218 J. G. WALLER OX A NEWLY DISCOVERED BRITISH SPONGE. 



Without the microscope it would not have been discovered ; for it 

 was by passing it over the shell, examining numerous incrustations, 

 that I noticed a small patch of bihamate spicules. The patch was not 

 bigger than an eighth of an inch, but so difficult to detach from the 

 shell's rough surface, that a large part was destroyed in the process ; 

 it belongs to the lowest form of the filmy sponges. The tiny object 

 therefore, which I secured and mounted, is rather under l-20th of 

 an inch in diameter ; but it would exceed this if it could all be 

 flattened out, by treble that size. As it is, however, fortunately it 

 exhibits all the characteristics of the sponge, which is one of the 

 most interesting of its kind it has been my fortune to see, and intro- 

 duces us to a form of spicule not hitherto found, as far as my pre- 

 sent knowledge goes, in the Spongidse. This alone would make us 

 rejoice in the discovery, though it is not the only point of interest. 



My specimen partially coats a fragment of oyster-shell, as also 

 portions of a structureless substance, like the glutinous lining of 

 the tubes of annelids, and seems to belong to Dr. Bowerbank's 

 genus Hymeraplua. The simplicity of structure which marks 

 this class is shown by its consisting of a membrane, strengthened 

 with spicules. In this example the membrane is very trans- 

 lucent, and scarcely visible when mounted in balsam. Upon 

 this is a skeleton made of clavate cylindro-arcuate spicules, 

 somewhat long and disposed in fasciculi (Fig. 1), these being loosely 

 connected by a few single spicules of the same kind. They measure 

 30-4000ths of an inch. (Fig. 2.) But one of the most distinctive 

 features is, that the membrane is bound together by a close inter- 

 lacing of contort bihamate spicules, very numerous, and making a 

 confused network. The normal size of these is 7-4000ths of an 

 inch. Intermingled with these are others of the same form and 

 character ; but nearly three times the size, very few in number, 

 whose purpose might fairly be supposed to clamp and strengthen this 

 network. (Figs. 3, 4, 5.) There are also anch orate spicules along with 

 this reticulated mass. Some large, tridentate, and bidentate of same 

 kind ; equi-anchorate, few in number ; others very small, more abun- 

 dant ; often very difficult to detect in the confusion of the bihamate 

 network ; and some very minute, which may more possibly be con- 

 sidered to be in an undeveloped state. The large forms are re- 

 markable in the unequal character of the flukes. In the bidentate 

 one flange is smaller than the other ; the tridentate has a fine 

 projection from one central tooth, and the other fluke has a 



