MADBEPOEAEIA OF THE DEEP SEA. 329 



In a note, the thick walls of the corallites are referred to, as well as the existence of 

 " traverses." The genus is admitted into the family of the Oculinidae by MM. Milne- 

 Edwards and Jules Haime; and they state, p. 102: — "The visceral chambers [of the 

 Oculinidae] only exhibit a small number of ' traverses,' or of incomplete 'planchers;' 

 but they [the visceral chambers] tend to become narrow and contracted inferiorly, and 

 even to fill up from below upwards by the growth of the wall, and sometimes of the 

 columella also." Subsequently (same page) the authors state that the wall has neither 

 " traverses " in its interior, nor epitheca on its surface. 



Now it is most important to distinguish between the words " traverses " and 

 " planchers," even when the latter are said to be incomplete. 



" Traverses endothecales " correspond to the English term of endothecal dissepiments ; 

 and they extend between the septa, closing up the interseptal loculi ; but they do not 

 extend across the axial space in one uninterrupted tissue, so as to shut off the part 

 below the calice above. 



"Planchers," or "tabulas," or "diaphragmes lamellaires," close the corallites at different 

 heights, and form horizontal divisions, which extend from one wall to the other of the 

 general visceral cavity, instead of occupying the breadth of the interseptal loculi above. 

 The following words, however, complicate the matter somewhat : — " they [the planchers] 

 differ from the ' traverses,' because " ils ne dependent pas des cloisons," although the 

 reference to the dissepiments which only close the interseptal loculi between the septa 

 follows. 



Tabulae or " planchers " are independent of the septa, but they are formed by endo- 

 thecal growth after the septal laminae have been perfected in the visceral chamber. 

 Then this is crossed by a development of hard structures within the membranes, and 

 all below is shut off. The tabulae extend across the axial space and between the septa, 

 and in the interseptal loculi. An incomplete tabula is an anomaly, but may exist as the 

 result of antecedent or of imperfect development. 



The specimens of this genus are very numerous in the collection of dredgings ; and I 

 have had the opportunity of examining others from the American expedition, which 

 were dredged off the Florida coast. M. Sars has sent me a Norwegian specimen ; and 

 Mr. Kent has shown me the beautiful series obtained by him south of Cape Finisterre, 

 in 600 fathoms. The drawings of the fossil species described by M. Seguenza have 

 been studied; and I have availed myself of a number of recent specimens obtained 

 from unknown localities. 



The result of my study of a great series of forms which must be included in the 

 genus, is to prove that the corallites do not fill up inferiorly, but that the wall increases 

 externally, that small dissepiments often exist, that perfect tabulae are by no means 

 imcommon, and that some specimens have no endotheca whatever. 



The variability of the species is immense ; and the corallites on the same stem are 

 also variable. 



3a2 



