Morphology of the Madreporaria. 113 



Aladrepora and Porites. The definite establishment of this 

 is a matter of the highest importance. 



Recently (1902), I have been able to confirm the observa- 

 tion of Pourtal^s (1871) as to the primary hexameral character 

 of the Rugose coral Lophophyllum proliferum (McChesney), 

 and by means of serial sections have definitely established 

 the metaseptal sequence. The structure of Paleeozoic corals 

 so nearly conforms with that of recent forms as to warrant 

 the assumption that the relationships of the polyp to the 

 corallum were the same as those now thoroughly understood. 

 The metaseptal arrangement in Lophophyllum is such as to 

 indicate that the mesenterial additions took place successively 

 at one region only within four of the six primary exocoelic 

 chambers ; two of tlie chambers are the middle exocoeles and 

 the other two are the exocoeles on each side of one of the 

 axial (directive) pairs of j)rimary mesenteries; no additions 

 were made in the two remaining primary exocce'es. So far 

 as the septal arrangement in Lophophyllum can hd taken as 

 representative of the Rugosa generally, the order is shown to 

 be most closely related to the living Zoanthese. In this group 

 of mainly tropical and colonial Actinians the metacnemes are 

 added in succession at one region within the primary exocoele 

 on each side of the ventral pair of directives, whereas in the 

 Rugosa they were added in two middle primary exocoeles iu 

 addition. 



With their entoccelic metacnemes, therefore, Porites and 

 Madrepora cannot yet be brought into conformity with either 

 the extinct Rugose corals or the living Zoanthid polyps, 

 seeing that these .are characterized by exocoelic additions. In 

 one feature, however, they all agree. The metacnemes in the 

 Zoanthese and in Porites and Madrepora arise successively, a 

 bilateral pair more or less simultaneously on opposite sides 

 of the polyp. In the end they are arranged on each side of 

 the polyp so as to constitute unilateral pairs consisting of a 

 complete and an incomplete moiety (anisocnemic), whereas 

 in ordinary Hexactinians the metacnemic unilateral pairs 

 are constituted of equal moieties which appear together 

 (isocnemie). 



The Cerianthids present in their mesenterial growth a 

 method which most nearly approaches that under considera- 

 tion, though differing in several important details. While 

 Actinian writers such as Boveri and McMurrich hold that the 

 first four pairs of mesenteries to arise represent the four 

 Edwardsian mesenteries, van Beneden (1897) considers that 

 no such relationship can bo maintained, and the results of 



