SPERMATOGENESIS OF EUSCHISTUS 739 
6 to 14) certain autosomes are regularly in contact with the 
nuclear membrane at the distal pole of the latter, where it touches 
the idiozome (id). This continues until the disappearance of 
the idiozome (figs. 47, 48), after which none of the autcsomes 
maintain such a position. In earlier stages (figs. 6 to 12) there 
appear to be two autosomes invariably in this position, but later 
it would seem that the ends of more than two touch the nuclear 
membrane at the idiozome pole (figs. 14-20, 23, 25, 27-29, 33- 
43). Otherwise autosomes do not end upon the nuclear mem- 
brane except in connection with the growing plasmosome (to 
be described in another place). Whether these are particular 
autosomes, or whether it is a matter of chance which of the auto- 
somes occupies this position, I have not been able to determine, 
but this is a regular phenomenon of every cell of these stages. 
The ends of such autosomes are coiled upon the nuclear mem- 
brane, making there a plate of chromatin and not a dise of sepa- 
rated granules as described by me in 1898. This ‘chromatin 
plate,’ as it may be called for convenience, is invariably ata 
particular pole, next the idiozome body, and probably has some- 
thing to do with the genesis of the latter as will be shown later. 
This plate is brought to an end by the autosomes withdrawing 
rom the nuclear membrane, and leaving no chromatin behind 
them. But I have seen no evidence that any chromatin is bodily 
eliminated from the nucleus at this point. 
Following upon the pachytene (figs. 48 to 45) comes the strep- 
sinema or diplotene stage, the earlier part of which about coin- 
cides with the maximum size of the sphere and its dissolution 
(figs. 46 to 57). The pachytene threads gradually unravel so 
that each geminus comes to appear distinctly double for the 
second time. The two components of each geminus then appear 
more or less spirally twisted around each other and their shorten- 
ing is due to this twisting rather than to any linear contraction. 
As arule the two components of a geminus begin to separate first in 
their middle portions, while their ends are still closely apposed (figs. 
50, 54,56). Three entire gemini are drawn in figs. 46, 47, 49, 54, 
55, two entire ones in figs. 50, 52, and four in fig. 56. In fig. 48 
five gemini are shown in their entirety, while the sixth, to avoid 
