328 Mr. E. Ray Lankester on the 



pointed out the similarity of these layers to the two primitive 

 layers of the Vertebrate embryo. 



Notes to B. — The difference in the two modes of origin of 

 the Planula may be due to the dropping of the invagiuation- 

 process as a shortening of the developmental process — that is 

 to say, in obedience to the tendency to a direct as opposed 

 to a recapitulative development. It is, however, to be noticed 

 in connexion with this that in the later development of 

 special organs we have examples where development occurs 

 sometimes by invagination and sometimes by simple accretion, 

 and where the bulk of the developing structure appears to 

 determine the invagination. 8uch, for instance, is tlie case 

 with the otocysts or auditory capsules of mollusks. In the 

 Nudibranchiates I have satisfactorily determined that their 

 cavity does not arise by invagination. On the other hand, in 

 the Cephalopod Zo%o I have found (what was previously sus- 

 pected but undemonstrated) that the otocyst is formed by an 

 invagination, the ciliated canal connected with it being a 

 remnant of its external communication. The development of 

 the nerve-centres also furnishes examples. In Loligo I have 

 observed that the cephalic ganglia originate each by invagina- 

 tion and formation of a groove and cavity. In Gasteropods 

 the corresponding ganglia form by simple thickening of the 

 outer layer of cells. The origin of the cerebro-spinal nerve- 

 centre of Vertebrates and certain Tunicates, as compared with 

 that of Arthropods and notably of certain Annelids [Lumbricus^ 

 and Euaxes as described by Kowalewsky), offers the same 

 contrast. 



It is remarkable that the origin of the primitive gastric 

 cavity by invagination has been more widely observed in the 

 higher groups, and that in most Coelenterata as yet studied the 

 cavity is formed directly. There are exceptions to this among 

 Coelenterata ; but in this subject it must be remembered that 

 we have as yet very few adequate observations. Among the 

 higher groups the observations of Kowalewsky have especially 

 established the occurrence of this primitive invagination in 

 Amjjhioxus, in Tunicates, and certain Vermes ; my own ob- 

 servations (as yet unpublished) have proved its wide-spread 

 occurrence in Mollusca, viz. in the Lamellibranch Cyclas 

 pjisilla, in several Nudibranchs {PoJycera^ Eolis, Doris, Pleuro- 

 hranchus) , in the Pulmonates Arion and Limax. The presence 

 of accessory yelk is what, more than any thing else, appears 

 among the Mollusca to be associated with the suppression of 

 the invagination-process. The anus of Rusconi in the de- 

 veloping Batrachia among Vertebrata represents the orifice of 

 invagination in a somewhat modified condition. 



