1. It gave matroclinous haploids with a frequency of 90-100%.
The frequency of haploidy could be higher than 100% because of
the occurrence of twins and triplets like n-n and n-n-n types.
2. All haploids were produced by pollination. Without pollination the endosperm developed only to a certain stage, and its development was not completed. By the method of genetic marking it was shown that normal endosperm was formed only after fertilization. We suppose that this is connected with the necessity for the endosperm to have ploidy equal to 3n or divisible by 3. Consequently, according to cytological data, the line is characterized by autonomous reduced parthenogenesis, but genetic data show that pseudogamous haploid parthenogenesis takes place.
3. Side by side with haploids, diploid parthenogenetic individuals were found. Their origin has not been ascertained for the present, but it is connected, most probably, with spontaneous diploidization of haploids.
4. The highest frequency of haploidy was observed by delay of pollination (5-10 days). In connection with this, the idea seems attractive to create apomictic lines which, by super early pollination (for example, before emergence of silks), would keep the possibility of hybridization and selection by traditional methods.
5. AT-1 has purple colour of leaves, seeds and roots, which is controlled, probably, by genes A C R B Pl. This makes the diagnosis of matroclinous individuals difficult, but allows discovery of androgenic plants easily. The latter were produced by crossing with Mangelsdorf's tester and other lines, having genes lg, gl.
During the first four years there was a constant danger of losing this line, because only haploids arose and diploids were rare. In addition the plants were strongly affected by Ustilago maydis. So we were obliged to make selection on resistance and on decrease of haploidy level. At the present time we have selected a resistant form. The frequency of haploidy, by early dates of pollination, lies in the limits of 2-3%. It can be increased ten times by delay of pollination. Together with this, pollination delay leads to decrease of seed productivity, probably as a result of anomalous autonomous endospermogenesis.
We ascertained that ability for parthenogenesis is a nuclear-encoded trait. It can be transmitted to other lines by egg or pollen. It shall allow in the future combining of different traits with parthenogenesis in the same genome. Taking this into account, at the present time we have accomplished the following:
1. AT-3 line, the analogue of AT-1 line, but having yellow
seeds, green leaves, and white roots, was produced. It allows
diagnosis of apomixis with purple, brown and embryo markers. In
hybrid progeny from crossing of this line with Mangelsdorf's
tester, lines were selected with different marker genes. It is not
excluded that it can be possible to determine the chromosomes
2. By backcrossing the analogues with CMS of T, S and C-types, it was shown that these cytoplasms did not influence manifestation of parthenogenesis. In the future male-sterile lines can be required to exclude selection on sexual reproduction.
3. We produced hybrids between AT-3 and line-donors of some non-reduction genes. On this basis we suppose to test the possibility of production of unreduced apomicts.
4. Most wild apomicts are polyploids. So we also carry out work on production of polyploid apomicts.
5. Crosses of AT-3 with Tripsacum dactyloides were made. We propose to test the possibility of (1) introgression from T. dactyloides to AT-3 of non-reduction genes only, (2) increasing of ability for apomixis in hybrids and (3) decreasing of quantity of non-desirable linked wild characters resulting from "borrowing" of parthenogenesis from maize.
Thus, we have demonstrated the possibility of occurrence of autonomous reduced parthenogenesis, conditioned by nuclear genes, which produced (and produces now) unique material for further work on synthesis of apomictic forms. This work was supported in part by a grant from the Russian Foundation for Basic Research.
Return to the MNL 71 On-Line Index
Return to the Maize Genome Database Page