neal: nervous system in squalus acanthias. 157 



stage has been selected to begia with, because it gives the strongest 

 evidence that I have seen of a segmented condition in the neural plate. 

 I shall describe first two embryos which represent fairly well the con- 

 ditions I have found at this stage. These two embryos are represented 

 in Figures 1, 2, and 3, Plate 1. The neural or " medullary " plate is 

 seen to be a spatula-like expansion of the anterior end of the embryo, 

 raised somewhat above the blastodermic area. Figure 1 (Plate 1) repre- 

 sents an embryo viewed from the dorsal side ; the neural plate, it will be 

 observed, is not perfectly flat, for its edges bend slightly ventrad and a 

 shallow depression extends along its median portion. The chorda, lying 

 in the axis just beneath the neural plate, causes a slight elevation of the 

 floor of the groove in the median line. Anteriorly the chorda passes into 

 the common tissue which later becomes difi'erentiated into entoderm, 

 mesoderm, and chorda. The anterior more expanded portion of the 

 neural plate has been called the "cephalic plate." At the posterior 

 portion of this cephalic plate its lateral wing-like expansions undergo 

 their greatest bending ventrad. The posterior or trunk portion of the 

 neural plate extends back into the tail folds and along the embryonic 

 rim. I do not wish to seem to imply by this statement that the tail 

 folds and the embryonic rim become included in the neural tube, because, 

 although in general I believe in the concrescence theory, I do not find in 

 the continuity stated above any evidence of addition to the posterior 

 part of the neural plate by a concrescence of the tail folds and the embry- 

 onic rim. 



In the dorsal view of the embryo shown in Figure 1, Plate 1, little or 

 no evidence is afforded in either cephalic or trunk regions in support of 

 Locy's contention that the edges of the neural plate are segmented. We 

 see only that the edges of the plate are slightly and irregularly lobed, 

 and not in the true sense segmented. For the lobes 07i the opjyosite mar- 

 gins of the plate do not correspond in number or position, neither do they 

 show any definite relation to the mesodermal somites, as seen in the cleared 

 specimen. 



Figure 2, Plate 1, shows the same embryo viewed from the ventral 

 side, and gives the strongest evidence I have seen of Locy's interpreta- 

 tion of the condition of the neural plate. The " segments " appear much 

 more marked in embryos of this stage when viewed from the ventral 

 side, for reasons already stated by Locy, who has well insisted upon the 

 importance of ventral views. There are several reasons, however, for re- 

 garding the structures which appear aloTig the edges of the neural plate 

 as not true segments. These so called segments, even in the cephalic 



