A cytological study of the egg of Cumingia with special reference etc. 249 
again segregated out in the original manner when the chromosomes 
reform”. The point seems to he justly urged by Foote and Strobell 
’09) that both the chromosomes and centrosome are rather the effect 
of cell activity than its cause. 
The Centrosome. 
The history of the centrosome in Cumingia is likewise ad- 
verse to the hypothesis of its genetic continuity. The primary aster 
consists of a distinct small deep-staining focal granule (centrosome), 
surrounded by a homogeneous pale centrosphere bounded by a "micro- 
some circle” and an outlying astrosphere. The centrosome then di- 
vides and an amphiaster is formed (Fig.7). As the amphiaster grows 
the centrosomes enlarge until the condition is attained illustrated in 
Fig. 10. The centrosomes are at first spherical but as the anaphase 
ensues they become irregulär (Fig. 11) and soon fragment and dis- 
appear (Fig. 12 and 14). Subsequently there is present only a centro- 
ephere through which are scattered pale-staining granules. This con- 
dition is somewhat similar to that described by Wilson (’06) for 
Toxopneustes where a "pluri-corpuscular centrum” is formed at this 
Stage. LiLLis’s (’Ol) findings in Unio are also similar though he re- 
ports here the continued presence of a small variable number of chro- 
matic granules. 
In Cumingia the original centrosomes completely disappear; nor 
are the products of their disintegration to be recognized in the finely 
granulär centrosphere of subsequent stages; nor is a centrosome again 
discernable nntil later cleavage stages (Fig. 33) when it reappears 
as a rather pale single central granule of the aster. The centrosome 
seems to have arisen de novo in the cytoplasm at the beginning of 
the maturation phase, as also the entire aster. One can of course 
not prove that it did not persist from the last oögonial mitosis all 
through the growth period but it is certainly unrecognizable until 
made conspicuous by astral rays. Again at fertilization the centro- 
some seems to arise de novo. Hundreds of spermatozoa and early 
male pronuclei have been studied at various stages (Figs. 16 and 22) 
but no aster or centrosome could ever be seen in relation to it. It 
is only after the definitive male pronucleus is formed (Fig. 27) that 
an amphiaster can be seen preceding the nucleus somewhat as Lillie 
(’Ol) describes for Z7mb, and lying for a while separating the male 
and female pronuclei by the distance of its long axis. Seither here 
