On Badylocalyx pumieeus {Stutclihury). By W.J.SoUas. 129 



the interior of the sponge. A similar function might perhaps be 

 assigned to the avicularia of the Polyzoa which hold fast for a 

 long while any little victim which may have been caught between 

 their beaks. 



First layer of reticulate skeleton. — Notwithstanding a close 

 search was made for them, no instances of framework spicules 

 existing in a free state could be found ; they could be seen in the 

 very first stages of cementation, but not earlier: certainly the 

 dermal spicules are very distinct, and never become involved in the 

 skeletal network, unless by rare exception ; the acerate spicules like- 

 wise, though occasionally involved, as a general rule remain free. 

 In the first stage of cementation we find two or three or more rays 

 of the framework spicules (Plate VIII., Figs. 1, 3, and 6) attached 

 to the rest of the network, from which the spicule seems to have 

 budded forth, the remaining rays projecting freely and usually out- 

 wards towards the exterior of the sponge; these free rays are 

 always more or less clavately capitate, and always microspined, 

 although they appear to have already become invested by a thin 

 layer of the ubiquitous siliceous cement. Some of these rays are 

 very persistent, retaining their freedom for a long while, especially 

 those which point directly towards the exterior of the sponge. 

 Near the centre of the attached spicule fine siliceous filaments cross 

 from one adjacent ray to another, subtending the angle formed by 

 them, so that when all six rays have been so connected together, a 

 hollow lantern joint results, which, when regularly developed 

 (Plate VII., Fig. 11), closely resembles the octahedral nodes of 

 Mylinsia Grayi or of a Ventriculite. Usually, however, its form is 

 much less symmetrical than this, owing chiefly to irregularities in 

 the form and distribution of the framework spicules themselves, but 

 partly also to the irregular way in which the connecting fibres join 

 them together. 



The rays of each spicule are bent in all directions, and the 

 entire spicules are scattered in great confusion, some lying one 

 way, and some another. The rays of adjacent spicules thus exhibit 

 no definite arrangement one with another ; sometimes the end of one 

 touches the middle of another ray, and where they touch they 

 unite ; sometimes two rays lie parallel to each other at a slight 

 distance apart, then transverse bridges of silica cross from one to 

 the other, and unite them into a fenestrated fibre ; frequently one 

 ray traverses the centre of another spicule, and thus multiplies the 

 number of fibres radiating from the resulting node of the finished 

 network. 



As the deposition of silica continues, the attached ends of the 

 spicular rays become covered up and disappear, the fenestra of the 

 open fibres are filled in, and solid more or less cylindrical fibres 

 result ; so, too, the open lantern of the nodes is in time obliterated 



VOL. II. K 



