I 



8P0SGIIDA. 367 



tremely short, and deal almost invariably with the mere external 

 characters of the forms to which they refer, and thus, owing to the 

 well-known variability and comparatively slight diagnostic import- 

 ance of these characters in the Spongiida, are almost useless, per se, 

 even for the identification of species. 



Thus we have in all some 90 species (allowing for synonyms), 

 more or less fully described, of Sponges chiefly from the southern, 

 south-eastern, and south-western coasts of Australia, and some GO 

 species described in the barest manner, probably (but not certainly) 

 in most cases from the northern coasts, and a few from the south. 



The present collection comprises upwards of 300 specimens, repre- 

 senting 110 species, besides 7 distinct varieties, of which more than 

 half are well preserved in spirit and the remainder are dry. The 

 districts searched consist of: — 1. Port Jackson, IN". S. Wales ; 2. 

 Several points on the north-east coast of Queensland ; 3. Various 

 islands and spots in Torres Straits ; 4. The Arafura Sea, between 

 Cape York and Port Darwin ; 5. Port Darwin, N.W. Australia. 

 The depths range from between tide-marks to 36 fms. (Arafura Sea), 

 but most dredgings did not exceed 20 fms. in depth. 



It is perhaps not surprising, after what has been stated as to the 

 previous work which has been done among the Australian Sponges, 

 to find that a large proportion (42 out of 110, or 38 per cent.) of 

 the species are certainly new to science, and that a considerable 

 number more may possibly prove to be so, having been assigned only 

 doubtfully to described species. However, the distribution of the 

 littoral species (when these are properly defined and limited) of 

 Sponges appears usually to be but moderately wide, and perhaps 

 less so than the shallow- water Alcvonaria ; both this and another 

 fact must be adduced to account for this large number of new spe- 

 cies, viz. that the sponge-faunas of the neighbouring and moderately 

 distant seas are even less known than that of Australia itself ; this 

 will be seen clearly from a statement of what has been done to 

 elucidate the fauna of the Indian Ocean which I have made under 

 the heading Geographical Distribution. I have in this dearth of 

 information taken pains to describe every form which warranted 

 description. I have been careful not to assign a new name where 

 the sponge might possibly have been already described, in order not 

 to run the risk of adding to the overburdened synonymy, but have 

 generally given a full description in such cases, so that no doubt 

 might remain as to the characters of that species, at any rate, with 

 which I have had to deal. 



Taxonomy of the Collection. — Of the 1 10 species obtained, 20 (or 

 more than one sixth) are Ceratosa, a number which is illustrative of 

 the largely tropical character of the localities from which they are 

 drawn ; the Dysideidoe include two new species and a most inter- 

 esting variety of a most important form, Pmmmopemma densicm, 

 Marshall, whose nature receives hereby confirmation and elucida- 

 tion ; the number of new Ceratosa (4) is not great, as most of the 

 species seem to have a wide range. 



The Silicea, as usual, far outnumber the other groups ; the 



