MUSEUM OF COMPARATIVE ZOOLOGY. 219 



The mode of development which I have described in the foregoing 

 pages, taken in connection with frontal sections, which show that the 

 pronephric thickening tapers gradually backwards into indifferent soma- 

 topleui'e, seems to me to be very strong evidence concerning the precise 

 origin of the duct. I believe I am justified in concluding that the seg- 

 mental duct hetioeen somites V. and IX. arises in situ from a thickening 

 of the soniatoplevre serially equivalent to that from xvhich in the anterior 

 region the j^ronephros is developed. Indirect evidence which can be 

 brought to bear on this question will be reserved for the fuller con- 

 sideration which can be accorded it, in connection with the following 

 stage (page 222). 



Stage IV. 



Plate I. Figs. 8, 9. Plate III. Figs. 18-36. Plate IV. Figs. 89, 39. 

 Plate Y. Fig. 46. 



I have placed in this stage embryos of frogs taken from five different 

 killings. They all belong to the fourth day after fertilization, and aside 

 from individual variation show an evident advance in organization on the 

 preceding stage. In all a distinct differentiation of muscular tissue has 

 begun, the auditory vesicle is wholly cut off from the epidermis, and the 

 ventral sucking (or more properly sticking) disks are well developed. 

 In the following description, I shall find it convenient to distinguish a 

 younger and an older set of embryos. In the younger set the embryos 

 are from 3^ to 3^ mm. long; they have about 14 protovertebrse and the 

 fundaments of 3 pairs of gills. The embryos of the older set are from 

 3 J to 3| mm. long; they possess about 17 protovertebrie and the funda- 

 ments of 4 pairs of gills. 



All the embryos of this stage have the pronephric pouch in its typical 

 form. A side view of this organ with the neighboring portion of the 



section to the vertical axis of the protovertebra. Cross sections in this region fre- 

 quently encounter two contiguous protovertebrae. If the plane of the section 

 traverse the communicating canal of a protovertebra, it would also pass obliquely 

 through the dorsal portion of the next anterior protovertebra. The latter would 

 then appear in cross section as a distinct mass immediately lateral to the neural 

 tube and the chorda, and would resemble the condition which a protovertebra 

 presents when cut near its anterior or posterior wall. Immediately below this mass 

 there would be found on the same cross section the ventral portion of the more 

 posterior protovertebra, with the corresponding part of its cavity. The latter, how- 

 ever, being apparently a direct continuation of the body cavity, owing to the exist- 

 ence of the communicating canal, would appear to represent the dorsal part of the 

 body cavity, and the fundament of the duct would thus seem to be farther 

 removed from the dorsal angle of the body cavity than it really is. 



