ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
617 
paper discusses tlieir characters from the comparative point of view. 
As to method, he emphasises the value of corrosive sublimate as a fixa- 
tive, and especially the necessity for mounting sections in glycerin ; 
such preparations show, he says, details invisible in balsam, and also 
allow of fresh staining, which may be necessary. 
Some only of the points of structure dwelt on can be mentioned here. 
Ganglion-cells in general consist of three parts, a homogeneous hyalo- 
plasm, a supporting network of spongioplasm, and a deeply staining 
substance having a varying relation to the spongioplasm proper. The 
neuroglia places the ganglion-cells in communication with one another. 
The nuclei of the cells contain two kinds of granules, some of nuclein, 
and others of paranuclein ; the latter are the nucleoli. Besides the 
chief nucleolus, there are also smaller slightly different structures, the 
accessory nucleoli. These may pass out into the cytoplasm and give 
rise to the structures sometimes described as centrosomes ; they probably 
form the chromatin substance already described in the cell protoplasm. 
The author believes that the specific function of ganglion-cells is depen- 
dent on this continued passage of accessory nucleoli from nucleus to 
cytoplasm. He 'does not believe that centrosomes occur in ganglion- 
cells. The structures described by Lenhossek as centrosomes in the 
cells of the spinal cord in the frog, are not true centrosomes. Their 
meaning is doubtful, but they may be remnants of degenerated ganglion- 
cells. 
Histology of Glands.* — Herr Erik Muller has found that the remark- 
able secreting capillaries of the gastric glands can be demonstrated by 
other means as well as by Golgi’s famous method. Fragments of the 
mucous membrane of the stomach of a dog were laid for 24 hours in 
Kopsch’s mixture (40 parts 3*5 per cent, potassium bichromate and 10 
parts of commercial formalin), and then for one or several days in the 
bichromate alone. They were then washed, and hardened in alcohol of 
increasing strengths. The sections were stained with Heidenhain’s iron- 
hsematoxylin. The results were admirable, and not only served for the 
discovery of several new points, but were valuable as an independent 
confirmation of results obtained by Golgi’s method. 
The results as to the fundus glands in the dog’s stomach are to show 
that the paths of the secretions are first by the central lumen and cross 
channels, which are surrounded by a distinct ectoplasmatic membrane, 
and secondly by the intracellular secreting capillaries. In both central 
and parietal cells the- secretion arises from granules which, before they 
are converted into fluid, pass through two stages, one in which they are 
stainable, and one in which they are not. The course of the secretion is 
different in the two sets of cells : — (u) in the first case this is poured 
direct into the gland-lumen ; (b) in the parietal cells it is drained off 
by their intracellular canals. 
In a further investigation of the mucous glands of the cat, the author, 
by means of the injections of pilocarpin, obtained cells free from secre- 
tion, and found that not only, as has been previously stated, do empty 
mucous cells and empty albuminous gland-cells closely resemble one 
another, but, further, neither are distinguishable from the cells of 
* Zeitschr. f. wiss. Zoo!., lxiv. (1898) pp. 624-47 (2 pis.). 
