514 mr. h. m. bernard on the affinities of 



Summary. 



Alveopora agrees in all essentials of skeletal structure with. 

 Favosites, and there is no other reason to doubt their genetic rela- 

 tionship than that which arises from the interval of time with only 

 the little-known Cretaceous genus Koninchia to connect them. 



A comparative study of Madreporarian skeletons leads to the 

 conclusioa that the skeleton of Alveopora is at a very low level 

 of development, and on that account alone might be regarded as 

 the survival of a Palaeozoic type. 



In this respect Alveopora stands far away from the Poritidas, 

 with which it is usually classed, this latter standing almost at 

 the opposite extreme of high specialization. 



The Madreporarian skeleton may be described as the rigid 

 secretion of the basal portion of the columniform body of a polyp 

 into which the flexible upper portion may be invaginated. lu 

 its earliest development a simple cup, it has become complicated 

 in various ways: primarily, by the development of radial infoldiugs 

 of the stiff external wall, comparable with the infoldings of the 

 chitinous cuticle of Arthropods; secondarily, (1) by further com- 

 plications of these infoldings so as to form an intricate " internal " 

 skeleton, which may render the primitive external cup unneces- 

 sary, and hence lead to its becoming vestigial ; (2) by a process 

 of repeated sheddings of the external hard secretions, and the 

 formation of new ones across and among the existing " internal " 

 skeletal structures (dissepiments and tabulse). 



The growing soft parts may overflow this primitive rigid cup. 

 The bagging thus occasioned may reach the substratum and widen 

 the base of attachment, or may merely hang down, secreting in 

 so doing a folded rim to the cup ('eutheca' of v. Heider). This 

 bagging is quite distinct from the peripheral region of the body 

 which surrounds the " internal " skeleton above mentioned, and 

 the word Randplatte cannot be applied to both. It would be 

 better to apply it to the bagging process, which has some 

 distinct morphological importance, whereas the alternative has 

 only topographical utility. 



I am disposed to believe that many " Astrseidse " will be found 

 to have epithecate walls, owing to colony formation starting from 

 this bagging of the soft parts over the primitive cup. All such, 

 if this is correct, will have to be separated widely from those 

 forms in which the theca is dissepimental iu origin. 



