SPERMATOGENESIS OF EUSCHISTUS ton 
condition of periodical or rhythmical occurrence, perhaps associ- 
ated with periodical discharges of nuclear substance, the same 
cell passing through a series of such conditions; in no other way 
at-least can the sporadic occurrence of synizesis be explained. 
For this is clearly a.normal phenomenon and not an artefact due 
to faulty preservation of the nuclei, for it is seen after the use of 
all fixatives and only up to the pachytene stage. It is then 
associated with the time when the nuclear membrane is exceed- 
ingly delicate. Figs. 23 and 24 from one cyst, and figs 30 and 31 
from another, illustrate the different appearance that cells of the 
same cyst may show. But even during the acme of such syni- 
zesis (figs. 24, 31, 33, 27, 40) there is no evidence that the auto- 
somes become fused or in any way lose their identity, all-the 
appearances indicate rather that they are then only very closely 
massed; for in thin sections of such masses the outlines of the 
individual autosomes may always be discerned. 
The leptotene condition with its slender but still isolated auto- 
somes (figs. 25 to 28) passes gradually over into the zygotene, 
leading to the condition where the autosomes lay themselves 
parallel in pairs (figs. 41, 42). It is difficult to be sure of the 
exact sequence of all the stages shown in figs. 29 to 42, yet there 
can be no doubt that the leptotene condition changes gradually 
_ into the zygotene. It will be recalled that no continuous chroma- 
tin spirem had been produced, therefore the bivalent autosomes 
cannot be formed by any transverse breaking of such a spirem. 
Already in the leptenema there may be some trace of a parallel 
juxtaposition of autosomes, and this would seem to have pro- 
ceeded further in the stages of figs. 29 and 30. Then we reach 
the interesting conditions of figs. 32 to 35 and 38 to 42; in figs 32, 
34, 42 all autosomes are shown in their entirety. In fig. 32 most 
of the autosomes show parallel arrangement, especially marked 
in the case of the pair that touch the plasmosome (Pl). Fig. 
34 is especially important; it exhibits, in addition to the two 
idiochromosomes (D, d), precisely twelve autosomes, four of 
which compose two gemini; in other words, this nucleus shows 
eight univalent and two bivalent autosomes. This figure, made 
with the greatest care and redrawn on different occasions, estab- 
