18 
BULLETIN OF THE BUREAU OF FISHERIES 
tion of chromatin develop directly into flagellates, since there is apparently a tend- 
ency for the chromatin to collect on opposite sides of the nucleus in many cells that 
are evidently in the schizogenic cycle. Later, as shown in Figures 50 to 53, the 
cells become elongated, with the nucleus at one end containing the paired chromatin 
masses symmetrically arranged, one at each side. Between the chromatic bodies 
there is always some residual chromatin, which gradually loses its affinity for 
stains. At a somewhat later stage the outline of the original nucleus has disap- 
peared, and each of the chromatic bodies now forms the karyosome of a distinct 
nucleus (figs. 54 to 56). These nuclei show a striking resemblance to the nuclei 
of the flagellates, each karyosome being surrounded by a membrane separated 
from it by a clear area. In addition, there is usually a deeply staining body, evidently 
the blepharoplast, between the paired nuclei. The position of the blepharoplast 
suggests that it is derived from the residual chromatin left in the nucleus after the 
formation of the karyosomes. It has been impossible to distinguish flagella on 
the intracellular parasites at any stage, but in a number of instances there were 
structures that appeared to be developing axostyles, although in no case were they 
very distinct. However, in spite of the fact that no clear case of the presence of 
axostyles or flagella has been demonstrated, it is nevertheless believed that such 
a series as shown in Figures 48 to 56 can be explained only on the assumption that 
they represent a gradual metamorphosis of the intracellular parasites into the free- 
swimming flagellates. 
Since no intracellular stage has previously been attributed to any flagellate at 
all closely related to Octomitus, it may not be out of place to briefly review at this 
time the evidence for the assumption that the intracellular parasites and flagellates 
are simply different stages in the life cycle of a single species. Undoubtedly the 
strongest evidence in favor of such a belief is the fact that various stages can be 
found among the intracellular parasites which, when arranged in series, show a 
gradual transition from the typical intracellular form to the fully developed flagel- 
late. While it is true that such stages are not common, they can scarcely be 
regarded as exceptional, since several can often be found on the same slide. 
Furthermore, it has been found that the intracellular parasites and flagellates 
always occur together. In no case have flagellates been found in fish in which the 
intestinal epithelium did not contain at least a few of the intracellular forms. It 
is true that the latter have been found in fish when no flagellates could be demon- 
strated in the intestine, but in all such cases flagellates were found in other trout 
from the same locality. When we take into consideration the wide distribution of 
Octomitus, the constant association of the flagellates and intracellular parasites 
becomes all the more significant. It is difficult to explain such a relation except 
on the ground of specific indentity. 
Of special significance is the fact that an undescribed species of Octomitus occurs 
in the fan-tailed darter {Etheostoma jiabellare) at White Sulphur Springs, W. Va., 
which, like 0. salmonis, lives in the anterior end of the intestine. Sections of the 
caeca of infected fish show the presence of a few intracellular parasites (fig. 57) 
that are evidently distinct from those in the trout. Presumably these are the intra- 
cellular stages of the species of Octomitus peculiar to the darter. 
