NORTH AMERICAN PORIFER.E. PART II. 531 



out, principally, however, owing to the fact, that they are more or less charged with intro- 

 duced material. The bundles project above the surface, and are arranged along the sides or 

 as divisions between the channels which connect the mouths of the numerous tubes, which 

 penetrate to the surface without any special order of arrangement. The bundles are smaller 

 than in any of the preceding species, and, in fact, are really in some parts only single fibres, 

 as in Spongelia. This minuteness, their stiffness and projecting character, make the sur- 

 face look as if it were set with bristles. 



Loc, Zanzibar, in Coll. Peabody Academy, and in Soc. Coll. 



Stelospongos cnbriformis Hyatt. 



The surface of the skeleton presents precisely the asj^ect of Spongelia. This is owing to 

 the regularity with which the bundles of fibres occur upon the surface, and their small size, 

 as well as to the regularity with which these are connected by ridges of associated second- 

 ary fibres ; which also, on account of their minute size, resemble the single secondary fibres 

 of Spongelia. The linear or radiatory arrangement of the associated primary fibres forms 

 long and more or less continuous tubes penetrating into the interior ; these are so perfor- 

 ated or pierced by connecting holes through the skeletal walls, that they appear very 

 irregular. These holes are nearly, or quite, equal in diameter to the tubes, and entirely 

 interrupt the primary fibres. The division walls between these holes are formed by large 

 secondary fibres stretching across from one set of primary fibres to the next. These are 

 again connected by a third set of fibres which run from one hoi'izontal secondary fibre to 

 the next. Sometimes this regularity is not apparent, and only an irregular network is 

 visible, in which no distinction can be made between the horizontal and connecting or ter- 

 tiary set of fibres. In some parts of the same sponge, also, all regularity is lost, and the 

 whole is a network of irregular size and shape, similar to that of the true Spongite. 

 This is especially the case with the following. 



Variety stahiJis. In this variety (PL xv, fig. 15 ; PI. xvii, fig. 25) the tissue is unusu- 

 ally dense, and the surface is not marked, except in spots, by the usual tissue of cellular 

 aspect, dne to the opening of closely contiguous tubes upon the surface of the skeleton. 

 The primary fibres are also indistinct in the interior, and a considerable part of the skele- 

 ton has the irregular mesh just described. The primary fibres are shown only upon one 

 side of the surface, and here, also, tubes are found. 



Loc, Key West, in Yale Coll., Havana, in Coll. Mus. Comp. Zool. 



A Havana specimen combines the structure of the former with that of the next 

 variety, having a similar surface but more decidedly tubular, a hollow interior and an inter- 

 mediate size to the mesh of the skeleton. There is also a specimen from Nassau in the 

 Society's collection having something of this intermediate character, though perhaps, strictly 

 sjjeaking, it ought to be included in the next variety. 



Variety typica. The surface of this sponge, PI. xv, fig. 14, is more perfectly cellular or 

 coral-like than that of any sponge I have ever seen. The vertical walls between the tubes 

 are formed of very dense tissue, and there are but few horizontal threads crossing the 

 mouths of the tubes ; in fact, in some parts of each colony none at all. The mouths of the 

 tubes vary greatly in diameter, but in some specimens reach the diameter of 5 nun. 

 The intimate structure of the skeleton is distinct from the Key West specimen of variety 

 stab ills, but is only a less variable form of that found in the others of the same variety. 



