Aim. To evaluate the effi ciency of inducing generative, reduced parthenogenesis and to better use the differentiating potential of the embryo culture under apomictic seed production in selection materials of sugar beet with cytoplasmic male sterility (CMS), and B) to isolate homozygous lines (dihaploids) without the use of polyploidizing substances. Methods. Apomictic (agamosper- mous) seed production in apocarpous pollen sterile lines from B. vulgaris subsp. vulgaris var. altissima (sugar beet) using classi- cal so-called Owen sterile cytoplasm and sterile cytoplasm from Beta maritimа and Beta patula as sources, was conducted under pollen free conditions and spatial isolation in the greenhouse breeding complex of the Yaltushkivska experimental breeding station (Yaltushki, Ukraine). The specifi cities of embryonic development of apomictic embryos were studied with the purpose of effi cient regulation of the induction of explants in vitro as donors of the culture of immature embryos. Fluorescent fl ow cytophotometry in combination with the computer program of the Partec Ploidy Analyser PA-2 (Partec GmbH, Germany, now Sysmex), were used to determine the degree of ploidy, enabling the selection of haploid and dihaploid lines in vitro. A genetic method was developed using the expression of morphological marker indices of nuclear genes of anthocyanin coloring (R+ r–) of regenerant plants in vitro and ploidy determination for differentiation by generative (reduced) parthenogenesis. The sampling technique that took into account the hormonal composition of cultural media and the level of genome ploidy, sample frequency and statistical analysis of the results was determined using the appropriate statistics; the percentage of regenerants, induced by different types of morpho- genesis and ploidy in vitro, was determined along with the measurement error to control the accuracy of the selected sampling (number of seed embryos). Results. The selected cultural medium No. 3, based on the basal medium according to Gamberg et al., 1968 (21), contained 6 BAP – 2 mg/l, 2.4 D – 0.5 mg/l, gibberellic acid – 0.1 mg/l, which ensured a success rate of 4.4 to 23.3 % of direct regeneration of shoots from the embryo culture, depending on the genotype of donors, and 4–10 % for induction and proliferation of callus. In ten experimental numbers of alloplasmic lines of sugar beet, the incidence of haploids and mixoploids among the regenerants from the embryo culture fl uctuated within the wide range of 14.8 – 62.2 % and exceeded the indices, ob- tained by other known methods of haploid parthenogamy, which had the values of 3.79 – 6.25 %. Conclusions. The homozygous lines and dihaploids were determined and set apart/stabilized in the process of micropropagation, where the differentiation of clones was made on the basis of total DNA content in interphase nuclei, using information of histograms generated in fl uorescent fl ow cytometry with the Partec Ploidy Analyser PA-II instrumentation. The medium, based on macro- and microsalts according to Gamberg et al., 1968 (21) was found to be the most effi cient; it ensured at least partially successful direct regeneration in the culture of embryos within the range of 4.40 ± 1.29 to 23.3 ± 3.45 %. The success of direct regeneration of apomictic material depended on the composition of the cultural medium used fi rst and foremost, and to a lesser extent on the stages of embryogenesis from day 12 till day 32, differentiated by the fi xation period for seed embryos starting from the beginning of fl owering. Homozygous lines were created without polyploid-inducing substances due to spontaneous transfer of some cells of haploid regenerant plants to a higher level of ploidy, that can be used in the breeding of sugar beet. Genetic determination of apomictic seed reproduction in alloplasmic lines and pollen free lines of sugar beet and the technologies of inducing dihaploids allow reducing the period of inzucht-crossing considerably to obtain homozygous lines, creating unique material for chromosome engineering and marker-oriented selection with target combinations of genes in homozygous state.
