26 BRITISH FOSSIL CORALS. 
tentacules these colour-bearing cells are arranged in small irregular groups, but elsewhere 
they become more numerous. 
‘The internal membrane lines the interseptal loculi, where its presence is rendered 
evident by its colour ; it is stopped inferiorly by the last sclerenchymatous dissepiments. 
The mesenteric folds formed by this membrane present a few colour-cells. The folds’ give 
attachment to the simple “ boyaux pelotonnés” which float in the large interseptal loculi 
along the smaller septa, and which often show themselves on the columella when the 
mouth is half open and the polype is slightly contracted. ‘Their walls are almost entirely 
composed of nematocysts of the largest size, and their surface is furnished with large 
and strong cilia; they are frequently affected by peristaltic movements, and they are 
attached to the tentaculiferous disc by strong muscular fibres.” 
39 
IV.—Repropuction anpD MuvLtiIrLication. 
Ovular Reproduction ; Gemmation ; Fissiparous and Serial Growth ; Reproduction. 
The mesenteric folds and the twisted tubular processes, whose ends are free in the 
visceral cavity, appear to be the organs which develop the male and female elements. 
It would appear that all corals are not bisexual, but the majority are so. Spermatozoa 
were asserted to exist in the ¢ubular processes, but their description tallied with that of the 
thread-processes of xematocysts. Milne-Edwards dispelled this illusion, and the true male 
elements have been discovered. The presence of ova in the mesenteric folds and in the 
tubular processes has been noticed and in the latter position by Michelotti and Duchassaing 
in large compound corals.” The ova are matured in the folds and processes, and then 
escape into the visceral cavity, and are expelled through the stomach and mouth. They 
have some power of active locomotion, and select favorable localities for their resting-place. 
The young polypes have faint traces of the future sclerenchyma, and grow rapidly when 
once fixed, provided they are well nourished. 
As growth proceeds, the structure of the wall determines the shape of the corallum ; 
and its simple or compound character is regulated by the particular methods of the mul- 
tiplication of the individual. Some corals are always simple or solitary, others for a 
considerable period, and some for a very short time. The kind of gemmation or budding 
determines the massive, dendroid, encrusting, &c., nature of corals. 
It appears to be very rare for buds to fall from the parent corallum and to form inde- 
pendent individuals. 
By gemmation is meant the development of corallites from the tissues of a parent 
corallum. A very small patch of the membrane in immediate contact with the scleren- 
chyma of the parent appears to pucker, and septa are rapidly formed within the enlarge- 
ment which occurs; tentacules have already appeared, and the small bud proceeds as if it 
AtPlate Il, fig. 2: 2 Op. cit. 3 Op. cit. 
