648 'Journal of Comparative Neurology aud Psychology. 



specimen, it must be destroyed in further proceeding to make the 

 one first attempted. Consequently the best plan in such a case is to 

 begin anew on another brain if there is plenty of material available. 

 The fibers of the brain are so easily removed that one has to be con- 

 tinually on guard against proceeding too quickly and attempting to 

 show too much at once. It is better to show a little in each dissec- 

 tion and to do a greater number to cover the ground. For this 

 reason I recommend beginning at the cortex, studying the superficial 

 fibers first, and making a series of dissections, each deeper than the 

 preceding one, until the whole brain has been gone through. It is 

 well to dwell upon some of the difficulties to be met with, and to say 

 something of the methods which I use to overcome or diminish them. 

 By careful observation during the dissection of over 200 brains, 

 human and others, I was able to gather certain facts, and abstract 

 with tolerable accuracy certain laws regarding the structure of the 

 white matter of the brain, which have been of great use to me in my 

 subsequent dissections. The chief difficulties are caused by inter- 

 crossings with other tracts. These intercrossings take place at various 

 angles, and on this depends the degree of difficulty. The nearer to a 

 right-angled crossing it is, the more difficult it will be to dissect. 

 If the fibers are almost parallel — or merely interlocking at an acute 

 angle — the tract can easily be followed, but if the angle at which 

 they cross each other is much greater than this, it will be impos- 

 sible to dissect them unless one tract is markedly larger than the 

 other, or its fibers are in isolated bundles. When one tract is much 

 smaller than the other, it becomes lost at the intercrossing and only 

 the larger one can be followed. The intercrossing and intermixing 

 of fibers from different systems is, for obvious reasons, greater near 

 the cortex than is to be found deeper, consequently the superficial 

 dissections and the terminals in the cortex of deep dissections are 

 less certainly made out than deeper tracts themselves. I have also 

 observed that fibers arising in adjacent convolutions or adjacent 

 nuclei and going to a distance to be distributed to adjacent areas soon 

 gather together and continue in a compact bundle, which is in some 

 cases thoroughly isolated from the surrounding structures until near 

 the points of distribution, when they spread out and proceed to their 



