A HUMAN EMBRYO OF TWENTY-FOUR PAIRS OF SOMITES. 145 



which portions are actually expanded and which constricted. A segment which 

 appears expanded in side view may appear constricted from the front or back. 

 Relative measurements of cross-sectional (oblique) areas of the chorda dorsalis 

 between the second and thirteenth segments seem to indicate that the expanded 

 portions lie within the body segments, but the evidence for this is not altogether 

 convincing. 



That segmental flexures of the chorda dorsalis exist is indicated by the fact 

 that for long distances, including extents of a number of segments, the chorda lies 

 approximately equidistant from the medullary tube. The ventral surface of the 

 medullary tube, as shown in plate 2, figure 1, presents dorsal and ventral curva- 

 tures; consequently the chorda must also follow these curvatures. This would 

 make the ventral curvatures of the chorda segmental. Minot 37 has described, 

 under the term "segmental flexures of the notochord," a series of dorsal and 

 ventral curves which he found to be constant in a number of mammals, man 

 included; these, he states, are so placed that their dorsal curvatures are segment- 

 ally arranged, but he states further that in young pig embryos the ventral curves 

 are segmental and that a shifting of the flexures takes place in embryos of about 

 12 mm. The flexures which are apparently present in my embryo, therefore, 

 accord with those which Minot describes for pig embryos younger than 12 mm. 



The chorda is fused with the entoderm of the digestive tube in one region 

 only, namely, in the posterior part of the pharynx at about the level of the third 

 to fourth body segments. Here it is attached in two places, each extending 

 through but a few sections and the two fusions separated from one another by but 

 three sections. The more anterior fusion is shown in text-figure 7. It is of 

 interest to note that Gage 12 and Thompson 44 both found a fusion of the chorda 

 to the entoderm in the same location and that Watt 48 shows the chorda? in the 

 twin embryos he describes to be least developed in this region. Watt believes, 

 therefore, that that section of the chorda opposite the posterior part of the pharynx 

 is the last in point of time to develop, and that embryos showing a fusion of 

 chorda and entoderm in this region are evidence in favor of this view. 



Throughout its whole course the chorda dorsalis is invested on either side 

 with mesenchyma. Dorsally and ventrally, however, this investment is not 

 complete. Ventrally it is incomplete not only at the points of fusion with the 

 entoderm, but both anteriorly and posteriorly to them. Anteriorly it lies almost 

 in contact with the roof of the greater part of the pharynx (text-fig. 6). Poster- 

 iorly it soon becomes separated from the pharyngeal wall by intervening mesen- 

 chyma. From the seventh body segment caudally the chorda lies almost in 

 contact dorsally with the medullary tube, there being from this point caudally no 

 mesenchyma on the dorsum of the chorda. Elsewhere the chorda is completely 

 surrounded by mesenchyma. The younger chorda? described by Watt showed 

 mesenchyma ventrally in one small region only, namely, at the level of the eighth 

 segment, while dorsally mesenchyma passed between the chorda and medullary 

 tube in two places at the anterior end and again opposite the first body segment. 

 The rapidity with which the notochodral processes of the sclerotomes separate 



