2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 56 



topore the primitive streak, ps, is seen as a collection of scattered 

 cells between the ectoderm and the entoderm, apparently formed by 

 proliferation from the ventral side of the ectoderm. 



A slightly later stage is shown in figure 2, a dorsal view of an 

 embryo with five pairs of mesoblastic somites. A sagittal section of 

 this stage is shown in figure 2A. The foregut is here more inclosed, 

 and the notochord, nt, having separated from the entoderm, en, 

 is seen as a distinct layer of cells extending from the foregut to the 

 blastopore. 



A transverse section through the headfold of this stage is shown 

 in figure 2b. The foregut is seen as a w4de cavity, cnt, depressed 

 dorsally, apparently, by the formation of the medullary groove and 

 the notochord ; it is wider laterally than in a dorso-ventral direction, 

 and its walls are made up of about three layers of closely arranged, 

 irregular cells ; the wall is somewhat thinner on the dorsal side, 

 just below the notochord. 



Figure 3 is a dorsal view of the next stage to be described ; al)out 

 fifteen pairs of somites are present. 



Figure 3A is a transverse section through this embryo near the 

 anterior end of the enteron, cnt, which cavity, cephalad to this region, 

 is bluntly pointed. As seen in the figure the enteron is here wide 

 from side to side, and is depressed dorso-ventrally except for a wide 

 groove in the ventral wall. This groove is lined with rather more 

 closely arranged cells, and marks the region where the mouth will 

 break through at a somewhat later stage. A short distance caudad 

 to this region the groove disappears and the pharynx is reduced to a 

 shallow slit extending almost to the superficial ectoderm on either 

 side ; then the slit-like pharynx becomes suddenly reduced in a lateral 

 and increased in a dorso-ventral direction, to assume the outline 

 shown in figures 3F. and 3c. At a point about one-third of the length 

 of the embryo from the tip of the head, the enteron opens to the yolk- 

 sac, so that what now may be called the foregut has this considerable 

 extent. There is, however, not the slightest indication of a tail-fold, 

 so that there is no inclosed hindgut at all. As is shown in figure 30, 

 the neurenteric canal, nc, still opens ventrally, though the medullaiw 

 canal, mc, has now no dorsal opening to the exterior. The medullary 

 canal continues for a short distance (about fifteen sections of five 

 microns thickness) posterior to the opening of the neurenteric canal. 



Figure 4 is a surface view of the next stage to be described. There 

 are here about twenty pairs of somites, though the exact number 

 cannot be determined. Although not visible externally in the surface 

 view shown, the gill clefts are beginning to form, and the first one 

 opens to the exterior as will be seen in sections of another embryo 



