IX] OF THE SKELETON OF SPONGES 447 



But if such a physical explanation of the forms of our spicules 

 is to be accepted, we must seek at once for some physical agency 

 by which we may explain the presence of the solid material just 

 at the junctions or interfaces of the cells, and for the forces by 

 which it is confined to, and moulded to the form of, these inter- 

 cellular or interfacial contacts. It is to Dreyer that we chiefly 

 owe the physical or mechanical theory of spicular conformation 

 which I have just described, — a theory which ultimately rests 

 on the form assumed, under surface-tension, by an aggregation 

 of cells or vesicles. But this fundamental point being granted, 

 we have still several possible alternatives by which to explain the 

 details of the phenomenon. 



Dreyer, if I understand him aright, was content to assume that 

 the solid material, secreted or excreted by the organism, accumu- 

 lated in the interstices between the cells, and was there subjected 

 to mechanical pressure or constraint as the cells got more and 

 more crowded together by their own growth and that of the 

 system generally. As far as the general form of the spicules goes, 

 such explanation is not inadequate, though under it we may have 

 to renounce some of our assumptions as to what takes place at 

 the outer surface of the system. 



But 'in all (or most) cases where, but a few years ago, the 

 concepts of secretion or excretion seemed precise enough, we are 

 now-a-days inchned to turn to the phenomenon of adsorption as 

 a further stage towards the elucidation of our facts. Here we 

 have a case in point. In the tissues of our sponge, wherever two 

 cells meet, there we have a definite surface of contact, and there 

 accordingly we have a manifestation of surface-energy ; and the 

 concentration of surface-energy will tend to be a maximum at 

 the lines or edges whereby the three, or four, such surfaces are 

 conjoined. Of the micro-chemistry of the sponge-cells our 

 ignorance is great; but (without venturing on any hypothesis 

 involving the chemical details of the process) we may safely assert 

 that there is an inherent probabihty that certain substances will 

 tend to be concentrated and ultimately deposited just in these lines 

 of intercellular contact and conjunction. In other words, adsorp- 

 tive concentration, under osmotic pressure, at and in the surface- 

 film which constitutes the mutual boundary between contiguous 



