My conclusions, which, I confess, are tentative and based mainly on studies of one kind of cell, the grasshopper neuroblast, may be summarized as follows.
The late prophase orientation of the chromosomes is a carry-over from the late telophase orientation. It is apparently maintained by means of the centromeres, which appear to be attached within a limited region of the nucleus throughout telophase, interphase, and prophase.
Metaphase orientation of the chromosomes may be explained as the resultant of two forces: a force involving the centromere and spindle, which is responsible for keeping the centromeres in the equatorial plane of the spindle, and a repulsion force involving the noncentromeric portion of the chromosomes, which results in a tendency toward uniform spacing of the chromosomes outside the spindle.
Anaphase separation of sister chromatids and their subsequent movement toward the poles of the spindle involves at least four distinct phases: (i) the initial poleward movement of the centromeres, which may be due to intrinsic repulsion or to a force acting between spindle and centromeres that produces an angle of almost 90 degrees between the separated and unseparated portions of the chromatids; (ii) the autonomous separation of the noncentromeric part of the chromosome; (iii) elongation of the spindle, beginning just after the sister chromatids are separated proximally and ending when the longer chromatids are about to lose contact distally; and (iv) the later movement apart of the daughter chromosomes, probably resulting from a pushing force exerted by elongation of the interzonal fibers.