340 E. W. MACBRIDE. 



blood system, and has no connection with the gut whatever. In 

 many Coelenterates the stage of the hollow blastosphere is 

 missed out, and segmentation results in a " morula " of which 

 the external layer is the ectoderm and the rest endoderm. I 

 think we must imagine that in the development of Geryonia 

 the shortening process has gone one step further, and that as 

 a result of segmentation we reach at once the stage of the 

 hollowed-out planula. 



Sedgwick's hypothesis was suggested from a study of the 

 embryos of Peripatus capensis. Their developmental his- 

 tory is, however, the very last place where one ought to seek 

 for indications of the ancestral meaning of the earlier stages 

 — at any rate if Sedgwick's own hypothesis as to the significance 

 of the embryonic phase be correct. All species of Peripatus 

 so far as is at present known are oviparous; in Peripatus 

 Novae-Zelaniaj, however, the eggs are large and yolky, and 

 the development conforms to the ordinary centrolecithal type 

 so characteristic of Arthropods — the peculiarities of which we 

 have described above. In Peripatus capensis nutriment is 

 supplied by the wall of the oviduct, and the yolk has in large 

 measure disappeared, at any rate its more solid portions ; but 

 the development still bears the impress of centrolecithal seg- 

 mentation, i.e., in the imperfect definition of the blastomeres. 

 It is obvious one might with equal justice expect to find 

 information as to the character of the ancestor of Metazoa in 

 the eggs of mammals. 



Let us now briefly rehearse the conclusions to which the 

 foregoing discussions seem to point. The earliest well-marked 

 larval stage which we have discovered is the blastula — a sphere 

 of uniformly ciliated cells. This " animal Volvox," as Huxley^ 

 calls it, may be regarded as a protozoon colony, not in the 

 sense of consisting of independent units any more than does 

 Volvox, but rather in the sense of being built up by the repeti- 

 tion of a unit as a result of what Lankester^ calls "eumero- 

 genesis,'' just as is the colony of a Hydromedusan. At first all 



1 'Anatomy of Invertebrates,' p. 678. 

 » Art. "Ilydrozoa," 'Encycl. Brit.' 



