416 M, K. A. Zittel on Fossil Hexactinellida. 



or soldered together like a lattice — an axial cross, consisting 

 of three central canals cutting each other at right angles, lying 

 at the foundation of all the siliceous structures. Besides the 

 true skeleton-spicules *, numerous " flesh-spicules," generally 

 of very elegant form, are frequently present. 



Suborder I. DiCTYONiNA. 



Skeleton-spicules fused together in such a fashion that every 

 arm of a sexradiate applies itself to the corresponding arm of 

 a neighbouring spicule, when both are surrounded by a com- 

 mon siliceous envelope. The connected skeletons consist of a 

 latticework with meshes of cubical and irregular form. Flesh- 

 spicules present or wanting. 



Family 1. AstylospongidaB. 



Sponge-body very thick-walled, not stalked, free (some- 

 times, however, attached by a broad base). Water- vascular 

 system consisting of radial canals directed from the surface 

 towards the centre, to which are usually added vertical tubes, 

 which likewise stand to the number of 8-10 in radial series. 

 Latticed framework rather irregular, with solid crossing nodes. 



Astylospongia^ F. Rom. Silurian. 

 Palceomanon^ F. E,om. Silurian. 

 Protachilleum^ Zitt. Silurian. 

 Eosjpongia^ Billings. Silurian. 



Family 2. Euretidae. 



Sponge-body cup-shaped, cylindrical, top-shaped or branched, 

 attached. Skeleton lattice-like, the crossing-nodes of the co- 

 alescent sexradiates not perforated. Surface naked, or pro- 

 tected by a thickening of the outer skeletal layer, sometimes 

 coated with a very delicate network of coalesced spicules, 

 which differ but little in form from those of the rest of the 

 skeleton. This meshed outer membrane also veils the ostia. 

 Root-structure similar to that of the rest of the sponge-body 

 Flesh-spicules present or wanting. 



* The skeleton-spicules were originally defined by Carter (I. c. p. 353) 

 as " large spicules which are only concerned in the formation of the sup- 

 porting structure or skeleton." In a memoir which has just appeared 

 (Ann. & Mag. Nat. Hist. 1877, vol. xix. p. 121), "On two Vitreohexactinel- 

 lid Sponges," the denoixunation " skeleton-spicules " is applied only to the 

 larger free siliceous structures of the Hexactinellida. The coalescent 

 sexradiates which form the connected latticed skeleton, and which I have 

 particularly named skeleton-spicules in this memoir, are called by Carter 

 ** vitreous fibre." 



