GANYMEDBS ANASPIDIS (nOV. GEN., NOV. SP.). 163 



front end, where there is usually a sort of pad of fine-grained 

 cytoplasm projecting back into the cavity (fig. 8). In the 

 main body, behind the neck, is another spherical cavity, 

 apparently separated from that of the ball proper, and con- 

 taining a fluid that is not quite clear, but of a loose reticulate 

 structure (figs. 6, 7). Enclosing the hinder part of these 

 may sometimes be seen a dark crescent of nearly homogeneous 

 material (fig. 9). 



So far, so good. In other cases, however, we find quite a 

 different appearance, thei-e being only one cavity present, and 

 all traces of a neck having vanished (figs, 3, 5). Closer 

 inspection shows that the cavity corresponds with that of the 

 true ball, as its contents are perfectly cleai*, and it has a 

 pad of cytoplasm anteriorly. The dark crescent may come 

 directly behind it (fig. 5), while the thick body cuticle 

 extends completely over it. The question then is, what is 

 the relation between these two conditions ? 



It seems obvious that the ball can be extruded at will 

 — but in what way? Is it evaginated (pleurecbolic) or 

 is it acrecbolic, and, if the latter, is it pulled out by 

 muscular or elastic action or pushed out by some other 

 means ; and how is it retracted ? An examination of many 

 G-regarines (a task necessary owing to the absence of living 

 material, but laborious from the small size of the ball — 

 8 — 10 /x across), has made it seem probable that it is acrec- 

 bolic, and pushed out by the accumulation of a watery fluid 

 behind it. As far as I can make out, the structures and 

 processes concerned are as follows : — The dark crescent {s.t. 

 in figs.) is a tissue which has the power of secreting a fluid 

 (w.) into a space anterior to it, thus driving the ball out 

 through an opening in the body cuticle. When the ball is 

 retracted, the elastic cuticle would be closed over the anterior 

 end ; and when extrusion has taken place, it would press in 

 and form the thin neck. One animal (fig. 2) shows what I 

 suppose to be an early stage of extrusion : the hole is just 

 being enlarged, so that the cuticle at its edge stands out as a 

 well-marked rim {cut. rim). In later stages (figs. 4 and 6) 



