222 BULLETIN OF THE 



centre of the mass (c. and c/.), which, as I have said, doubtless corre- 

 spond to the central protoplasmic area (cd.) seen in the preceding sec- 

 tion. The most prominent cells of that section are here represented by 

 two faint circles of protoplasm (b. and e.). In the next following section, 

 not figured, the duct terminates as a single non-nucleated mass, probably 

 corresponding to the dorsal cell in Figure 25. This remnant lies in a 

 distinct depression of the somatopleure (Fig. 26,/.). This depression 

 continues backwards through the space of three sections. Instead, then, 

 of terminating in a thickening of the somatopleure, the end of the duct 

 lies in a groove of unmodified somatopleure. There is no tissue directly 

 behind the duct for its further growth, and the inference is natural that 

 the somatopleure is mechanically depressed before the growing tip of the 

 duct. In fact, I believe this to be actually the case, and that in this 

 region the duct does grow by a simple cell proliferation within its own 

 mass. 



The key to the situation is to be found in the location of the pos- 

 terior end of the duct in this specimen. An enumeration of the somites 

 shows that the sections figured lie at the hinder end of somite XL To 

 show the bearing of this fact, I shall anticipate some of the results 

 of a study of Stage V. In a series of frontal sections of the latter 

 stage, I have succeeded in locating with reference to the successive 

 somites the position at which the duct opens into the cloaca. The open- 

 ings are in the same vertical plane with the middle of somite XII. The 

 posterior end of the duct, then, in the specimen which I have just de- 

 scribed, is within the distance of half a somite from its final termination. 

 In order to empty into the cloaca, the duct has to grow inward from its 

 position at the lateral margins of the protovertebrce to a position much 

 nearer the median plane. It is difficult to comprehend how the duct 

 could make this extension, except by proliferation of its own cells. It is 

 just in this region that I find evidences of such a mode of growth. If the 

 inference I have drawn from the facts adduced be correct, it seems to 

 me to add strength to the conclusion I have reached in regard to the 

 general mode of formation of the duct, inasmuch as it has been shown to 

 be possible to detect free growth where it exists. 



That the duct arises in the way I have described, and is not developed 

 from the ectoderm, is shown, moreover, by certain indirect evidence which 

 may be properly discussed at this point. As I have already stated, the 

 duct is developed in such intimate connection with the somatopleure 

 that I have been led to believe that it arises throughout its entire length 

 from a proliferation in situ of that layer. In almost all of my prepa- 



