FORMATION OF NOTOCHORD IN AMPHIBIA. 295 



on either side of the notochord are coming together under 

 the notochord. A few sections beyond (Fig. 11), the no- 

 tochord is almost entirely cut off from the archenteron, as only 

 one or two heavily pigmented cells lie between the two parts of 

 the dorsal endoderm. In the middle region of the embryo (Fig. 

 12), the endodermal cells have united under the notochord and 

 the notochord is a rounded chord of cells entirely separated from 

 the surrounding tissues. 



In Ra/ia palnstris, therefore, as well as in Bitfo lentiginosns, 

 the notochord is composed entirely of mesoderm in the anterior 

 part of the embryo, and of mesoderm and chorda-endoderm in 

 the posterior region. The early stages in the formation of the 

 notochord are very similar in the two species ; but in Rana there 

 is no upward growth of yolk cells as in Bufo to form the per- 

 manent dorsal wall of the archenteron. 



Most of the embryologists who have studied the early de- 

 velopment of the Urodela agree with Jordan (10) who de- 

 scribes the formation of the notochord in the common newt 

 as follows : " The cells of the median dorsal wall of the archen- 

 teron assume a somewhat columnar form and are gradually 

 pushed up and pinched off until they are completely separated 

 from the endoderm and come to lie above it in the mid-line.'' 

 This view is held by Hertwig (7), Scott and Osborn (20), Field 

 (5), Eycleshymer (4), Brachet (2), and Schwink (19). 



Lwoff (13) is, perhaps, the most prominent of those who op- 

 pose this view. In his study of Axolotl, Lwoff finds that the 

 mesoderm and the notochord are derived from cells invaginated 

 from the surface of the egg at the blastopore rim, and he states : 

 " Bei den Urodelen bildet sich die dorsale Wand des Darmes, 

 ebenso wie bei Petromyzon, verhaltnissmassig spat, namlich 

 nachdem die Chorda sich von den seitlichen Mesodermplatten 

 gesondert hat. Die Entodermzellen wachsen von rechts und 

 links einander entgegen, vereinigen sich unter der Chorda und 

 bilden aufsolche Weise die dorsale Wand des Darmes." This 

 description of the manner in which the permanent dorsal wall of 

 the archenteron is formed in the Axolotl agrees remarkably well 

 with the results of my investigations on Bnfo. Lwoff 's sum- 

 mary of the results of his study of the Anura based on an 



