MUSEUM OF COMPARATIVE ZOOLOGY. 251 



tapers gradually into undifferentiated somatopleure. The backward pro- 

 longation of the thickening is the first fundament of the segmental duct, 

 and may be traced at least as far back as somite VI. Both portions of 

 the thickening appear to arise in the same way ; namely, by cell prolifer- 

 ation in the somatopleure. 



It is a matter of some difficulty to ascertain when the first trace of a 

 lumen appears. Before the two walls actually separate, the nuclei fre- 

 quently show an arrangement which is suggestive of an evagination ; 

 but one cannot always trust such appearances. Later, a line of pigment 

 can be traced from the body cavity for some distance into the interior of 

 the thickening, and finally the two walls separate, leaving a clearly defined 

 lumen. In all cases, the two regions of continuity with the ccelom are 

 opposite the middle of protovertebra3 III. and IV. respectively ; and 

 there is no indication whatever of a continuous fold. 



Although the pronephric mass thus shows evident signs of segmen- 

 tation, yet, as is to be seen by a comparison of segmental and interseg- 

 mental regions (Plate VII. Figs. 55 and 56), the proliferation is not 

 interrupted in the latter regions. In frontal sections through prone- 

 phridia in which a definite lumen has begun to appear (compare Plate 

 VII. Fig. 55), there can be seen two narrow canals leading from the 

 cavities of protovertebrae III. and IV. and extending outward as ccelomic 

 diverticula into the pronephric mass. From this condition the hasty 

 conclusion misdit be drawn that the narrow canals are in fact outgrowths 

 from the protovertebral cavities. This however, in my opinion, is not the 

 case. If the relations of the mesoderm in such a transverse section as is 

 shown in Figure 55 be regarded, it will be seen that a frontal section 

 through the pronephric region (in the figure cited, a horizontal section a 

 little below the level of the letters coelJ) would cut through the proto- 

 vertebral cavity near its floor, and at the same time pass through the 

 lumen of the pronephric thickening. Since, moreover, these two spaces 

 are continuous by means of the communicating canal, it might at first 

 appear that the latter belonged to the pronephric tubule. The fate of 

 that portion of the tube, however, shows this interpretation to be incor- 

 rect, and that it was only by means of the communicating canal that the 

 lumen of the pronephros communicated with the protovertebral cavity ; 

 for when the separation of the protovertebra 1 from the lateral plate takes 

 place, the communicating canal, which is assumed to be the stalk of the 

 pronephric diverticulum of the protovertebra, becomes closed, and the 

 pronephros is thereby left in communication with the body cavity alone 

 (compare Mollier, '90, Taf. XII. Figs. 10 c, 10 (/., tr x and tr 2 ). 



