200 THE VOYAGE OF H.M.8. CHALLENGER. 
In a young specimen of this coral, without reference to which the true nature of the 
complicated structures in the adult could hardly have been determined, the corallum is 
almost symmetrically biconvex and perfectly circular; but the columella is elongate, as 
in the adult. The columella is papillar rather than laminar, and represents the rooted 
base of the adult columella. There are sixty perfect costa present at the margin of the 
calicle, and the commencement of two more in each system (¢.e., twelve in all). There 
are twelve concentric rows of perforations in the base. The septa are not nearly so 
much contorted as in the adult, and the synapticular junctions between them not being 
as yet formed, the division of the septal cavities into successive chambers is not apparent 
as in the adult. No superficial fusion of the primary with the curved extensions of the 
tertiary and quaternary septa has as yet taken place; hence no six-rayed star or flower 
appears in the centre. Stout transverse granular projections are present on the upper 
margins of the septa, the sources of future synapticular connections. Only four cycles are 
complete in cach system. Two members of the fifth cycle are present in each system, and 
are those nearest the primary septa on each side. An additional pair of quinary septa is 
just commencing to grow in each system, and branching off from the peripheral ends of 
the quaternary septa which are next the secondary, but on that side of them which is 
nearest the primary, @.e., opposite to the secondary. A very short branch of the costa 
belonging to the interspace between the quaternary and secondary septa passes into each 
interspace formed between these newly-grown quinary septa and the quaternary adjoining. 
In the development of a flower-like series of oval chambers around the elongate 
columella, this coral most strikingly resembles Stephanophyllia florealis of Quenstedt,! 
which is a fossil of the ‘“‘ White Jura” (=Oxford Clay) formation. The oval chambers 
here in the first series around the columella are twelve in number, and alternate with 
twelve in the second series. In Stephanophyllia complicata, six alternate with six. In 
Stephanophyllia florealis there appears to be no trace of the bisection of these chambers 
by the straight primary and secondary septa as in Stephanophyllia complicata ; but the 
fossil specimens are, according to Quenstedt, always so much mutilated that their struc- 
ture can only partially be made out. Possibly the twelve chambers of Stephanophyllia 
Jlovealis may represent the twelve of Stephanophyllia complicata, formed by the bisection 
of the six inner by the primary septa, which might appear as figured in Stephanophyllia 
Jlorealis in much-worn specimens. Comparison with actual specimens may determine 
this point. In Stephanophyllia florealis forty-eight septa only are distinguishable ; but 
it is highly probable that structures so slight as the quinary septa of Stephanophyllia 
complicata might be indistinguishable in a mutilated fossil. The coral appears to fall 
into that division of the Stephanophyllias distinguished by Milne-Edwards as having the 
? Milne-Edwards and Haime, judging from Quenstedt’s figure, Handb. der Petrefact, p. 657, pl. lix. figs. 12, 13, 
1852, wrongly supposed that Stephanophyllia florealis should be referred to the genus Thecocyathus, Hist. Nat. des Coral. 
t. i, p. 49. F 
