Actiniae. 303 



from body-wall to a'sophagus, with lubes of the iliiuiil)er i)rojecting 

 between them. It will be noted Irom the two sections iigured 

 that this consecutive series of horizontal sections includes a pair 

 of directive mesenteries. Each chamber is completely filled with a 

 compact mass of rounded cells, exceedingly rich in food yolk and 

 measuring from 0*3 to 0*4 mm. in diameter. Usually distinct cell 

 membranes are present, but in one specimen examined the limiting 

 membrane is very indistinct. I examined three specimens of about 

 this stage and find each somewhat different in the character of these 

 cells. The difference is mainly in the appearance of the cell mem- 

 brane and the greater or lesser granular appearance of the cell 

 contents. In one specimen the outer layer of cells abutting on the 

 wall of the brood chamber are roughly cuboidal in shape and very 

 much freer from food yolk than the more centrally placed cells, 

 suggesting a more rapid assimilation of the food yolk into 

 protoplasm in these cells. 



I cannot but regard each of these cell masses, notwithstanding 

 the relatively enormous size of the constituent cells, as a deve- 

 loping embryo. It is known, although considerable uncertainty 

 still prevails regarding the earlier developmental processes, that 

 in many cases among the Anthozoa cleavage results in the 

 formation of a solid morula, and Kowalevsky's observation (4) on 

 Actinia parasitica (Adamsia rondeletii) supports this. He says : 

 •' Cleavage is regular, but as the result of it there arises not a 

 blastodermic vesicle, but only an aggregation of cells, which 

 becomes covered with cilia and swims about as a larva." The 

 researches of E. B. Wilson on the development of Rcnilla (8) show 

 that here also a solid morula is formed, consisting of irregularly 

 rounded cells, heavily charged with food yolk, and in which at first 

 no differentiation exists, but where later an outer layer gradually 

 becomes marked off in character by appearing less granular, throuo-li 

 the conversion of the contained food yolk into protoplasm, and so 

 forms the ectoderm. Unfortunately, except the specimen showing a 

 slight differentiation of the outer layer of cells, I have found no 

 intermediate stages between the condition just described antl the 

 larva with all the primitive layers formed. 



A specimen of U. sulcata possesses four embryos in the brood 

 chambers, which may be described as bilaminar, pyriform, or more 

 elongated vermiform planuLnp, possessing an ectoderm and an 

 endodermic epithelium, with a sustentative lamella (mesoglcea) 

 secreted between them, but with no trace of cilia so characteristic 

 of free-sicimming planulre. The interior is filled with a dense mass 



