699 



left in the mid - dorsal line an extrusion of yolk cells. This may 

 be found anywhere along the length of the medullary plate of the 

 embryo, from a point just behind the anterior commissure of the brain 

 to the anal opening. Such an embryo will show a condition of spina 

 bifida. 



I have carefully studied in living eggs, in hardened preparations 

 and in sections, all intermediate stages in the process of closure. In 

 many cases the dorsal lips of the blastopore fuse around the point of injury 

 and progress backwards at a rate slower than the normal overgrowth 

 of the dorsal lip. Meanwhile the lateral and posterior edges of the 

 blastopore have begun to fuse along the median line, from behind for- 

 wards. According to the relative rate of growth of these two ends, 

 that are approaching each other, the position of the yolk plug will be 

 determined. In a few cases the dorsal lip remained stationary and 

 then the yolk plug appeared immediately behind the anterior com- 

 missure. All intermediate stages, in the exposure of the yolk, between 

 this point and the anal opening are found. In some cases where the 

 injury is small, perfect embryos may result. 



Sections through these embryos show that wherever the yolk plug 

 is left protruding the medullary plate and notochord are formed in sepa- 

 rate halves on each side of the yolk plug, and the doubling may 

 extend for some distance both in front of and behind the yolk plug. 

 The reason for the failure of the lips of the blastopore to entirely 

 grow over the yolk mass seems to be due to the delay in the pro- 

 cess of closure, so that the medullary plates begin to form before the 

 yolk is enclosed. After this process has set in, the subsequent closure 

 is prevented or delayed. 



These experiments do not exclude the possibility that a bifurcation 

 of the material of the dorsal lip may have taken place in these em- 

 bryos, owing to injury to the cells situated in the middle line. I tried 

 to test this possibility by the following experiment. Instead of a single 

 injury to the dorsal lip, a series of points were stuck in and on each 

 side of the dorsal lip of the blastopore, so that a line of extruded 

 cells stretched across the blastopore. The result was the same as in 

 the preceding experiment, but more striking, owing to the less fre- 

 quent fusion of the lateral lips of the blastopore immediately behind 

 the region of injury. The possibility of a splitting of the material of 

 the dorsal lip, and a lateral backgrowth of the same, is made highly 

 improbable by the last experiment. 



If then, as the experiments seem to show, the material of the 

 dorsal lip does not grow backward, we ought to find evidence of this 



46* 



