342 THE NERVOUS CYTOLOGY OF PERIPLANETA ORIENTALIS. 



the occasional substitution of physiological salt solution for water, which apparently 

 in no way affected the result, and the shortening of the manipulation to five minutes 

 and under, after practice.) The ganglia were placed at once in vom Rath's fixing 

 fluid * for one-half hour, thoroughly washed in water, run through graded alcohols 

 to absolute, then through creosote and xylol to paraffin. They were embedded and 

 cut into sections from 1.5 to 3.3 micra in thickness. They were attached to the slide 

 with albumen fixative, mounted in xylol damar, no stain having been used. Other 

 ganglia were prepared in the same way, with the exception that they were only left 

 in the fixing fluid ten minutes. 



2. Observations — The cells lay naked in their connective-tissue spaces, whose 

 walls were regularly somewhat wavy and shrunken in appearance. The nuclei were 

 surrounded by a distinct membrane, which was usually indented at one or two points 

 and occasionally badly shrunken. The nuclei themselves presented a very fine and 

 uniformly granular appearance and contained one or more dense nucleoli. The cell 

 bodies presented, throughout their central and major portions, a blurred granulation, 

 varying in density and in the size and number of the granules. Their periphery, 

 however, exhibited regularly at some part, and not infrequently throughout the 

 entire circumference, a lighter zone, composed of fibrillae (PL XXV, Fig. 1). These 

 were almost always observed at the point of origin of the nerve-fibre, lying parallel 

 with the long axis of the cell. They sometimes extended from this region partly or 

 all the way along one border of the cell, giving the appearance of an unsymmetrical 

 entrance of the nerve-fibre and more or less completely enveloping one side of the 

 granular portion of the cytoplasm (Figs. 2, 3). Often they formed a broken net- 

 work, with its main strands centripetally disposed, and occasionally they lay in irreg- 

 ular whorls (Figs. 4, 5). The grouping of individual fibrillae into bundles caused 

 the strands presenting the above appearances to vary markedly in thickness. The 

 fibrillae regularly invaded the granular area to some extent, thus making a gradual 

 transition. In a few of the cells, subjected to the action of the reagent for only ten 

 minutes, the peripheral network was more even and could be seen to acquire more 

 and more granules along its strands until it was lost in the granular area (Fig. 6). 

 The cells of various sizes presented essentially the same appearances after this, as after 

 all other plans of technique. 



In cross-section the nerve-fibres (Figs. 7, 9) exhibited in most specimens a 

 shrunken sheath, and within it an irregular and unsymmetrically disposed bunch 

 of granules, connected to the sheath by two or three fine strands. In more favorable 



* Vom Rath's Fluid : Sat. aq. sol. picric acid, 80 cc; 10% aq. sol. osmic acid, 10 cc; 10% aq. sol. platinic 

 chloride, 4 cc; glacial acetic acid, 0.8 cc. 



