleading, since except 

 for the gauzy looseness of the texture, there is nothing par- 

 ticularly of the kind thus suggested. The maximum 

 dimensions of the spicules, some of which are not quite 

 correctly stated in the previous account, are as follows : — 



Megascleres. — 



(i.) Principal styli, 300-305 x 13//. (Spicules "i5/<" ia 



diameter have not been observed). 

 (ii.) Auxiliary styli, often basally spined, 350-370 x 7- 



10}l. 



(iii.) Special dermal styli, usually basally spined, and, as. 



a rule, less than 135 x 5 ;<. 

 (iv.) Acanthostyli, 70-80 x lojx. 



Microscleres. — 



(i.) Larger chelae, 14-18 /-( long, occasionally (though 



very rarely) contort. 

 (ii.) Smaller contort chelae, 8-12 /<. 

 (iii.) Toxa : obserxed lengths, 70-255 i'. 



Locs. — Off Port Jackson, 49-50 fms. ("Thetis"); Oyster 

 Bay, Tasmania, 60 fms. ("Endeavour.") 



Genus Thalassodexdrox, Lendcnfeld. 



1888. Thalassodendron^ Lendenfeld. Descr. Cat. .Sponges- 

 Austr. Mus., 1888, p. 222. 



The genus Thalassodendron was proposed for "Desmaci- 

 donida? with a supporting skeleton composed of horny fibres, 

 partly without spicules in the interior and echinated by scarce, 

 pretty smooth styli." The definition attached no importance 

 to the nature of the coring spicules and regarded as essential 

 a feature which we now know to be lacking in systematic 

 value. As a consequence it happens that the six species, 

 described — T. dis;itaia, 7'. iypica, 7'. rubens, varr. dura et 

 latnella, T. paucispina, T. hrevispina and T. vhninaUs — fall 

 into several different genera. Thiele^ has already pointed out 

 that T. digitata possesses the requisite characters of his genus. 



1 Thiele — Kieselschwamme von Ternate, ii., 1903. p. 962. 



