December 1899] MULTINUCLEAR CELLS 435 



cells of many Monocotyledons (9), the tapetal cells in the sporangia of 

 Angiopteris (4), the generative cells of the vessels in Dioscoreaceae (3), 

 the older parenchyma cells of Taraxacum officinale (15), the large 

 parenchyma cells of Cercus niv.ltangularis (18), the young, elongated 

 pith-cells of Ochrosia coccinca (18) have all been shown, by various 

 observers, to be furnished with a plurality of nuclei. 



All our experience teaches us that wherever a number of nuclei 

 appear (whether these be sooner or later separated by a cell-wall, or 

 remain together in a multinuclear cell), they arise from the division of 

 an original mother - nucleus. When a nucleus divides into two 

 daughter-nuclei, it does so by one of two ways. 



Either it becomes constricted here or there, and without more ado 

 breaks into two or more parts, or it first passes through a complicated 

 series of preparatory stages in which certain of its internal parts 

 describe the most curious " figures," and then only separates into two 

 dauo-fiter-nuclei. In the former case the division is said to be direct 

 or simple fragmentation, in the latter it is described as indirect or 

 karyokinetic (7). It is generally supposed that a nucleus which is 

 fragmenting has lost the power of dividing activity by karyokinesis. 

 The great German cytologist, Strasburger, writing in 1880, says: 

 " According to my entire experience karyokinetic division and frag- 

 mentation cannot be brought together, and certainly one cannot replace 

 the other" (17). 



Eecently, however, the Italian observer, Buscalioni (2), has shown 

 that this separation of the two forms of division is by no means neces- 

 sarily the case, and that in the development of the embryo-sac of Vicia 

 Faba, Lupinus, Fritillaria impcrialis, and Lcucojum vcrnum, and in the 

 laticiferous tubes of Urtica, fragmentation and karyokinesis may take 

 place side by side with one another, or the same nucleus may first 

 divide directly and then indirectly. Moreover, both Buscalioni and 

 Dixon (5), as well as Miss Sargent (14), have observed a curious con- 

 dition of the nucleus in which some of the preparatory stages of karyo- 

 kinesis are gone through, but before the process is complete the nucleus 

 divides directly. Whether this is really an intermediate stage pointing 

 to the fundamental identity of the two processes, as the authors appar- 

 ently suppose, is doubtful. The facts clearly indicate, however, that 

 the two varieties of division are by no means incompatible with one 

 another (19). 



Some observations which I recently made on certain vegetative 

 cells of some Gramineae give additional support to this view. 



It does not seem to be as widely known among botanists as it 

 should be that in certain members of the Gramineae, especially in Zea 

 Mays (Indian corn), multinuclear cells of the most pronounced 

 character are of frequent occurrence. 



If a section be made through the stem-region of a young plant so 

 as to pass through the enveloping leaf-sheaths, the parenchyma cells of 



