324 
ZOOLOGY: E. B. WILSON 
themselves in a ring shaped group to form a 'nebenkern-organ' lying at 
the base of the nucleus and surrounding the blepharoplast (centriole) 
and the outgrowing axial filament (figs, h — k). Two hundred cases thus 
far tabulated give 73% with six chondriosome-spheres, 16% with five 
and 11% with seven. Other numbers have not been observed. 
The sperm thus passes through a stage in which it possesses a neben- 
kern closely similar to the definitive middle-piece of certain nemer- 
tines, pelecypods and annelids (Retzius) ; but this is only a temporary 
condition. Becoming closely applied to the axial filament, the ring of 
chondriosomes is in later stages progressively drawn out backwards to 
form the envelope of the flagellum, apparently extending as far as the 
end-piece of the latter. There are some indications that a remnant 
of the chrondriosome-ring may finally be cast off along with the general 
protoplasmic remnant which, as usual, is sloughed off from the elongating 
flagellum, but this has not yet been clearly established. I have not yet 
been able to determine whether the envelope develops a spiral structure 
analogous to that so clearly seen in Centrurus. 
From the foregoing account it is evident that chondriosome-material 
having the same origin, fate and (presumably) physiological signifi- 
cance may be distributed to the germ-cells by processes widely differ- 
ent even in nearly related animals. In one of these scorpions the dis- 
tribution is effected by a definite process of division, in the other by an 
operation that has at least the aspect of a hit-or-miss segregation, and 
one that gives only an approximate equahty of result. On its face this 
would seem to indicate that a wide distinction should be drawn between 
chondriosomes and chromosomes in respect to their power of division 
and their relation to heredity. It is of course possible that in both the 
cases here described the chondriosomes may multiply by division at an 
earher period (a point now under investigation). It may also be sus- 
pected that division of the chrondriosome-ring is not an independent 
or autonomous act but depends upon the division of other elements 
with which it is associated. In either case the facts raise interesting 
questions concerning the power of division on the part of the several 
cell-components and the relation of this power to the principle of genetic 
continuity in general. 
^Certain special exceptions, such as the supernumerary chromosomes or the unequal 
small bivalents of Orthoptera, are here disregarded since they are readily exp'icable in 
accordance with the general rule. 
^Arch. Zellforschung,9, 3, (19X3), ■ • 
