Mr. II. J. Carter on the Hexactinellida?. 353 



whole structure together; for, indeed, if divested of this filigreed 

 development, as is the case with the surface-spicules of Dacty- 

 localyx Bowerbankii, Johnson (which are simply trifid-headed 

 shafts with the shaft projecting- internally and the trifid heads 

 horizontally), they would be reduced to the state of similar 

 spicules in the Geodinidaj and in Pachastrella (that is, held 

 together only by the amorphous sarcode of the dermis), and 

 would thus form a group in my fifth division. 



We shall see by-and-by that, while the glassy skeleton of 

 the Hexactinellidaj is formed by a vitrification of the sarcode 

 or horny fibre, that of the Lithistidse is formed by a vitrified 

 extension of the spicule. 



Assuming this, I shall for the present give lists respec- 

 tively of the Hexactinellidaj and Lithistidaj which have already 

 been described, grouping them, according to the best of my 

 judgment, as their characters seem to indicate, and following- 

 each by a short commentary. After which I shall take the 

 branched tube-net sponges (that is, the Aphrocallistidas, in 

 which we must include Farrea and Aulodictyon) for the special 

 subject of this paper, returning to the other groups for more 

 detailed description at a future time when it may be more con- 

 venient. I shall also, for the present, use the terms "sex- 

 radiate" for the HexactinellidaB, and " quadriradiate " for the 

 Lithistidse. 



As the spicule of the Hexactinellidse and Lithistida3 is, like 

 that of all other sponges, developed from a single point or 

 minute cell, which, by linear extension, becomes the central 

 canal on which the vitreous layers of the spicule are sub- 

 sequently built, we find that in the Hexactinellidaj the spicules 

 are produced by a more or less uniform radiation from the 

 central cell immediately ; while in the Lithistida3 the spicules 

 are formed by a more or less uniform triradiation subsequently, 

 or from one extremity of the primary cell after it has un- 

 dergone linear extension, — that is, that this half of the shaft 

 is transformed into three arms, and that, too, probably com- 

 mencing a little beyond the central cell, as an inflation of the 

 shaft, often just below the head, leaving a kind of neck or more 

 constricted portion, would seem to indicate. (See especially the 

 large trifid spicules of the Geodinidas &C.) 



With reference to the Hexactinellidas, it is also necessary to 

 premise that there are here, as in most of the Spongiadse, large 

 spicules which are only concerned in the formation of the sup- 

 porting structure or skeleton, which we shall term " skeleton- 

 spicules," and minute spicules which belong entirely to the 

 sarcode, which will be termed " flesh-spicules " — also that the 

 latter here often afford a valuable character in conjunction with 



