400 Mr. H. J. Carter on Veep-sea 



two of these linear projecting shafts having been broken oft' 

 leads me to infer that tlicse might have been anehoring- 

 spicules which had lost their heads, as the latter are not to 

 be seen on either embryo. The whole of the embryo and its 

 spicules are, of com'se, of microscopic minuteness, as they can 

 only be seen with 5 -inch compound power, equal to nearly 

 400 diameters, although quite as clearly as if the spicules had 

 been of the largest size. 



I have been thus particular in describing these embryos 

 taken from the base of a Geodia and corresponding in the 

 form of their spicules with those of that Geodia j because the 

 name of " ovarium " lias been applied to the " siliceous balls " 

 of the petrous crust by Dr. Bowerbank, and that of "ovisacs" 

 by the late Dr. Gray in his " Notes on the Arrangement of 

 Sponges " (P. Z. S. 9th May, 1867), while many others have 

 adopted a similar terminology ; so that, had not Dr. Johnston 

 {Brit. Spong. 1842, p. 202), Schmidt, and those wlio have 

 particularly examined these bodies from their earliest appear- 

 ance to their full development, which every specimen of 

 Geodia presents in abundance, been perfectly satisfied that 

 they could not be considered reproductive elements of the 

 Geodia under any form, these two embryos would prove that 

 the " siliceous balls " are nothing more than sponge-spicules 

 of this particular form ; besides, they have just the appear- 

 ance and general character of the embryos of Tethya cranium ^ 

 which I described and illustrated in 1872 (Ann. & Mag. Nat. 

 Hist. vol. ix. p. 426, pi. xxii.). 



The description of Geodia nodastrella above given may 

 appear prolix ; but it is the first time that I have had the 

 opportunity of giving a typical one ; and the species are so 

 much alike that this in its general characters may serve for all 

 the rest. 



Geodia megastrella^ n. sp. (PI. XVI. figs. 46 and 46'.) 



General form hemispheroidal, elongate, flattened at the base, 

 where it is adherent to the surface of the large fragment of 

 Corallisies Bowerhanhii on which it has grown, presenting a 

 large circular aperture on the summit. Colour now grey. 

 Surface, where not rubbed off, the same as in the foregoing 

 species ; but the stellate of the dermis (fig. 46', /i, m) furnished 

 with minutely spined capitate rays instead of simple nodes, 

 and the siliceous ball very large, being 13-1800ths inch in 

 diameter (fig. 46', (j). Pores as in the last species (fig. 46, c). 

 Vent single, on the summit, consisting of a large circular hole 

 partly closed by a diaphragm of sarcodc (fig. 46, d). Internal 



