Saratov State University
Development of haploids on haploids of parthenogenetical maize lines in crosses n x 2n by different pollination delay terms --Smolkina, YV, Tyrnov, VS It was demonstrated earlier that kernels with both haploid and diploid embryos may develop on haploids of parthenogenetical maize lines pollinated by pollen of diploid plants. Haploids on haploids were produced successively in 5 generations (Tyrnov, MNL 71:74, 1997).

In our work we investigated the influence of different terms of haploid pollination delay on setting of kernels and ploidy level of plants, developed from them. We used parthenogenetical lines AT-1 and AT-3, described earlier in MNL, 1997 (Tyrnov, 71:73-74; Enaleeva, Tyrnov, 71:74-75). For haploid producing, these lines (and further the haploids themselves) were pollinated by an embryo marker, which is characterized by low haplo-inducing ability (about 0.05%). Therefore, almost all haploids arose as a result of a genetically conditioned predisposition to parthenogenesis of maternal forms. Haploids were discovered by the method of genetic marking, morphometric and cytological analysis (Khokhlov, Tyrnov, Methods of Diagnosis of Haploids. Haploidy and Selection. Moscow, pp. 14-25, 1976). The experiments included "ideal" and "not ideal" variants. The ideal variants were characterized by an equal number of haploids in different experimental seasons (2000, 2001), by equal sizes of plant and ears, and by similar conditions of cultivation. Pollination was made in 2-3 and 7-8 days after appearance of silks. Not ideal variants included the smaller plants, with different sized ears. The terms of pollination can be conditionally divided into "early" (1-3 days) and "late" (5-9 days). Sometimes pollination was made on the first day, and repeatedly on the second or third day. Such differences are connected with the use of haploids, which for some reason were not discovered in other experiments. Haploids might grow in thickset crops and in the shadow of powerful diploids. It is possible that some haploids were a part of twins such as n-n and n-2n. We suppose that variants that are not ideal are not less valuable than ideal ones, since in further work it is impossible to be sure that haploids will always appear in perfect conditions.

Ideal variants of both lines on 20 haploids, early and late in terms of pollination, were examined yearly. The differences in the different years were not discovered. So, we give the data for 2 years. In line AT-1, 248 kernels were set on 40 haploid ears, on average 6 on one ear, with a variability from 1 to 20. With pollination delay, only 18 kernels were set on 40 ears. In line AT-3, according to pollination terms, 186 and 16 kernels were set. Consequently, pollination delay led to a reduction of kernel setting of more than 10 times. With early pollination, kernels were produced on all 80 haploids. Nine ears gave one kernel, 14 gave two. Other ears had 3-9, 14, or 16-21 kernels. With pollination delay in line AT-1, 65% of the ears did not form kernels, in line AT-3 72.5%. The maximum number of kernels on an ear did not exceed 2.

The differences were observed in ploidy level of seedlings, obtained by growing kernels which set on haploids. Early pollination of the AT-1 and AT-3 lines resulted in 3.6 and 2.7% haploids, late pollination resulted in 78 and 75%. Taking into account that on a majority of ears pollinated late only haploids have been set, we may speak about a 100% frequency of parthenogenesis. However, it is necessary to remember that we are discussing single specimens, not mass production of haploids with such a frequency. Nevertheless, this index can be used as a diagnostic sign in selection on parthenogenesis.

Enough similar results were obtained in variants that were not ideal. With late pollination, kernels set 8-15 times more seldom, but the occurrence of haploids among them reached 50-60%. With early pollination, the frequency of haploidy did not exceed 5%.

Besides haploids in all variants, the occurrence of twins was noted with a frequency of 1-12%. The types of twins were n-n and n-2n. The phenomenon for haploids was noted first. It connects with parthenogenesis and probably can be its diagnostic sign.

It was noted above that in line AT-1, haploids on haploids were produced successively in 5 generations (1986-1990). Beginning with 1990, an analogous experiment was conducted with haploids of the AT-3 line. We could produce haploids on haploids this time in 11 generations. Signs of degeneration were not established.

Hence, some conclusions may be made from the data obtained. With pollination delay, signs of reduced parthenogenesis manifest most clearly: the set of haploids on haploids, 100% occurrence of haploidy, polyembryony of n-n and n-2n types. However, taking into account that about 70% of ears do not form kernels, and on the rest only single kernels develop, pollination delay will be expedient only in those cases when a line gives many haploids or if a single haploid can be cloned.

The work was supported by a grant from the Russian Foundation for Basic Research (01-04-49385).

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to the MNL 77 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page