1908.] SPICULES OF CALCAREOUS SPONGES. 673 



almost pure calcite ; but the laj^er formed last of all, namely the 

 sheath, is again an " impure " layer. 



Having regard to the mode of formation of these spicules 

 discovered by Minchin (1898 and 1908) and Woodland (1905), it 

 would be a tempting hypothesis to refer the two substances 

 secreted to the activities of the two formative cells ; the apical 

 formative cell or "founder" may be supposed to lay down the 

 " impure " substance, while the basal formative cell or "thickener " 

 secretes the purest calcite. On the other hand, the formation of 

 the sheath must also be ascribed to the thickener. 



The continuity, generally to be observed, of axial filament and 

 sheath, and the similarity of their staining reactions are points 

 in favour of considering these two sti-uctures to be of similar 

 nature. "We have referred above to Biitschli's arguments in 

 favour of regarding the sheath as being chiefly of inorganic 

 nature, a conclusion for which there is much to be said, and 

 which may be extended to the axial filament. The fact, however, 

 that both filament and sheath have an affinity for special stains, 

 is in favour of their containing a certain amount of organic 

 matter, and we may regard sheath and filament as consisting of 

 an organic basis richly impregnated with inorganic non -crystal- 

 line materials. At this point we must leave the question of the 

 nature of these structures to receive more exact and definite 

 solution from more competent observers. We claim merely to 

 have demonstrated the following proposition : — The spicules of 

 ■calcareous sponges leave after decalcification a residue in the form 

 of stricctural constituents, sheath and axial filament, ivhich can he 

 coloured hy special stains. 



In conclusion, attention may be drawn to some points relating 

 to the morphology of the spicules, upon which the axial filaments 

 throw some light. It is seen that in the rays of the triradiates, 

 the filament is broad and even band-like at the base, and tapers 

 to a fine point at the apex. Comparing with this the monaxon 

 of A. falcata (fig. 12, PI. XXXIV., figs. 13, 14, PI. XXXYII.), 

 it is seen that the filament is broad and band-like at the blunt 

 distal end of the spicule, and tapers to a fine thread at the 

 pointed proximal end. This supports the conclusion, based by 

 Minchin (1908) upon developmental data, that the distal pro- 

 jecting ends of the monaxons are homologous with the central 

 ends of the rays of the triradiates. 



Comparing, however, the monaxons of C. contorta^\\h. those of 

 A. falcata (figs. 9, 10, PI. XXXV.), it is seen that in C. contorta 

 the filament is band-like towards the middle of the spicule, but 

 tapers to a fine thread at each end. This strongly suggests that 

 the monaxons of this sponge are not really primary monaxons, 

 but are secondarily derived from triradiates and aie to be re- 

 garded as biradiates as suggested by Minchin (P. Z. S. 1905, ii. 

 p. 10). On the other hand, the monaxons of A. falcatOj would 

 appear to be true primary monaxons. 



43* 



