LACAZE-DUTHIERS ON CORAL. 365 



nation, the author confesses he can say little more than that it takes 

 place not only within the body of the polyp but also within the 

 ovary, and that this is always the case in both the Alcyonaria and 

 Zooantharia. In stating that unimpregnated ova are never dis- 

 charged from the Zooantharia, we think the author goes too far. It 

 is perfectly true that the Actiniae, to which reference is made on 

 more than one occasion, commonly give birth to their young in a 

 more or less developed condition, but we have reason to know that 

 certain species also discharge ova which show no trace of segmenta- 

 tion having taken place, and which have no power of independent 

 motion. This locomotive power is one of the earliest characters 

 displayed by the embryo polyp, and the one mentioned by the author 

 as being specially indicative of larval condition. In the case of 

 Caryophyllia among the true coralligenous polypes, we can also speak 

 to the fact of the occasional discharge of ova, in which the ger- 

 minal versicle may be distinctly visible, and with none of the or- 

 dinary signs of impregnation having taken place. Such instances 

 have occurred in small salt-water tanks, in which the water had been 

 unchanged for several years, and we see no reason to differ from the 

 author when he states his grounds for believing that the habits and 

 behaviour of marine animals in the aquarium do not present any 

 variation from what might be observed in them if they could be 

 studied at the bottom of the sea. In the same tank that contained 

 the oviparous Caryophyllia, we have also seen a specimen of JBalano- 

 phyllia produce young in the earliest larval condition, their shape 

 being slightly oval, and not undergoing any material change before 

 they became attached to the stones. 



"We may pass over the author's account of the gradual de- 

 velopment of the coral polyp, and turn to his interesting de- 

 scription of the production of the coral {poly pier) itself. After 

 acknowledging how much naturalists of the present day are indebted 

 for many of their discoveries to the perfection of modern micros- 

 copes, he points out that no trace of spicules can be found until the 

 first polyp — the direct produce of the ovum (oozoiie), has been fully 

 matured, and that Donati was certainly mistaken in believing he saw 

 them in the larvae. The spicules of the sarcosoma have from the 

 first a constant and characteristic arrangement, as had been 

 previously observed by Valenciennes, and this special character was 

 of great use in tracing the growth of the actual coral. The long 

 standing dispute as to whether the origin of the coral was in the 



