Some Points in the Spermatogenesis of Mammalia. 157 



structure, the archoplasmic metamorphosis, and tlie reappearance of 

 the centrosomes, are eifected without the spermatid losing its spherical 

 or polygonal character; but after a while the cells begin to be lengthened 

 out, along an axis corresponding with that of the tail (Figs. 22, 24, 25). 

 As soon as this becomes manifest, a definite orientation of the cellular 

 constituents comes rapidly into view; I believe this to be due primarily, 

 if not entirely, to the pressure induced through the enlargement of 

 contiguous cells, and there seems good reason to believe that the 

 elongation of the cell as a whole originates through a similar cause. 



In every spermatid, the residual archoplasm now wanders further 

 fi'om the nucleus, taking up a final position along the axial thread of 

 the tail (Figs. 22, 24, 25 a). The chromatic body follows suit, but comes 

 to rest on the nuclear side of the residual archoplasm, quite close to 

 the nucleus and the structure at the base of the thread. The centro- 

 somes come also into close connection with the basal portion of the 

 axial thread, but their position is not nearly so uniform as that of 

 the chromatic body. The centrosomes, chromatic body, and the basal 

 structure of the axial thread, become now much confused together, 

 and this confusion of parts is heightened by the chromatic body dividing 

 into two portions of unequal size (Fig. 22). Between these two por- 

 tions the axial thread is seen to pass like a faint line, to a termination 

 in the little basal structure to which I have repeatedly alluded, and 

 which may be distinguished as a Cercosome (cc). 



It will be remembered that the spermatid centrosomes become 

 apparently confused with the two portions of the chromatic body and 

 this Cercosome, in such a way that the indi\idual identification of the 

 lesser structures becomes extremely difficult 



Concurrently with these changes, two others of marked significance 

 occur. The first relates to the nucleus, and appears as a collar-like 

 fold, which rises up around the basal attachment of the axial-thread; 

 as it grows, it more or less encloses the two portions of the chromatic 

 body and the delicate "shaft" of the tail, in a shallow cup (Fig. 22). 

 One side of the nuclear depression thus arising is ultimately prolonged 

 into a stiff projecting flap (Fig. 22, 24 h) which serves as a kind of 

 support for the protoplasmic sheath or shaft of the ..Hanptstiick" of 



