803 W. J. 80LLAS ON THE STHUOTUBB AND 



culcs, which arc thus naked, being effected solely by the interlocking 

 of their extremities (PL XXYI. fig. 3). 



In many instances the rounded apophyses can be distinctly seen 

 in our phosphatized specimens, lying within but quite distinct from 

 the containing sockets, while in others the effects of fossilization 

 have gone so far as to obliterate the distinction between the two, 

 and to fuse the articulated clusters into a more or less solid homo- 

 geneous mass. 



In some parts of the sections (PI. XXVI. fig. 1 b) simple circular 

 areas occur, unconnected with the radiating arms ; these are due to the 

 section having passed through the shaft of a spicule at some distance 

 above its point of trifurcation. In others, again, the arms are seen 

 without a central circular area (fig. le), owing to the section having 

 passed through the plane of the arms beyond the end of the shaft, 

 and thus separated the shaft with the upper half of the rays from 

 the lower half, which remains imbedded in the slice. 



In the spicules of recent Lithistids one may often observe, in the 

 axis of the shaft, a central canal, which, at the point of radiation of 

 the arms, gives off three branches, one for the axis of each arm ; and 

 in a section of a fossil sponge which was sent to Mr. Carter as taken 

 from a Poh/pothecla, he has observed * not only all the details of 

 Lithistid structure which we have described in Slphonia. but has 

 also detected this very quadriradiate canal as well, occupying the 

 axis of the spicular shaft and rays exactly as in the recent forms. 

 In my specimens of phosphatized Siphonia?. I have carefully searched 

 many times for these canals under a magnifying-power of from 140 to 

 500 diameters, but always without success ; they have apparently 

 disappeared during the mineral replacements to which the substance 

 of the spicules has been subjected. Their absence, however, is of 

 no consequence in our inquiry, since the characters which remain 

 are in every detail so exactly those of a Lithistid spicule as to mako 

 the production of further evidence on this head unnecessary; and in 

 so considering them I have the unreserved support of Mr. Carter. 



Besides the Lithistid network the sections also exhibit a number 

 of simple finely-pointed acerate spicules (PI. XXYI. fig. o & 5 a) of 

 variable size, but sometimes attaining very respectable dimensions, 

 ranging in length from 0*02 to 0*045 of an inch, and in breadth 

 from 0-0015 to 0*007 of an inch. 



Arrangement of the Spicules. — The spicules of the skeletal net- 

 work, though they do not exhibit a very regular arrangement in 

 detail, are not, however, scattered without order through the sponge, 

 but present on a large scale a definite and regular disposition. 



Longitudinal Section (PI. XXYI. fig. 1). Selecting a band of net- 

 work lying between two radiating or inenrrent canals, one finds it to 

 consist of several scries of spicules, the filigreed or botryoidal ends of 

 which for each series lie along lines radiating from the exterior 

 towards the centre of the sponge, i. e. in the same direction as the 

 radiating canals ; the smooth arms of the spicules are likewise ar- 



* Ann. & Mag .Nat .Hist. ser. 4, vol. xii. p. 349, ' ; On the Hexactinellida and 

 Lithistidse." 



