Shinkishi Hatai 



axis cylinder at its end enlarges greatly and forms the so-called " axis cylinder plate." 

 An enlargement of the axone terminal may be seen in Figs. 5 and 6. Especially in 

 Fig. 5, where the nerve-fibers enter into the granular layer of the cerebellar cortex, 

 there is to be seen an enormous enlargement of the axones to several times their 

 original diameters. A detailed description of the structure of the axone terminals and 

 their relation to the surrounding neurones will'be given later. 



2. Structure of the dendrites. — The internal structure of the dendrites shows a 

 close resemblance to that of the cell-body. Besides the ground substance which stains 

 faint red, as in the case of the cell-body proper, it contains Nissl granules. Unlike 

 the axone the dendrite contains but a small amount of the ground substance, and, fur- 

 ther, the size of the individual neurosomes is approximately the same as that of the 

 cyto-microsomes, where they stain more faintly than the neurosomes in the axone. In 

 other words, the neurosomes in the dendrites do not show much differentiation from 

 the cyto-microsomes. The reticulum, however, presents a marked alteration, exhibiting 

 in some cases (Plate XIII, Fig. 4) a fibrillar arrangement. A most interesting feature 

 of the dendrite is the nodules or gemmules which develop along its periphery. By 

 the Golgi technique they stand out like pin-head prolongations or knobs. The 

 presence of these gemmules on the dendrites has been denied by Hill (1896), while 

 the other investigators regard them as very important and constant structures in 

 certain forms of nerve-cells (Van Gehuchten, 1897, Cajal, and others). Still another 

 interpretation has been made by Demoor (1896, 1898), who considers the moniliform 

 appearance of the dendrite as a condition in which the gemmules are partially retracted 

 and regards them as important for the normal activity of nerve-cells. I agree 

 with the view which regards these structures as a constant character of certain forms 

 of the nerve-cells. This knob-like structure can be seen not only in specimens pre- 

 pared by the Golgi technique, but also in those prepared by my own method. In this 

 case we can see clearly the internal structure of the gemmules and their relations to the 

 main body of the dendrite. The gemmules are nothing more than a local extension of 

 the ground substance of the dendrites, and a more or less modified reticulum can be 

 seen within them in many cases (Plate XIII, Fig. 4). It is difl&cult, however, in some 

 instances to distinguish the gemmules from the surrounding structures, when a large 

 number of the neurosomic chains forming the axone terminals surround the dendrite 

 very densely. Careful observation shows that the neurosomes in the gemmules stain 

 less deeply than those forming the terminals. The accumulation of neurosomes to 

 form gemmules is shown in Fig. 4, which has been drawn from the cells in the cerebral 

 cortex of the adult white rat. 



IV. TERMINATION OF THE AXONE ON THE DENDRITES AND CELL-BODIES 



1. Termination of the axone on the cell-body. — The actual termination of the 

 axone on the cell-body as well as a diffused network of the nerve-fiber terminals 

 surrounding the cell-body and forming the so-called "pericellular network" has been 



183 



