94 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 
outside developing a vacuole of their own like a pseudonucleus. These 
as well as the first body are rare exceptions destined ultimately to 
disappear, and consequently take no part in the formation of the 
spermatozoon. To give such temporary, accidental bodies a name 
can only lead to confusion. 
«/ 
Development of the Spermatozoon. 
With the preceding orientation of the essential parts of the sperma¬ 
tid — centrosome, nucleus, nebenkern, and cytoplasm — in mind, the 
transformation of the spermatid into a spermatozoon can be expressed 
in a few words. The spermatozoon is simply a greatly elongated 
spermatid, consequently a greatly elongated and compressed cell. 
The originally spherical spermatid (pi. 13, figs. 39, 39a) early begins 
to elongate. It becomes irregular, wedge-shaped, and pointed (pi. 
17, fig. 125-132). The giant spermatids arising normally from the 
giant spermatocytes, are almost from the beginning in relation with 
the cyst wall (pi. 13, fig. 39). They all sooner or later acquire such a 
relation (pi. 14, figs. 40, 46). 
That part of the cytoplasm in which the nebenkern is located, grows 
out into the tail of the spermatozoon. The cells are now so oriented 
that the tail grows toward the center of the cyst (pi. 14, fig. 40). As 
this elongation continues, the cyst bulges out at one point, toward 
which the growing tails are directed (pi. 14, fig. 41). There is evidently 
considerable lateral pressure. Growth of the cell is in the direction 
of the free pole of the cell, which owing to its relation to the enclosing 
cyst cell and the surrounding spermatids is necessarily the tail end. 
The axial filament .— While the cell is thus being compressed into 
the elongated form, there grows out from the nucleus a fiber which 
penetrates the nebenkern, and extends apparently along the path 
made by the original intermediate fibers. It soon emerges from the 
cytoplasm at the pointed tail end as a long flagellum (pi. 17, fig. 133- 
135). 
Originally there may be three or four of these fibers; but they finally 
unite; and at the point of union a knot or small enlargement some¬ 
times appears. This is apt to be mistaken for a centrosome. The 
point of union varies, and consequently the knot is sometimes seen 
in the center of the nebenkern (pi. 17, figs. 133, 139), at the distal or 
