70 PROCEEDINGS OF THE AMERICAN ACADEMY 



Three or four hours after the appearance of the ring, when the 

 blastoderm covers about one third of the yolk, and the anterior end 

 of the embryo has nearly reached the ectodermic pole, we may obtain 

 a good view of the formation of the chorda. For this purpose the 

 ovum is so placed that we can get an optical transverse section of the 

 embryonic portion near the middle. At this point the entoderm is 

 only one cell deep, and the formation of the chorda can be seen with 

 the greatest ease. A median band of the entoderm becomes plainly 

 cut off from the lateral portions by vertical and nearly parallel planes 

 of division. This median band, which is perfectly flat and of even 

 thickness with the lateral portions, is the Anlage of the chorda. 

 Leaving the history of the chorda at this point, we have to inquire 

 how the alimentary tract is formed. As the lateral portions of the euto- 

 dermic layer thicken up, they divide into two strata : the lower stratum 

 is one cell deep, and represents the so-called " secondary entoderm " ; 

 the upper, which is several cells thick, represents the mesodermic plates. 

 The lower strata abut at first against the inferior angles of the chorda, 

 but close up under the chonla somewhat later, thus forming a single 

 layer of flattened cells. About the time the blastopore closes, a median 

 strip of this layer, corresponding in width to the chorda, begins to 

 thicken, soon becoming two or three cells deep. It is this thickened 

 strip that gives rise to the alimentary tube. Precisely how the solid 

 band is converted into a tube, we are not at present able to state. 



Comparing the foregoing account with that given by 0. Ilertwig ''''^ 

 in his last paper on the development of the mesoderm of the amphibia, 

 one important difference is seen. Ilertwig comes to tlie conilusiou 

 that his " Chorduentoblast" forms not only the chorda, but also the 

 roof, of the alimentary canal, the lateral and ventral portions of the 

 canal being derived from the " Darmentoblast." We think that what 

 we have described as tlie entodermic ring corresponds to the chorda- 

 entuhlust of Rana ; and it seems plausible that the periblast should 

 correspond to the '• Darmentoblast " (*' yolk hypoblast of Scott and 

 Osborn "). On this view w^e should expect the periblast to take some 

 share in forming the alimentary canal, which cannot be admitted if our 

 observations are correct. Ilertwig's observations do not bring the 

 development of Triton and Rana into agreement on this point; and 

 until this is done, we cannot expect to see the differences between the 

 teleostean and amphibian development fully reconciled. But as there 



60 O. Hertwig. Jena. Zeitschr., XVI., pp. 247-:V2:], 1883. 

 6' Scott and Osboni. Quart. Journ. Micr. Sci., XIX , 1879. 



