THE GENTJS ALTEOPOEA WITH THE TAYOSITID^. 513 



real cause of the difficulties experienced by Prof. Lacaze- 

 Duthiers (J. c.) whenever the epitheca comes into the dis- 

 cussion. 



On the other hand, in his account of the development of his 

 specimens o£ Balanophyllia and of Leptopsammia, Prof. Lacaze- 

 Duthiers records not only what he had already recorded of Lis 

 specimens of Astroides, that the first calcareous secretions belong 

 topographically to the septa, but that an epithecal wall soon 

 appears round the young polyp. He then noted that the septal 

 skeleton overran this (1st) wall, which tended to limit the size 

 of the base too much, and that then another epithecal wall (2od 

 wall) appears, which may be again overrun, when a third appears 

 round the base, which has been in this way greatly expanded. 

 This observation seems to me of considerable importance, and I 

 am fortunately in a position to confirm it. 



An examination of almost any group of young single corals 

 developing on the corroded parts of other specimens vrill show 

 the periodical overflowing of the epithecal cups as the individuals 

 outgrow them. Pigure 11 shows one such, overflowing excen- 

 trically. The first wall has been obscured by the columella. 

 One quite overrun wall is, however, distinct, and the last formed 

 is being overrun. Cases in which the successive walls are quite 

 concentric have also been noted, and are probably the normal 

 method of expanding the base of attachment. The need for 

 widening the base of the primitive epithecal cup, which must 

 otherwise continue to expand, cone-like, as the polyp grows, fully 

 explains this remarkable process. In some cases it is probable 

 that the overflow was strictly limited and local, and was then re- 

 covered over by epitheca, in which case the conical cup would be 

 propped up by rootlets, as in Uliizotrochiis. I would further 

 venture to make the suggestion that if the primitive conical coral 

 skeleton frequently adopted this overflowing method of widening 

 its base, the overflowing being on one side only, the radial arrange- 

 ment of the septa would be interfered with, as we see to be 

 the case in fig. 11. Persistence of the bilateral symmetry thus 

 acquired could, I think, be made to explain some of the more 

 anomalous arrangements of the septa among Palaeozoic corals 

 (e. g. Streptelasma). 



