THE president's address. 33 



mother-cell, and that the apical ray of the quadriradiate is 

 another unit added to the system by yet another mother-cell. 

 It is difficult to explain the origin of the monaxon and quadri- 

 radiate spicules, but it seems quite possible that each arose as 

 a mutation, which, proving advantageous to the sponge, was 

 fostered by natural selection. 



If time permitted it would be easy to show you in detail how, 

 from this simple type of skeleton, various adaptive modifications, 

 the nature of which has been determined largely by mechanical 

 requirements, have led up to the far more complex skeleton in 

 each of the higher families of Calcarea, but we must content 

 ourselves with only a few of the numerous interesting facts that 

 recent researches have brought to light. It is clear that the 

 arrangement of the skeleton must always be conditioned to a 

 great extent by that of the canal-system, and, at first at any 

 rate, the canal-system has certainly led the way. Progressive 

 evolution has been brought about in this case by those processes 

 of colony-formation and integration to which I referred in my 

 last presidential address. Radial budding from the simple 

 tubular body of a primitive Leucosolenia gives rise to the 

 Sycon type, in which the collared cells that originally lined 

 the whole of the gastral cavity become restricted to the radial 

 tubes or chambers. The skeleton of the central tube becomes 

 more or less specialised to form the gastral cortex. In the walls 

 of the radial chambers the triradiate spicules take on a more 

 definite orientation. One ray becomes elongated and directed 

 towards the distal end of the chamber, and the other two tend to 

 surround the chamber, with the angle between them facing the 

 opening into the central cavity. The entire spicule thus becomes 

 " sagittal," with differentiated basal and oral rays, and inas- 

 much as the oral rays tend to form a series of rings around the 

 chamber the arrangement of this part of the skeleton is said 

 to be articulate. The monaxon spicules are confined to the 

 blind end of the chamber, where they form a projecting tuft, and 

 as all the ends of the chambers lie at approximately the same 

 level a new dermal surface is formed, protected by the projecting 

 monaxons. 

 In the meantime the walls of adjacent radial chambers come 

 JouRN. Q. M. C, Series II.— No. 78. 3 



