dredged up from the Gidf of Manaar. 145 



the terminations are not round like those of most Discodermice, 

 but pointed, prong-like, and angular, as they are delineated 

 (PI. VII. fig. 47). 



Discodermida. 



Finding that the Discodermida grow by the transformation 

 of a simple disk on the surface to the complicated structure of 

 the interior, I have endeavoured to illustrate this in a series 

 of figures which are taken from a portion that was boiled to 

 pieces in nitric acid and the residue mounted in Canada 

 balsam (PI. VIII. fig. 48, d, &c). Indeed all the species 

 have been studied in this way, when the gradationary forms 

 of the disk thus separated render it very easy, as before stated, 

 to see and follow the changes of form which it undergoes ; 

 besides which, this may be corroborated by looking at the 

 specimen in situ in its natural state through a microscope 

 (PI. VIII. fig. 48, a, b, c). In addition to the disks the 

 Discodermida appear to be always characterized by one or 

 more forms of minute flesh-spicules in great abundance, of 

 which a curved acerate, or straight bacillar one, micro- 

 spined, is perhaps the most prevalent (fig. 48, h, and 49, c, d) • 

 but these are not confined to the dermal sarcode in which the 

 disks are developed and imbedded, although apparently most 

 abundant there (fig. 48, a, &c), but almost as plentifully dis- 

 tributed throughout the whole structure. Again, the filigreed 

 ends of the branches of the full-formed spicule are not angular, 

 but more or less spherical, like bunches of grapes interlocked 

 with one another, after the manner of the clasping of hands 

 (fig. 48, I). At the same time, however, I am not able to 

 explain the fact that in Kaliapsis cidaris, Bk. (Proc. Zool. 

 Soc. 1869, pi. xxv. fig. 2, &c.),of which I possess some good 

 specimens in situ, and which is a genuine Discodermia, a 

 " diadem-like " form of spicule similar in this respect to that 

 of Corallistes aculeata, exists under the discophorous layer. I 

 do not pretend to follow this transformation of the disk, which 

 is succeeded, as usual, by the fully formed Lithistid spicule of 

 the interior, but now only to announce the circumstance. In 

 specimens of Corallistes aculeata, in situ, the spikes may be 

 seen without any disks whatever, even when the surface is 

 overgrown and thus protected by a Microciona ; nor in the 

 mounted residue after boiling in nitric acid of specimens of 

 this species is there a trace of a disk under any kind of form 

 to be observed ; while in most specimens of Discodermice the 

 Microciona grows upon the disks themselves. 



In describing the structure of the different species of Disco- 

 dermida, I shall commence with the disks, as these are the 



