Data of the preceding article provide some instructive information about the time of the occurrence of chromosome nondisjunction under the following assumptions (cf. Shadley et al., MGNL 58:158): a) seed size reduction is due to an aneuploid chromosome constitution in the endosperm, b) only a single nondisjunction event occurs at either the micropylar or the chalazal pole during the embryo sac development, c) after the first post-meiotic division, the two nuclei move to the opposite pole, d) each of the four nuclei at either pole has an equal chance to become polar nuclei before cellularization, e) any of the four micropylar nuclei can randomly develop into the egg and f) all of the nondisjunction events take place at the same cell division.

Nondisjunction does not occur at the first division. If it did, nuclei at one pole would have a chromosome constitution complementary to those of the other pole, i.e. nuclei at one pole being hypoploid and those at the other pole being hyperploid. The endosperm would be euploid, since the hyperploid polar nucleus compensates the hypoploid one. No seed size reduction would be expected. The consistent occurrence of small seeds contradicts this hypothesis.

Nondisjunction may take place at the second division. If a chromosome fails to disjoin at the micropylar pole during this division, half of the kernels would carry a hyperploid embryo and the other half, a hypoploid embryo. Among the former, 2/3 of them would be in contact with a hypoploid endosperm, and 1/3 of them with a hyperploid endosperm. Among the latter, 2/3 of the embryos would be associated with a hyperploid endosperm and 1/3 of them with a hypoploid endosperm. On the other hand, nondisjunction at the chalazal pole would give kernels carrying only an euploid embryo. Half of these would have a hyperploid endosperm, and the other half a hypoploid endosperm. Altogether, for the small kernels with an aneuploid endosperm, the rate of euploid, hyperploid and hypoploid embryos is 2/4, 1/4 and 1/4 respectively. The present data (13:4:5) fit this expectation (2:1:1) very well (P>0.5).

The agreement of the present data with the supposition of nondisjunction at the second cell division can only be accepted with reservation. This is due to the fact that the assumption of an aneuploid endosperm as the basis of seed size reduction may not be valid. As mentioned in the previous article, the size of the small kernels selected for chromosome counting is half the size of the normal kernels or smaller. To this writer's knowledge, reduction in seed size caused by a hyperploid endosperm has never been reported to reach 50%. Therefore, it is more realistic to assume that the endosperm of the small kernels is hypoploid. Under this assumption, the expected ratio of euploid, hyperploid and hypoploid embryos of the smaller kernels would be 3:2:1 (1/4:2/12:1/12), which is statistically consistent with the data (P>0.25).

Nondisjunction does not occur at the last division for the following reasons: If it occurs at the micropylar pole, 2/4 of embryos would be euploid, 1/4 of embryos hyperploid and 1/4 of embryos hypoploid. 2/3 of the euploid embryos would be associated with an aneuploid endosperm. and 1/3 of them with an euploid one. On the other hand, 2/3 of the hyperploid and hypoploid embryos would be associated with an euploid endosperm and the remaining 1/3 with an aneuploid endosperm. When nondisjunction takes place at the chalazal pole, all embryos produced are euploid, but half of the endosperms are euploid and the other half, aneuploid. Taking all into consideration, the frequency of aneuploid endosperms for euploid, hyperploid and hypoploid embryos is 10/24, 1/24 and 1/24 respectively. The ratio is 10:1:1, which is greatly different from that of the observed data.

Taking into consideration that the endosperm of the small kernel is hypoploid, nondisjunction is more unlikely to occur at the third division. Under this assumption, a hypoploid endosperm would never be associated with a hypoploid embryo, since, by being so, the embryo sac would have to carry two hypoploid nuclei at the micropylar pole. When a chromosome fails to disjoin at this division, such a situation would never occur. Yet, 5 (22%) embryos were found to be hypoploid in this study.

The results likewise rule out the possibility that the egg nucleus is a sister nucleus of one of the two polar nuclei (with the complementary constitution), since no small kernels would be expected to have an euploid embryo. The fact that 13 out of 22 embryos are of this type rejects this supposition.

In conclusion, the data of this study fit satisfactorily the supposition of nondisjunction at the second post-meiotic division. The aneuploid nature (hyperploid or hypoploid) of the endosperm of the small kernels does not affect this conclusion.

Bor-yaw Lin

Return to the MNL 59 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage