208 R. F. Tomes—On Heterastrea, Lower Lias. 
light other characters. A great development of endothecal structure 
sometimes almost filled up the loculi, and the walls of the corallites 
were observed to be of great thickness. Horizontal sections gave 
the proper interpretation of these thickened walls, and it was often 
seen that there was not only a fine line where the walls of the coral- 
lites came in contact, but that there was occasionally a visible 
interval between them. In this particular they resembled Septastrea, 
while at the same time their increase was by gemmation. Closer 
observation revealed the fact that there was both gemmation and 
fissiparity in nearly all the species examined; and, in a word, that 
they possessed characters which were inconsistent with both Isastrea 
and Septastrea, and formed a group of themselves. 
Before proceeding to define the characters of the group above 
indicated, I take the opportunity of making some remarks which 
bear, though only generally, on the species under consideration. 
Having repeatedly, both on former occasions and in the present 
communication, followed Milaschewitsch in the use of a word which 
is rendered in English by the word rejuvenescence, I am induced to 
give the following explanation of the process, because it is absolutely 
necessary that there should be a very clear conception of its impor- 
tance in one of the genera of which I shall speak. That genus is 
Elysastrea. ‘That it is not merely intermittent growth, and that it is 
not gemmation, has been most clearly shown by Milaschewitsch, but 
what really gives rise to it has not been explained by him. It was 
certainly noticed by Ehrenberg, and was long ago known to Dana, 
though he does not appear to have fully understood its nature. In 
the third volume of the American Journal of Science (second series) 
he speaks as follows :— 
“In some Cyathophyllidz the process of death goes on inter- 
ruptedly, as explained by Ehrenberg. The tissues of the polyp 
disappear at intervals from the sides of the corallum, or become dead, 
leaving a row of unoccupied cellules; then the animal goes on to 
increase from its contracted size, without refilling the cellules; the 
corallum consequently becomes covered with encircling ridges, or 
appears as if formed of a series of inverted cones. In some cases, 
as in the species referred to the genus Strombodes, the living portion 
becomes retracted at intervals to the very centre, all the rest dying, 
and afterwards the animal grows again and spreads to its original 
take place within the calice, as well as on the margin of the wall which surrounds it. 
Thus in the descriptions of IJsastrea endothecata and I. latimeandroidea, Prof. 
Duncan says it is marginal (‘‘on the margin’’ are the words made use of in the 
description of the latter species). More recently, however, he has asserted that ‘‘ the 
gemmation of Jsastrea certainly does not take place between the walls of corallites, 
but within the calicular margin; it is between the margin and the centre of the 
calice.”’ I am not aware that it has ever been spoken of as occurring between the 
walls, though I, as well as Prof. Duncan himself, have spoken of it as occurring both 
inside the calice and on the margin of the wall.. MM. Milne Edwards and Haime 
(Hist. Nat. Coral. vol ii. p. 626) speak of it in these words, ‘‘ Les polypierites se 
multiplient par gemmation calicinale et submarginale,’’ while M. de Fromentel says 
itis ‘‘submarginale.’’? The present communication by removing the Liassic species 
from the genus Jsastrea altogether, will materially lessen the difficulty of future 
classification. 
