BLACKMAN. — SPERMATOGENESIS OF THE MYRIAPODS. 513 



later stage (Figure 33). The chromosomes are of a distinctly dumb- 

 bell shape, the constriction at the middle of each one representing the 

 line of transverse cleavage at which point the chromatids separate 

 during the following metaphase. During the succeeding stages (Figures 

 33, 34) and in the prophase of the following mitosis (Figure 35) the 

 chromosomes retain this shape. In the metaphase of the second 

 spermatocyte (Figure 36) the dumb-bell shaped elements are arranged 

 with the constriction parallel to the equatorial plate and the separa- 

 tion of the component chromatids occurs at this place. This is the 

 reduction division of the chromosomes, and by it the parts of the 

 bivalent element are separated, doubtless at the same point at which 

 they united during the telophase of the last spermatogonium. 



In the meantime the nuclear spindle of the first spermatocyte remains 

 entirely independent of the centrosomes and their system of astral rays. 

 During the division and separation of the chromosomes the mantle 

 fibres elongate considerably and their polar ends approach the centro- 

 somes (Figure 32), but at no time do they abut upon it. Thus the 

 chromosomes are evidently not separated by the contraction of the 

 mantle fibres, but these seem merely to act as runways along which 

 the daughter chromosomes move as they approach the poles. This, 

 however, does not necessarily mean that the centrosomes do not have 

 an important function in the separation of the chromosomes, for the 

 influence of the chromosomes over the nuclear spindle is at all times 

 apparent. The nuclear spindle is oriented parallel to a line connecting 

 the two poles of the cell, and at all stages the mantle fibres converge 

 toward the centrosomes. It seems probable that the separation of the 

 daughter chromosomes is accomplished by an attractive force exerted 

 upon them by the centrosomes rather than that they move apart by a 

 force inherent in themselves. At any rate it cannot be disputed that 

 the centrosomes determine the direction of this migration. 



In a former paper (Blackman, -.05) I have given at some length 

 reasons for believing that the centrosomes and astral rays play an im- 

 portant part in the cytoplasmic cleavage. This work upon Lithobius 

 strengthens the belief there expressed by showing that similar phenom- 

 ena occur here also. The astral rays are not so conspicuous as in 

 Scolopendra, but in all other respects they are essentially the same, 

 and in Lithobius, just as in Scolopendra, I believe that the constriction 

 of the cell is brought about by a force which is inherent in the centro- 

 some and is exerted upon the cell membrane through the astral rays. 



The further changes in the maturation divisions offer but few points 

 of special interest, and will be treated very briefly here. After the 

 separation of the chromosomes in the first spermatocyte division, the 



VOL. XLII. — 33 



