March, 1908.] 
On the Cytology of Synchytrium. 
2 79 
chances of securing favorable sections of both the daughter nuclei 
are rather remote. In the present case with sections 2 microns 
thick, the chance is only about one in twenty-five, neglecting 
the thickness of the nuclei though they are of about the same 
thickness as the sections which reduces the chances very mater¬ 
ially. It is evident, then, that it is not easy to get a full series of 
nuclei in which both members of the pair show. Most of the 
drawings therefore delineate only one nucleus. 
While centrosomes are not demonstrable on the spindles .at 
metaphase or early anaphase (fig. 1 ef. Stevens 07 figs. 18—21), 
in what is interpreted as telophase (fig. 2) there are found enor¬ 
mous asters. The manner of their appearance has not yet been 
made out. Though usually located near the former axis of the 
spindle they are by no means accurately placed at its poles as can 
be seen by consulting almost any of the figures. In structure 
also it may be seen that they vary greatly. Sometimes there is a 
single deeply staining granule (centrosome proper) at the centre 
of the radiations (figs. (>, 7, 8, 11); sometimes the deep staining 
taken at the centre seems to be due simply to the convergence of 
radiations (fig. 5). In only one case and that not very distinct, 
was there seen anything approaching a clear centrosphere around 
the central granule. But very often there is more than one 
granule. The different granules may be located at the focus of a 
single aster so as to look like a dividing centrosome (fig. 2). 
They may be scattered about the focus without any very defi¬ 
nite relation to the rays( fig. 3). More often the different gran¬ 
ules are the centres of separate asters so that there appear two 
centrosomes (fig. 2, 4). Rarely there may be more than two dis¬ 
tinct centrosomes; figure 9 shows a case where there were three 
connected by heavy fibrous bands of kinoplasm, while each has 
its own aster. Figure 10 shows an anomalous condition where 
the centrosomes are located at nearly opposite poles of the nucleus 
while their rays meet so as to extraordinarily resemble the am¬ 
phiaster common in animal mitosis. Judged by itself this nu¬ 
cleus would seem to be in the prophase of division for in addition 
to the amphiaster the chromatin is arranged similarly to the 
spirem of the prophase. Such an interpretation seems, however, 
entirely inadmissible, since the chromatin in prophase does not, 
so far as is now known, assume such a spirem and the spindle, 
being intranuclear, has no relation to any such amphiaster. 
This condition was seen only in the single nucleus located in a 
cyst where all the other asters conform to the usual type. Fig. 
7, however, shows a condition interesting in comparison; though 
this may be simply a case where the second nucleus of a pair was 
cut out of the section, leaving a part of its aster. Here, in addi¬ 
tion to the conspicuous aster at one pole of the nucleus, is another 
on the opposite side which though faint and lacking a verv definite 
