CAPILLARY VESSELS. 155 



structure, they have arrived at a more complicated stage of 

 their formation, and I regard sucli fibres as distinct from their 

 cell-membrane. 



2. Very distinct cell-nuclei occur at different spots upon 

 the walls of the capillaries, both of the young and fully-deve- 

 loped tadpole. They appear to lie cither in the thickness of 

 the wall, or on the internal surface of the vessels, on which 

 they often form a projection. (See fig. 11.) They admit of a 

 double explanation. They are either the nuclei of the primary 

 cells of the capillaries, or nuclei of epithelial cells, which in- 

 vest the capillary vessels. It is true that epithelial cells occur 

 in vessels which have a great resemblance to capillary vessels, 

 if they are not actually such, as may be very distinctly seen 

 in the vessels of the membrana capsulo-papillaris in a foetal 

 pig of from four to six inches long, where some of them pro- 

 ject, in the form of half-spheres, into the cavity of the vessel; 

 but there were no epithelial cells perceptible surrounding the 

 nuclei in the capillaries of the tadpole's tail. On the contrary, 

 these nuclei frequently seemed to lie free upon the internal 

 wall of the vessel, and must have been much more abundant 

 had they been nuclei of epithelial cells. That these are the 

 nuclei of the primary cells of the capillaries is, therefore, most 

 probable, although this exclusive argument by no means decides 

 the question. 



3. In the tail of very young tadpoles, the capillary network 

 presents, besides the ordinary cylindrical canals which have an 

 equal diameter, and in which the blood flows in a regular cur- 

 rent, other vessels of an irregular form. Unfortunately I 

 neglected to make a drawing of them ; they accord, however, 

 in all essential particulars with the capillaries of the germinal 

 membrane of the hen's egg represented in pi. IV, fig. 12, 

 except that the meshes of the vascular network arc much 

 larger in the tail of the tadpole. They are not regularly 

 cylindrical. They are generally widest in situations where 

 branches are given off, sometimes wider even than the ordi- 

 nary capillary vessels. (See a, b in figure 12.) The branches 

 diminish very rapidly as they leave those broad parts, and 

 widen again as they approach another dilated portion They 

 present every degree of narrowing from vessels in which it 

 could scarcely be remarked, to those which are reduced so 



