CYTOPLASMIC CONSTITUENTS OF NERVE CELLS 415 



p. 182). He finds that the nucleoli in the peripheral cells of 

 the ventrolateral portion of the cord of chick embryos assume a 

 bipolar arrangement; that many minute granules, staining darkly 

 by Cajal's method, appear at either pole of the nucleus; and that 

 the first neurofibrils arise about the periphery of these granules. 

 I, also, have found that the first indications of neurofibrils appear 

 in a part of the cj^toplasm which stains especialh' darkly by 

 Cajal's method. 



The Nissl substance may apparently be excluded from the dis- 

 cussion, since, according to Marcora ('11, p. 946), it first appears 

 in the spinal ganglion cells of chick embryos of about six days' 

 incubation. Collin fixes the time even later for he states ('06, 

 p. 259) that it begins to manifest itself after about ten days' 

 incubation. Moreover, I have failed to demonstrate this material 

 in the nervous system of chick embryos of from 15 to 20 somites, 

 which is the stage during which neurofibrillation begins. 



The canalicular apparatus may, likewise, be eliminated. Mar- 

 cora ('11, fig. 22) has demonstrated it in the cells of the neural 

 tube of a duck embryo of three days' incubation. Repeated 

 efforts, on my part, have not given any indication of its presence 

 in the earliest stages of the formation of neurofibrils, although 

 indications of it possibly occur in figures 21, 25 and 26, where a 

 system of clear, unstained canals may be seen in the cytoplasm. 



It is my opinion that further investigation will show that both 

 the canaUcular apparatus and the chromidial substance appear 

 much earlier in development than has hitherto been supposed, 

 for they and the neurofibrils are, to some extent at least, indica- 

 tive of the same thing, namely, the functional maturity of the 

 cell. Furthermore, the tendency in the past has been for inves- 

 tigators to trace back these constituents of the cytoplasm into 

 earher and earlier stages of development with each advance in 

 technique, and our present technique is by no means perfect. 



It is impossible to rule out, in a similar fashion, the ground 

 substance from consideration because it is, like the mitochondria, 

 inseparably connected with all stages in the life of the cell. 

 The neurofibrils are therefore, in all probability, developed from 

 the ground substance, or from formed elements within it as yet un- 



