272 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 



tinctly trilobed, as though constricted, having the form of a linear series 

 of three heads (Fig. 13). In the typical condition, each of the three 

 parts is of equal size, and all are arranged in a straight row. It often 

 happens, however, that the three masses are not arranged in a straight 

 line, the result being various odd shapes. The elbow-like form (Plate 3, 

 Fio-s. 14, IC), so common in Uuio, arises in Haminea by a lateral displace- 

 ment of one of the end masses. In one instance the chromatin was found 

 in the early stage in a typical tetrad condition, each of the four masses 

 'being held together by linin (Plate 2, Fig. 9). 



The chromosome vesicles contain in addition to chromatin and linin a 

 substance, probably a fluid, which takes a faint plasma stain. This sub- 

 stance is more abundant in the peripheral region of the vesicle, which in 

 section gives to the wall of the vesicle an irregular outline on its inner sur- 

 face (Plate 2, Figs. 9, 11, 12 ; Plate 3, Figs. 15, 17). As the chromatic 

 matter increases in amount, the substance taking a plasma stain dimin- 

 ishes until there is no staining reaction in it. 



A broad band of fibres extends from each vesicle to the centrosome. 

 In the earlier stages there are granules on the fibre (Plate 2, Fig. 9). 

 AVhen the vesicles lie in the centre of the germinative vesicle, the 

 arrangement of the fibres gives the appearance of a central spindle. 



With the concentration of the chromatin into a single definite mass, 

 the walls of the chromosome vesicles become less distinct and soon dis- 

 appear, simply fading away, probably being dissolved in situ. As soon 

 as the -walls have disappeared, the chromosomes lie free at the equatorial 

 plane of the spindle (Plate 3, Fig. 16). The maturation figure now has 

 the typical appearance, except that the chromosomes still retain their 

 tripartite condition. 



There are three possible ways in which to account for the formation of 

 the chromosomal vesicles. It is possible that the bodies which I have 

 termed plasmosomes (p. 266) may be karyosomes, and that each of them 

 may have given rise to a vesicle ; but this view loses its force when we 

 remember that these plasmosomes may or may not be present in the 

 ovarian egg, and that the other vesicles are formed independently of 

 the first one. 



It is more probable that we are to look for the origin of the vesicles in 

 a metamorphosis of the linin itself; in which event there are two possible 

 explanations. First, there may be a direct rearrangement of the linin 

 to form a vesicle. This is probably true in part, but does not seem fully 

 to explain what happens. Secondly, the chromatin masses located at 

 the nodes may by vacuolation gradually form quite a regular vesicle. 



