THE president's ADDRESS. 35 



cavity, we find that further progress in skeletal development 

 is determined by the requirements of the wall as a whole, and 

 no longer by those of the individual radial chambers. We have 

 here a beautiful example of integration. 



In some cases the dermal and gastral cortices become, as it 

 were, bolted together, across the intervening chamber layer, 

 by a special system of large subdermal and subgastral spicules, 

 whose principal rays extend in opposite directions from the dermal 

 and gastral cortex respectively, so as to interlock with one another 

 in the chamber layer. At the same time there is a more or less 

 strongly marked tendency towards reduction of the articulate 

 tubar skeleton of the individual radial chambers, which is replaced 

 functionally by the so-called inarticulate type developed in 

 relation to the sponge-wall as a whole. 



The evolution of the inarticulate type of skeleton appears to 

 have followed a different course in the two families Heteropiidae 

 and Amphoriscidae. The case of the Heteropiidae is the more 

 remarkable and forms a very curious illustration of adaptability 

 to special requirements in both form and arrangement of the 

 spicules concerned. We have all stages of the process represented 

 in existing species, and there can, I think, be no doubt as to the 

 facts of the case, which have been described by my colleague 

 Mr. Kow and myself in our paper on the Classification and 

 Phylogeny of the Calcareous Sponges,* nor as to the interpreta- 

 tion of those facts at which we have arrived. Briefly stated, 

 the sequence of events has been as follows. 



In an ordinary Sycon the blind distal extremities of the radial 

 chambers are conical in shape and, in conformity with the 

 curvature of the wall of the chamber, the outwardly directed 

 basal rays of the most distally situated triradiate spicules 

 become tilted towards the apex of the cone. One of the paired 

 oral rays now becomes directed more towards the gastral cortex, 

 while the other grows away from the chamber-wall into the 

 newly developed dermal membrane. With the development of 

 a true dermal cortex the basal ray and the last-mentioned oral 

 ray both take up their position in this cortex, lying parallel to 

 the dermal surface, while the first-mentioned oral ray becomes 



* P.Z.S., September 1913. 



