448 COLLECTIONS TEOM MELANESIA. 



apparently partly formed by upgrowths from short stems placed 

 beside it, which, though now single, appears to have been originally 

 compound also ; one surface of this expansion is almost level, but 

 honeycombed densely with openings with rounded' edges, varying 

 from 1 to 4 millim. in diameter ; the other surface is broken up into 

 seven more or less pronounced vertical ridges, irregular and fre- 

 quently interrupted, 3 to 10 millim. in height ; this surface and its 

 ridges are also strongly honeycombed, and between the openings 

 usually project points and ridges of sponge-tissue. It measures 

 130 millim. (Sj inches) in breath, 120miUim. (4|^ inches) in height. 

 The second specimen is strap-shaped, 24 to 30 millim. across, some- 

 what abruptly bent at one point, and terminated by two small lobes ; 

 like the other specimen, one surface is comparatively level and is 

 honeycombed rather minutely (openings -3 to -5 millim. in diameter), 

 while the other is rugose, from the presence of several demi-canals, 

 about 2 to 4 millim. across, which run from the middle to the margin 

 of the frond ; the surface between them is minutely honeycombed and 

 drawn up into a few sharp points and ridges. The colour is darkish 

 grey, varying to greenish in both specimens. The spiculation is 

 essentially that of the above-mentioned form of C. reinwardti, but 

 the skeleton smooth acuate is only '0063 to -0079 millim. in dia- 

 meter, and the short, stouter, smooth acuate is wanting ; this slight 

 difference in spiculation appears to justify the separation of this 

 form under a distinct varietal name. It is perhaps as nearly related 

 to the original form as the first-mentioned specimens. The remark- 

 able fan-shaped specimen appears (having regard to the midtiple 

 character of its base) to be made up of several " persons " which 

 have united to form a single symmetrical frond. 

 Hab. Bird Island, N.E. Australia, coral-reef. 



78. Clathria frondifera. 

 (Plate XLII. fig. i ; and Part II. of this Eeport, Plate LIII. fig. J.) 



Halichondria frondifera, Bowerbanh, Proc. Zool. Soc. 1875, p. 288. 

 Amphilectus frondifer, Vosmaer, Notes Rm^. Mus. Netherl, li. p. 116. 



A very common species, especially in Torres Straits. At first 

 I was inclined to separate the specimens from Bowerbank's 

 species on account of the greater development of ceratinous sub- 

 stance investing the lines of skeleton-spicules. In the Australian 

 specimens this occupies from half to two thirds of the diameter of 

 the fibres, whUe in the type specimen from the Straits of Malacca 

 its place is often almost entirely taken by spicules, and extends to 

 half the diameter of the fibre only in some cases. The consistency 

 of the sponge is hence much tougher and more elastic in the 

 Australian specimens. The main skeleton-spicules are much stouter 

 in the Malacca specimens. 



Dr. Bowerbank has omitted to describe a very fine tricurvate flesh- 

 spicule which occurs both in his specimens and in the present ones. 

 His description is also misleading in not stating, what his own type 

 specimen satisfactorily exhibits, and what the Australian examples 



