OBSERVATIONS ON THE STRUCTURE OF THE SPLEEN, 369 
become gradually separated so as to be lymphoid corpuscles, 
and are finally carried away by the blood-current. In sup- 
port of this it may be mentioned that the number of the 
small nuclei and the nucleated elements which corre- 
spond to the lymphoid corpuscles is the smaller the more 
thoroughly the pulp had been previously washed and during 
the injection of the blood-vessels of the spleen. Besides, 
when describing the structure of the pulp of the human 
spleen we shall become acquainted with another important 
fact in support of the above theory. 
What has been said until now of the structure of the pulp 
is consistent only with the theory that the venous radicles of 
the pulp are represented by a labyrinth of spaces in the 
matrix, advocated, as mentioned above, by Wilhelm Miller, 
Frey and others. 
With reference to the stroma of the pulp of dogs I have 
to mention the presence of multinuclear masses belonging 
to, or forming a part of, it. MKolliker describes (1. c., p. 493) 
amongst the pulp-cells of young animals “ finely-granular 
cells containing numerous (4 to 10 and more) nuclei, grouped 
together. ... These peculiar multinuclear cells, which 
very much remind one of the multinuclear cells of the 
marrow of bones, and which he found in the blood of the 
liver of embryos, Kolliker believes to originate in the splenic 
pulp. The same observer found them also present in the 
blood of the splenic veins. ‘The multinuclear cells show 
(l. ¢., p. 25) the peculiarity that their nuclei originate simul- 
taneously by budding from one single nucleus. 
These observations of Kolliker 1 am able not only to con- 
firm, but I am also ina position to add to them the following : 
the multinuclear cells are only peculiar enlargements of the 
general matrix of the pulp, which, as we have seen, is a 
honeycomb of membranous structures. Thus also the multi- 
nuclear cells are flattened, branched structures in connection 
with the branched membranes of the stroma, as is accurately 
represented in figs. 1—4; they are of different sizes and are 
composed of the same pale transparent irregularly granular 
substance as the general matrix. ‘The nuclei are seen to 
exhibit very active budding (see figs. 1—4). The smallest 
cells of this kind—z.e., the youngest—possess a large nucleus 
which stains deeply in hematoxylin, and which shows 
several more or less deep constrictions and knobs (see fig. 
4). The number of these latter increases in proportion to 
the size of the cell. Figs. 2 and 3 are very instructive 
examples. These flattened multinuclear enlargements of the 
general stroma are met with most frequently in the splenic 
