KUTCHIN. — PERIPHERAL NERVOUS SYSTEM OF AMPHIOXUS. 619 



The ventral nerves were studied in both species of Branchiostoma, 

 and no differences were found which could not be readily explained by 

 the uncertainty of the methods employed. The nerve fibers of the 

 ventral nerves show such, uniform smoothness in good preparations 

 that it seems doubtful if any actual "varicosities" exist. The num- 

 erous smooth fibers illustrated in Figures 28, 29 (PI. 5), and 46 (PI. 8), 

 and the uniform smoothness of fibers impregnated with gold chloride, 

 certainly give grounds for such a conclusion. The small swellings 

 sometimes present in my Golgi preparations are of irregular occurrence, 

 and are usually found in whole specimens, or those impregnated 

 according to the slow methods. It is true that occasionally structures 

 resembling bipolar ganglion cells (PI. 8, Fig. 50) are to be found toward 

 the proximal ends of fibers in the ventral nerves, but one hesitates 

 to regard these structures as cells when their presence here would 

 indicate a marked variation from the typical condition. In Figure 

 50 an apparent nucleus is present in the cell-like body attached to 

 the nerve fiber; but every investigator who has used the methods of 

 Golgi knows what bewildering artifacts can be produced. 



According to my preparations the motor fibers of Amphioxus leave 

 the inner border of the spinal cord sheath (Fig. 25, PI. 5) as rather 

 slender, smooth fibers, apparently differing somewhat in size. These 

 fibers are often curved or wrinkled in the sheath, and frequently bend 

 sharply as they leave the covering of the neural tube. They are 

 continued to the muscle border (Figs. 25, 26) as comparatively smooth 

 threads, changing little in caliber. The spreading out of the ventral 

 nerves has been described so often that it need not be taken up here. 

 Upon reaching the side muscles certain motor fibers increase in size, 

 while others do not change, depending apparently on their distance 

 from the muscle fibers which they innervate. As a fiber approaches 

 its distal end, it gradually thickens, and finall\- ends with the peculiar 

 plate- or cone-like structure, first noted by Heymans et van der 

 Stricht ('98). In contrast to the observations of Dogiel (:02), 

 little branching is e\-ident in the course of these fibers. Figures 28, 

 29 (PI. 5) and 46 (PI. 8), which show a large number of motor fibers, 

 do not reveal branching. In Figures 28 (PI. 5) and 4.3 (PI. 8) single 

 fibers may be traced from near the internal border of the muscles to a 

 short distance from its exterior border and in one case (Fig. 43) to the 

 termination of the fiber. These long, single fibers are significantly 

 frequent in various parts of the myotome. Branching certainly 

 occurs (PI. 8, Figs. 48, 49), but my preparations do not reveal the 

 extensive division in the region of the muscles described by Dogiel. 



