412 BOWEN. 



in Figure 20 more or less from the indications of its length given by the 

 bunch as a whole. Smear preparations show the actual length to be 

 somewhat greater. 



The study of smears gives a much better idea of the mature sperm 

 than sections, since the former method furnishes an abundance of 

 complete, isolated heads. Sperm heads from smears (Fig. 21) show 

 one very marked difference from those in sections, the width of the 

 head being apparently much less. The thickness of the head in 

 sections is doubtless largely due to the stain, which not infrequently 

 makes dense chromatic structures appear larger than they really are, 

 especially if the extraction of the stain is not extreme. Two other 

 points of difference are also brought out by smears. One of these is 

 the exceedingly long and delicate (its width is much exaggerated in 

 Fig. 21) whip-like lash at the tip of the head, terminating in a minute, 

 slightly elongate thickening. This lash is presumably of acrosomal 

 origin. The other point is the definite expansion of the basal end of 

 the head (Fig. 21), which seems sometimes to stain more intensely 

 than the rest of the head. This probably corresponds to the some- 

 what similar portion at the base of the normal sperm heads (see Fig. 

 177, Bowen '22a), for which reason I believe that it probably repre- 

 sents the centrioles (" centrosomal middle-piece"). This would fall 

 in line with the fact already noted, that the centrioles in the large cells 

 are proportionately larger, while the available chromatic material is 

 constant in amount. 



The structure of the sperm tail was described in my second " Study," 

 but one feature of the tail of the large sperm calls for additional com- 

 ment. I refer to the enlargement already noted as a constant occur- 

 rence near the tip of the tail (Text-figs. 3B and 5C). Smears fixed 

 in osmic acid fumes indicate that this enlargement is due to two bleb- 

 like swellings on the threads, probably of mitochondrial (nebenkern) 

 origin, which pass along either edge of the ribbon-like tail (Fig. 22). 

 The meaning and fate of this swelling is unknown. Figure 22 also 

 shows very clearly how the marginal threads gradually narrow into 

 the termination of the sperm tail. 



The features described in this section as accompanying the forma- 

 tion of large sperms, are as constant in occurrence as is the poly- 

 megalous condition itself. In only one case (a specimen. No. 6, of 

 Murgantia), have I found a departure from this rule. In this indi- 

 vidual, the early stages are similar to those in the normal sperm, even 

 including the formation of the groove along one side of the head. But 

 at some point in the elongation of the head, the sperm revert to the 



