568 Transactions of the Society. 



The greatest number of spicula with moniHform and conical 

 enlargements of the axial canal, have these penetrations from with- 

 out, to which attention has just been directed. This is illustrated in 

 Figs. 1, la, 7, Ih, 9a, 4c, and 13, 14. In some instances the 

 perforating tubules enter enlargements of the axial canal. It is 

 now necessary to inquire into the cause of the open condition of 

 the ends of axial canals and of the perforations. 



An open state of the axial canal at the ends of spicula, is very 

 rarely seen in perfectly formed and adult spicula of ordinarily 

 located spongida, and occlusion is brought about during growth. 

 It may have happened that some of the spicula now under con- 

 sideration are immature, but it certainly is not the case with many 

 which are figured. In none is there any evidence of solution of the 

 spicule from without, down to the ends of the axial canal, for these 

 are of the usual shape. Several spicula, such as Figs. 2, and 

 2a, 2&, have no perforating cyhndrical tubes on them, so that the 

 open condition of the axial canal is solely in relation to the series 

 of enlargements of its calibre. These may be immature forms, but 

 the curiously bent spinulate 11a is not immature, and there is an 

 evident opening of the canal. 



With regard to the enlargements of the axial canal, Bowerbank 

 has shown that the inner layers of silica of spicula are more readily 

 dissolved than the outer by reagents, and Sollas has enlarged our 

 knowledge on the subject. But the remarkable shapes produced 

 within the spicule, bear no relation to the method in which the 

 silica was originally deposited. 



Where a perforating tubule or tubules exist, there is not so 

 much difficulty in accounting for the solution of the silica around 

 the axial canal. Mr. Carter, in a letter to me, is good enough to 

 remind me of my communication to the Koyal Society * on the 

 Thallophytes which perforate the walls and septa of hard corals, 

 and to express a belief that these penetrations are produced by 

 somewhat similar organisms to Aclilya penetrans nobis. What he 

 has seen I have already noticed, and I can confirm its truth. 

 There is, of course, an indisposition to credit that a delicate cell- 

 structure could dissolve silica and penetrate into its mass; but if 

 carbon dioxide is the solvent in the case of the coral, it will be 

 efficacious in that of the organic colloid silica of the spicule. 

 Certainly there are no ramifications within the spicula as there are 

 within the corals afi'ected by the AcJili/a, and the sides of the 

 penetrations are sometimes, but not always, as bright and polished 

 as the outside structure of the spicule itself. The following, how- 

 ever, are the appearances which may be seen in many of the 

 penetrated spicula in the deposit now under consideration. But 

 before noticing them, it is necessary to explain that the spicula, of 

 * Proc. Koy. Soc, xxv. (1876) p. 238, pi. vii. 



