14 BRITISH FOSSIL CORALS. 



The septal columellae may be mistaken for tlie fascicular and essential ; but a longitu- 

 dinal section will show that the inner edges of the septa forms tbe organ, and that it does 

 not arise from the base. 



The parietal columellae are very common, and their structure is illustrated (Plate IV, 

 fig. 13; Plate VII, fig. 9). 



The calicular surface of the columella may be prominent or depressed, papillary or 

 spongy ; and the organ may be very dense or consist of very lax tissue. 



The columellae of the following genera may be studied with regard to this variety : — 

 Parasmilia, Eusmilia, Dendrosmilia, LitliophyUia, Circophyllia, Jiliabdopliyllia, Mceandrina, 

 Manicina, Diploria, Heliastrcea, Solenastreea, &c. 



As a general rule, when pali exist, they are in close contact with the columella, and 

 as they spring from the base they often look like lateral processes of essential columellae. 

 It will be observed, in the descriptions of living corals, that the columella fills up much of 

 the visceral cavity, and is developed by the inner layer of the soft tissues. Playing a very 

 important part in the economy, and being in relation both with the septa and pali, the 

 columellae are structures whose variations in form are of generic import. 



CostcB. — The costae may be considered in a general sense to be the continuations of 

 the septa beyond the wall.^ 



In some Tarhitiolim the continuity between the costae and the exsert septa is very 

 evident, and both of the structures are much higher than the upper margin of the wall.' 

 But it is very probable that this exsert condition of the septa and costae is to be referred 

 to the corallum having attained its full development as regards height ; the further up- 

 ward growth of the wall was arrested, and only the combined costo-septal apparatus 

 grew on. For when the costae of the same specimens are broken oflT low down, it is 

 tolerably evident that the wall intervened between their bases and those of the corre- 

 sponding septa. 



It would appear that the costae and septa are not developed by the same parts 

 of the soft tissues except when they are exsert and above the wall; and the want 

 of correspondence between the septa and costae about to be mentioned is in consequence 

 of this. 



It is probably quite correct to give the costae an origin independent of the septa, and 

 to assert that the}/ are frequently separated by the thickness of the wall from the septal 

 laminae. 



The costae are developed by the inner layer of the tissue which covers the wall 

 externally, and the outer surface of the wall and the exothecal structures are also formed 

 by it. The costae follow, as a rule, the cyclical development of the septa, and are called 

 primary, secondary, &c. 



1 Plate I, figs. 2, 6, 7, 11, 15, 18. - Plate I, tigs. 6, 14, 15, 18. 



