In the fifth stage the central thickenings of the spindle separate into two sets, which travel symmetrically outwards towards the clear masses, growing in size during the process. They remain however united for a short time by delicate filaments--named by Fol connective filaments--which very soon disappear. The clear masses also increase in size. During this stage the protoplasm of the ovum exhibits active amoeboid movements preparatory to division. In the sixth stage, which commences when the central thickenings of the spindle have reached the clear polar masses, the division of the ovum into two parts is effected by an equatorial constriction at right angles to the long axis of the nucleus. The inner vitelline membrane follows the furrow for a certain distance, but does not divide with the ovum. All connection between the two parts of the spindle becomes lost during this stage, and the thickenings of the fibres of the spindle give rise to a number of spherical vesicular bodies, which pass into the clear masses and become intermingled with the granules which are placed there. The radii of the stars now extend round the whole circumference of each of the clear masses. In the seventh stage the two clear masses become elongated and travel towards the outer sides of their segments; while the radii connected with them become somewhat bent, as if a certain amount of traction had been exercised on them in the movement of the clear masses. Shortly afterwards the spherical vesicles, each of which appears like a small nucleus and contains a central nucleolus, begin to unite amongst themselves, and to coalesce with the neighbouring granules. Those in each segment finally unite to form a nucleus which absorbs the substance of the clear mass. _The new nucleus is therefore partly derived from the division of the old one and partly from the plasma of the cell._ The two segments formed by division are at first spherical, but soon become flattened against each other. In each subsequent division of these cells the whole of the above changes are repeated. The phenomena which have just been described would appear to occur in the segmentation of ova with remarkable constancy and without any very considerable variations.