562 Stewart Paton 



thods most favorable for the determination of cell boimdaries, bc 

 carefuUy studied, it is apparent that in certain areas, particulary at 

 the poiüt where the ventral roots develope, there are undifferen- 

 tiated portions of protoplasm. The same is true of the marginal 

 tract (Randschleier). The affirmation frequently made that these 

 areas are completely tilled up by the processes of neuroblasts, with 

 the addition of the intermediate supporting tissue derived from 

 spongioblasts, lacks confirmation. There is no evidence that at this 

 period any of the products of these latter Clements have yet reached 

 even the inner border of the group of neuroblasts lying in the 

 antero-lateral region of the cord. 



The description of the intimate connections so far reported to 

 exist between the priniary ventral root fibrils and cells cannot be 

 Said to be convincing, and they are demonstrable only in those instan- 

 ces where the methods of fixation and staining fail to differentiate 

 sharply between the protoplasm extruded from a cell to form a 

 process and the neurofibrils. The sections where these apparent 

 connections are most clearly shown are those in which silver is 

 precipitated in comparatively large quantities and where only a 

 slight differentiation of the tissues, even nnder the most favorable 

 conditions, is possible. Such a raethod for example is the one recently 

 employed by Ramon y Cajal. A careful comparative study of the 

 results obtained by the use of this method and the one described 

 in Section 3 shows that the differentiation of the tissue obtained by 

 the latter is far greater than by the former. The excessive shriukage 

 that occurs when the former is used is an additional drawback in 

 studyiog the finer structure of the nervous system. 



The preci pitation of silver taking place at the ends of the 

 neuraxone in prepavations stained by Ramon y Cajal, and described 

 by this investigator as bulbous ends (Wachsthumskeule) are clearly 

 artefacts, and show the uureliability of bis method when used as 

 a means for studying the histology of the nervous system. The 

 svvollen ends of the processes represented in tiie illustrations accom- 

 panyinghis latest contribution (Anat. Anz. 30. Bd. ])ag. 113 — 144 1907 

 Fig. 11 [c] and Fig. 12) are obviously artiticial products. Within the 

 processes the individuai fibrils are glued together by the excessive 

 deposit of silver, while the distal ends are surrounded or encapsn- 

 lated by au iniperfectly stained coating of protoplasm in which they 

 are embedded. The apparent independence of the neurolibrils is well 

 shown in my Fig. 7, Piate 23. 



