28 BULLETIN OF THE BUREAU OF FISHERIES. 



regenerative tissue. The amocbocytes have a heterogeneous origin, some representing 

 the wandering cells of the normal sponge, many more representing transformed (reduced) 

 pore cells. It may be remarked that our knowledge concerning the very existence 

 of specialized pore cells is exceedingly inadequate except in the case of the ascons. I 

 have, for instance, been utterly unable to find them in monaxonids such as StylotcUa 

 and Reniera."^ It is evident then that the process of regressive differentiation can not 

 pursue quite the same path in Stylotella that Maas maps out for the Calcarea. The 

 large question involved is of course : Do the several kinds of cells, preserving their nature, 

 struggle with one another for the mastery, certain kinds or one kind absorbing and 

 digesting others, and so growing and forming the regenerative mass? In support of 

 this idea it is to be noted that Maas and Miiller agree in finding that the choanocytes 

 are absorbed and digested by amcebocytes. Or when the flagellated chambers, the 

 canal epithelium, the epidermis perhaps, all break up into cells which wander away 

 from one another and help to form the mesenchyme-like syncytial tissue of the reduced 

 trabecula, does the protoplasm of all these cells undergo a reconstruction, a sort of 

 rejuvenescence, whereby they all pass into the condition of the unspecialized, generalized 

 plasm of that species, the masses of this plasm fusing intimately to form the regenerative 

 tissue? This is the interpretation of the facts which I have favored in my paper on 

 the subject.* 



Turning now to the phenomena that follow upon the sudden violent isolation of 

 sponge cells, their rapid fusion to form masses physiologically similar to those produced 

 in the slow process of regressive differentiation that goes on in confinement, the same 

 question meets us. Amoebocytes, hyaline elements, and choanocytes all combine to 

 form the plasmodial masses."^ Do the amcebocytes absorb and digest the other elements ? 

 Or do all the cells as a result of the shock pass into the generalized protoplasmic state 

 and persist as parts of the regenerative mass? A careful histological study might 

 enable one to answer this question. Meantime it seems to me that the latter hypothesis 

 receives support from my recent observations on the fusion of isolated cells in hydroids."* 

 In hydroids the body is made up of two specialized layers and there are comparatively 

 few cells present which correspond in this matter of regenerative ability to the amcebo- 

 cytes of sponges. I have found that a Eudendrium colony may be cut into pieces and 

 pressed out after the fashion described in this paper, and so broken up into cells, minute 

 cell masses, and possibly cell fragments. Fusion between these elements goes on and 

 plasmodial masses are formed which secrete a perisarc. Such masses throw out hydro- 

 rhizal outgrowths which in successful cases develop perfect hydranths. The same 

 phenomena were observed in Pennaria when only the stem was cut up, the regenerative 

 mass being thus exclusively derived from the ccenosarc. In these cases it might, to be 



« Wilson. H. v.: A study of some epithelioid membranes in monaxid sponges. Journal of Experimental Zoology, vol. rx 

 1910, 



*> Wilson, H. v.: A new method by which sponges may be artificially reared. Science, n. s.. vol. xxv. June 7, 1907. 



c Wilson, H. v.: On some phenomena of coalescence and regeneration in sponges. Journal of Experimental Zoology, vol. 

 V, 1907. 



d Wilson. H. v.: On the regenerative power of the dissociated cells in hydroids. Proceedings of the American Society of 

 Zoologists, Science, n. s.. vol. x.xxiu. Mar. 10, 1911. 



