Potash on the Siliceous Skeletons of Sjionges. 291 



one of the anchoring spicules of this sponge ; it was perfect 

 from one end to the other, inchiding the barbed head and den- 

 ticulated scalar lamina. The great length and size of this 

 spicule had previously led me to imagine that, although appa- 

 rently nothing but a simple spicule, it might yet upon investi- 

 gation be found to consist of several spicules imbedded in a 

 siliceous fibre. 



A careful examination was therefore made to determine this 

 point ; with sharp scissors the spicule was divided into con- 

 venient lengths and then boiled in potash. The axial canal 

 soon became clearly visible and could be continuously traced 

 from one end of the spicule to the other ,• it was obviously 

 quite simple, and radiated once only, viz. in the barbed head, 

 Avhere it forms a sexradiate cross : no other canal in addition 

 to this could be detected in the spicule, which thus, contrary 

 to my expectation, was proved to be not spicular fibre but 

 a simple spicule after all. With the development of the 

 axial canal several faint lines or cracks, circumferential in 

 direction, appeared on the exterior of the spicule, and were 

 immediately traversed by others running longitudinally ; the 

 outermost layer then peeled off in strips between the circum- 

 ferential cracks ; and as the jjrocess continued, other laminte 

 were exposed and similarly scaled away. The ends of the 

 various lengths of the divided spicule soon lost their cylindrical 

 form and became conical, the alkali dissolving backwards the 

 successive concentric lamina? of which the spicule was composed, 

 the outermost first and consequently most extensively, and the 

 rest in successive order, till at length the original strand on 

 which all the rest had been formed was exposed as a single 

 thread with an axial canal which, as happens in all similar 

 cases, was expanded towards its termination, funnel-shaped (as 

 in fig. 5, PI. IX.). 



i^othing can exceed the elegance with which caustic potash 

 exposes the concentrically lamellar structure of this spicule : 

 each individual lamella is so perfectly separated from its 

 fellows that one can count the number of which the spicule is 

 composed at various points of its length ; thus in the figure 

 (PI. IX. fig. 5) ten lamina3 are to be seen. 



The sponges next examined belong to other orders than that 

 of the Hexactinellidae. 



Geodia arabica. 



In spicules with sharp points, like the anchoring and aceratc 

 spicules of this sponge, the axial canal extends to within a 

 very short distance of the pointed extremity ; so that as the 



