86 K. MITSUKURI. 



is the part of the medullary substance {p, p) which is 

 continued posteriorly beyond the end of the suprarenal, as 

 paired cords on the ventral side of the aorta. In fact, the 

 part marked p on the left side of fig. 9 is continued from the 

 right suprarenal, which has ended several sections in front. 

 That this structure is really a part of the medullary sub- 

 stance there can be no question, as it is stained brown by 

 bichromate of potash. It extends backwards much more 

 than the length of the suprarenals themselves. For instance, 

 in an embryo of twenty-four days, the right suprarenal 

 appears in about forty sections, and the left in about thirty, 

 while these cords extend posteriorly for about fifty-five sec- 

 tions from the termination of the left suprarenal. The cords 

 unite with each other several times, soon, however, sepa- 

 rating again. Posteriorly, they gradually lose their brown 

 colour, and seem then made up of nerve fibres. Nerves may 

 also join them, where they still exhibit the typical brown 

 reaction. It is probable, from the comparison of the sec- 

 tions of embryos hardened in picric acid and potassium 

 bichromate, that so long as the histological structure shown 

 in /), fig. 9, is observable, the cords are stained brown. I 

 have carefully traced the cords forward both in the longitu- 

 dinal and transverse sections, and find that they become con- 

 tinuous with the nervous mass {n) shown in fig. 8. This is 

 obvious from the series of sections in fig. 10. A is the outer- 

 most, the sections gradually going towards the median axis 

 of the bod\ of the embryo. In A the cortex (c) is quite 

 large, while the part marked p (same as^, fig. 9) is con- 

 tinuous with the medullary substance (m). The nervous 

 mass [n, of fig. 8) is here also marked n. In B the cortex 

 (c) is disappearing ; in C it has completely disappeared, and 

 the part marked p has divided into two parts {p and a). 

 Lastly, in D, we see that the nervous mass {n) has united 

 with the small mass marked a in C. Thus there can be 

 no doubt that the structure p, figs. 9 and 10, is continuous 

 with the nervous mass n, fig. 8, and must be of a nervous 

 nature. And yet it has not a single typical ganglion cell, 

 and there is a considerable difference between its microscopic 

 appearance and that of the nervous mass (?z,figs.8,9, and 10). 

 It is possible that the cells that are found in^ are of the same 

 nature as the smaller cells in the nervous mass. The only 

 conclusion that I can arrive at is that this part of the peri- 

 pheral sympathetic system becomes early distinguished from 

 the other parts by an enormous development of connective- 

 tissue cells, and by a total absence of ganglion cells, and that 

 all this is preparatory to its transformation into the raedul- 



