rrnnanj Srptal Phtii cf tlic. Iiti'josa. oGl 



'I lie linver Palitozoic corals in flic above list all had to be 

 .scclioiicHl lor the microscope bifore the early sitpta were 

 visible, ('oiiseqiieiitly fewer specimens were examined than 

 in ihf Carbtniiferons corals, of whieli nearly two hnndred were 

 obtained, showing Stages I. to III. But enough were 

 obtained to show that these early Pahuozoic forms agreed 

 exaetly in their initial stages with the Carbouiferous ones, a 

 conclusion of interest in view of Gordon's suggestion that 

 there were four primary sejjta in the early llugosa. But, 

 using the length of the se[)ta as a guide to their age, an 

 examination of Gordon's own figure* of an Ordovician 

 Streptthisma shows that it is in perfect accordance, from 

 Stage I. onwards, with the septal sequence above described; 

 the tendency to a spiral growth of the primaries is perhaps 

 slightly more marked than usual, though this may well be 

 an accident of jireservation. 



Except for Dibunopliyllum and Ciiclophyllum, all the 

 corals examined were small forms. This is, of course, simply 

 due to the fact that in large, heavy forms the delicate tip is 

 more easily destroyed by rolling or movement of any kind. 



In no instance have I found any essential divergence in 

 the arrangement of the primary septa in the genera examined ; 

 and from the uniformity dis[)layed in this respect, even in 

 corals far removed in point ot time, it seems reasonable to 

 suppose that the sequence will be found to extend to the 

 liugosa as a whole, and that Duerden and Pourtales were 

 right in considering these corals to b^ primarily hexamerous. 

 Similar instances of the formation of the early septa in 

 bilateral pairs have been observed in certain Mesozoic Hexa- 

 corallidsf, but, indeed, the great majority of the Zoantharia 

 seem to have a primary hexanierous plan, attained in most 

 cases by the insertion of bilateral pairs. It is in the later 

 stages that, as Duerden remarks, " divergences are introduced 

 which are to be regarded as of the most fundamental im- 

 portance in coral and actinian morpliology," 



While these results bring the Kugosa into closer associa- 

 tion with modern corals and dismiss the idea that they are 

 primarily tetramerous, and as such built on a more primitive 

 basis tlian other Madreporaria, yet there are reasons to 

 sujipose that septal formation in these ancient corals took place 



* Loc. cit. fig. 16, p. 120. 



t See especially M. M. Ogilvie, " Microscopic and Systematic Study 

 of Madreporarian 'Types of Corals," Phil Trans, clxxxvii. p. 291, diag. F 

 & G (1896) ; and also pp. 97 & 105 of ' Die Korallen d. Stramberger 

 Schicbteu," Stuttgart, 1897 (by the same authoress). 



