78 ZOOLOGY. 



mouth, is found in their midst in such a way that its two 

 ends have a tentacle opposite each, and the other four dis- 

 posed two on one side and two on the other. Within, the 

 organs arise at points corresponding to the position of those 

 outside. The semi-partitions, twelve in number, begin as 

 mere ridges, which extend in pairs from the anterior end of 

 the stomach along the oral wall toward its border." Adult 

 Actiniae sometimes, though rarely, subdivide longitudinally, 

 but it is not uncommonly observed in the corals, in which 

 cases only the heads and stomachs divide, the general cav- 

 ity remaining common to the two. 



The development of Actinia meseiribryantliemum has been 

 traced by Lacaze-Duthiers. The young Actinia attains 

 maturity without any metamorphosis. The egg is supposed 

 to undergo segmentation within the ovary. In the state in 

 which the embryo was observed by .Lacaze-Duthiers it was 

 oval and surrounded by a dense coat of transparent conical 

 spinules. Soon the two primitive germi- 

 nal layers (ectoderm and endoderm) 

 were observed. Two lobes next appear 

 within the body ; these subdivide into 

 four, eight, and finally twelve primitive 

 lobes. This stage is represented by the 

 corresponding stage of the coral (Fig. 55, 

 E). Not until after the twelve primitive 

 lobes are fully formed do the tentacles 

 Fig. si.- -ciliated larva begin to make their appearance. When 



(gastrula) of a Polyp, a, , , ~ . . , 



primitive opening or bias- the first twelve tentacles have grown out, 



topore ; 6, stomach ; c, , j> -, , -, 



ectoderm; a, endoderm! twenty-four more arise, and so on, until 

 with its increasing size the Actinia is 

 provided with the full number peculiar to each species. 

 Lacaze-Duthiers observed the same changes in two species 

 of Sagartia, and in Bunodes gemmacea. Fig. 51 represents 

 the ciliated gastrula of an unknown polyp allied ivKalliphobe. 

 While Metridium and Bunodes are types of the ordinary 

 form of Actinoids, certain forms, like Halcampa producta 

 Stimpson (Fig. 52), are quite long and live fixed in the 

 mud or sand. Allied to Halcampa is Edioardsia, which 



