MORPHOLOGY OF THE MADREPORARIA. 47 



the fossula, as a sort of crypt for the sexual products, is prob- 

 able enough, considers that this need not have been the cause of 

 its origin. Rather, he would explain its presence on the mechan- 

 ical assumption that at an early stage in its development the coral 

 falls over, and in recovering the upright position the soft polypal 

 parts detach themselves from the base of the prototheca in such 

 a way as to bag down, and thus produce a pit or depression on 

 the floor of the calice. Bernard thus expresses it : " The fossula 

 has a very simple explanation, if the assumption of the falling 

 over is correct. As the soft parts detach themselves from the 

 base of the prototheca they might be expected to bag down, and 

 they will continue to be acted upon by gravitation and drawn 

 over towards the convex side of the coral until the vertical posi- 

 tion has been regained. It is possible that this bearing over to 

 the side may be due to the efforts of the polyp itself to bend up, 

 but gravitation in a causa efficicns." 



Such a purely mechanical explanation is very unsatisfactory 

 when we consider how hypothetical is the conception of any 

 general falling over of the prototheca, and also of the influence 

 which this would have upon the polyp itself as well as upon the 

 corallite ; it is difficult to conceive of the bagging down of the 

 polypal tissues always at a definite region of the calice on each 

 side of the cardinal septum. The view which I have submitted 

 above, that the simple cardinal fossula is correlated with the pres- 

 ence in the living polyp of a gonidial groove or siphonoglyph, 

 is more in harmony with the facts of anthozoan morphology, 

 while there can be no question as to the significance of the 

 grouping. 



Bernard fails to see the evidence for the existence of more than 

 one true fossula in any coral examined by him. He finds this 

 usually on the convex or "dorsal" side, this being the side 

 which I have here termed ventral, as being more in agreement 

 with the accepted anthozoan terminology. The occasional pres- 

 ence of the fossula on the concave side he attempts to explain as 

 dependent upon a shallower more open prototheca than that in 

 which it forms on the convex side. 



Some of Bernard's other statements seem so contrary to all 

 that we know of the nature and development of the Rugosa that 



