NORTH AMERICAN PORIFER^. PART II. 509 



finer than the finest varieties of the hitter. Here, again, the phatter or saucer-shape, which 

 is a modification of tlie cup-shape, is absent. Spongia equina exhibits similar degrees of 

 variation in the texture of the surface and the form. There are no proper cup-shaped 

 specimens among the American varieties of sub-sjiecies gossyplna, but in place of these the 

 fistular form. These occur generally associated in clumps, more or less densely filled up 

 into heads, and solid, but sometimes the tubes are almost isolated, as in the specimen 

 figured, PI. XVI, fig. 7. The younger specimens of this species have a very loose and open 

 texture, due to the approximation and large size of the openings, and to a less degree 

 this is also to be remarked in the Gerbis sponge. The former approximate in aspect to 

 the coarser qualities of the American species, and so also does the latter, which has very 

 nearly the same color and a.spect as the dark colored Key West specimens, but is not so 

 coarse or dark. It seems, then, that there are three sub-species of commercial value in 

 the Mediterranean, and although there are more than three in the list of the American 

 forms which follows, there are really only three, or at most four, sub-species, which find 

 their way into the New York and European markets. The coarsest varieties of the Euro- 

 pean sponges are finer, firmer, and more elastic than the finest of the corresponding Amer- 

 ican sub-species. This is directly traceable to the larger amount of foreign matter included 

 in the primary threads, the looser mesh of the tissue ; the fibres are also comparatively 

 coarser and the large cloacal channels more numerous throughout the mass. Nevertheless 

 we can select specimens which it would be difficult to distinguish, such as the fragment, 

 PI. XVI, fig. 1, taken out of the side of a large Florida or Nassau sponge, of such excep- 

 tional fineness that only the very loose texture, form, dark color, and collection in which it 

 was found, enabled me to separate it as an undoubted Carribean form. Compare this with 

 fig. 8, and also compare fig. 16, a piece from the side of an old specimen of a Carribean 

 form of sub-species tubuUfera with the under side of fig. 18, a true cup sjionge from the 

 Adriatic of rather coarse quality. There is precisely the same sort of relationship here as 

 I have observed between the different qualities of sponges existing in the same seas. Thus 

 the different varieties of sub-sj)ecies gossyplna differ in an exactly similar w^ay from each 

 other, and from the third form, sub-species cerebrifor'mis. They differ in texture, in 

 surfoce, and also in habitat, the finer kinds, as stated previously, l^eing found in the deeper 

 water, equallj' removed from excessive heat and excessive sedhnent. These three sub- 

 species run together by means of specimens of the coarser varieties, which cannot be 

 distinguished from each other with any certainty, in the same manner as the corresponding 

 sub-species in the Mediterranean and Carribean Seas were connected, through the coarser, 

 and not by the aid of the finer varieties. It is evident, however, that besides the general 

 differences noted on previous jaages, that the cup-shaped form is not found in the American 

 sub-species, whereas it is the prevalent form of the Mediterranean sub-species. A cursory 

 examination of a large collection will, however, satisfy any one that the shape does not 

 necessarily correllate with a finer or a coarser skeleton, but probably with a more or less 

 extended base of attachment and local peculiarities, such as currents, and the kind of bot- 

 tom, etc., which have not been investigated in this connection. 



In order to obviate the technical difficulties attending the labelling of specimens accord- 

 ing to the trinomial system, it has been suggested to me that the sul)-species name should 

 be inclosed in brackets, thus, Spongia officinalis [tuhuUfera D. et M.] Linn., var. rotund((. 

 To this, however, though it would as completely represent the true affinities of the sponges 



MEMOIRS BOST. SOC. NAT. HIST. VOL. II. 128 



