No. 3.] THE EMBRYOLOGY OF THE EARTHWORM. 423 
may be traced forwards into the head-cavity, and backward to a 
point near the dorsal median line; here they pass into the ecto- 
blast, but I have not found any communication either with the 
exterior or with the head-cavity. The important problem of the 
relation of these organs to the permanent nephridia I have en- 
tirely failed to solve ; this question can be studied to far better 
advantage in forms like Crtodrilus, where the head-kidney is 
much larger. 
We may now turn to the permanent nephridia. I have stated 
(p. 390) that these organs arise in connection with a contin- 
uous cell-cord of ectoblastic origin that forms part of the middle 
stratum of the germ-bands and lies alongside of the neural cord 
(Figs. 49, 59, 63, 72, etc.). Each nephric cord terminates be- 
hind in a pair of teloblasts derived from the ectoblast. The 
entire nephric cord is formed by the continued divisions of these 
“nephroblasts,” which agree precisely with the neuroblasts in 
structure, action, and mode of origin. In early stages each 
nephric cord consists behind of a double row of cells, each row 
terminating in one of the nephroblasts. Passing forwards the 
rows are no longer separated by any definable limit, and the 
nephric cord consists of an irregular series of cells which 
passes upwards beside the neural row and is lost at the sides of 
the stomodzum ; beyond this point I have never been able to 
trace it, either in sections or in surface-views. The general re- 
lations of the nephroblasts and the nephric cords to the remain- 
ing elements of the germ-bands are at first precisely similar to 
those of the neuroblasts and the neural cords, as may be seen in 
Figs. 53, 54, 71, 72, 90, 91. At first they lie at the surface of 
the body and form part of the general ectoblast; afterwards 
they sink beneath it so as to lie between the outer ectoblast and 
the mesoblast. The nephric cells closely resemble those of the 
general ectoblast and of the neural rows, but as a rule stain 
more deeply than the neural cells and less deeply than those of 
the ectoblast, so that all these structures appear with remarkable 
distinctness in cross-sections, double-stained with borax-carmine 
and Kleinenberg’s hzematoxylin. 
In surface views of carefully stained germ-bands spread out 
in glycerine (Fig. 63) the nephric rudiments are clearly seen to 
lie along the line of the nephric cords. In both cross and longi- 
tudinal sections they may be traced directly into connection 
