672 THE AMERICAN NATURALIST. [VoL. XXXIV. 
tissue, the author found them in typical form, not only in the mon- 
aster stage, but even as early as the spireme stage. In the former 
the central corpuscle, stained black in iron-haematoxylin, was sur- 
rounded by a typical polar star. Also in the diaster stage the centro- 
somes continue to be evident; the axis of the central spindle is at 
first straight, but later may become bent; the interzonal filaments, 
with their equatorial thickenings, become pressed together to form 
the deeply staining ** Zwischenkórper," which here, as in other cases, 
is not dependent on the formation of a cell wall In the [cup- 
shaped] depréssion of the nuclei of the diaster stage a remnant of the 
centrosome and the achromatic cone fibres is still visible, and after 
the formation of a nuclear membrane around the two masses of chro- 
matin, establishing the two daughter-nuclei, traces of the * Zwischen- 
körper " and central spindle are still to be seen. 
During karyokinesis the surrounding protoplasm acquires a granu- 
lar appearance. The granules, which are deeply stained in iron- 
haematoxylin, appear in the monaster stage ; in the diaster stage and 
the following anaphase these increase in number, so that the mitotic 
figure is surrounded by coarsely granular protoplasm. 
In addition to the karyokinetic, there is also an amitotic nuclear 
division, and while in general it may be said that the latter method 
replaces the former, the author is unable to say just when the one 
ceases or the other begins. Both processes are, indeed, to be seen 
in older embryos at the same time and in the same muscle. The 
first indication of the non-mitotic division is to be seen in the nucle- 
oli, of which there are at first one or two to each nucleus. The 
nucleolus elongates, becomes narrower in the middle, at length 
dumb-bell-shaped, and finally divided ; the two nucleoli then move 
apart. This process may be repeated. The chromatic substance 
meanwhile becomes collected into lumps of irregular form, which 
are, however, uniformly distributed through the nucleus. 
The actual division of the nucleus may take place in one or the 
other of two ways : First, by the formation of a thin, flat partition, 
which is perpendicular to the axis of the elongated nucleus or, often, 
oblique to it. Sometimes only a single partition is formed, but 
often the nucleus is divided by such partitions into as many as six, 
or even more, portions — the new or daughter nuclei; secondly, by 
the formation of invaginations from [z.e., constrictions of] the periph- 
ery of the nucleus this very much elongated body is finally divided 
into two, or sometimes several, not always equal, parts, each of whic 
contains a single nucleolus or, less often, two. The daughter-nuclei 
