220 BULLETIN OF THE 



segmental duct is shown in Figure 39 (Plate IV.). In this drawing, 

 the outlines were obtained by reconstruction from a series of cross 

 sections. The prouephric pouch is suspended from the dorsal angle 

 of the body cavity by the nephrostomal funnels. Elsewhere it is wholly 

 cut off from the mesoderm, and merely rests conformably on the outer 

 surface of the somatopleure. The precise relations of the parts can be 

 understood by referring to the series of cross sections shown in Figures 

 18 to 22 (Plate III.). Figure 18 represents a section through the left 

 pronephros in the region of the first nephrostome. The location of the 

 plane of thi's section in the reconstruction is indicated by the dotted 

 line IS, in Figure 39. The lateral plates are here wholly cut off from 

 the protovertebrse, splanchnopleure and somatopleure being continuous 

 with each other at the dorsal angle of the body cavity. Figui'e 19 

 shows the structure of the organ between the first and second nephro- 

 stomes. In this and the following sections it was found advisable to 

 depict the pronephric structures of the right side in oi'der to exhibit 

 in each case the section which most clearly showed the structural con- 

 ditions. The next drawing (Fig. 20) in the series represents a section 

 through the second nephrostome. In the preceding section, — not fig- 

 ured, — the three portions into which the lumen is here divided are 

 continuous. The constriction between the middle and the ventral lumen 

 is artificial ; for the cells occasioning this local closure do not belong to 

 the proper wall of the pouch, but form a group within the cavity. In 

 several instances I have observed such groups of cells lying entirely free 

 in the lumen of the pouch (Plate V. Fig. 45). In the present case, 

 howevei*, the mass is very intimately connected with the adjoining walls. 

 This condition is preserved through a distance corresponding to the 

 thickness of two or three sections, but the mass terminates by becoming 

 free from both walls, so that in cross section it has the appearance of an 

 " island " of tissue occupying the h;men of the pocket. The occurrence 

 of these islands within the cavity of the pouch is of significance in 

 determining precisely how the organ is developed. It is difficult to com- 

 prehend how they could be formed, provided the canals were produced 

 by a fold of the somatopleure. On the other hand, they are perfectly 

 intelligible on the assumption that the canals arise by the rearrange- 

 ment of a solid mass of cells into a peripheral layer with a central 

 lumen. According to the latter view, the islands would represent 

 residual portions of the pronephric thickening which had not been trans- 

 formed into peripheral wall. 



Picturning now to the section last under consideration (Fig. 20), the 



A 



