| The diversity of the wild Avena species holds promise for oat improvement if interspecific introgression is feasible. The production of octoploid oats could enhance intergenomic intra-allelic interactions buffering the phenotype over a wide range of environments. Interspecific 2x/6x and 4x/6x hybrids were treated with colchicine producing octoploids and decaploids. Agronomic and cytological behavior of A(,1), A(,2), and A(,3) decaploid and octoploid plants were studied. Replicated yield trials were conducted to compare octoploids derived from diploid and tetraploid introgression with hexaploid cultivars. Mass selection for seed size and density was conducted to assess the use of a rapid screening technique for decaploid population improvement.; Significant differences between decaploid families within a cross and between families among crosses were observed for height, biological yield, seed weight, and harvest index in 1983 and height, heading date, and harvest index in 1984. Means of the decaploid plants ranged from 2.2-20.4 grams per plant for biological yield and 0-3.9 grans for seed weight. Mean harvest index ranged from .01-.32. Octoploid means varied from 12.5-60.8 grams per plant for biological yield and 1.7-17.7 grams for seed weight. Harvest index ranged from .12-.35. Cytological studies of root tip cells from decaploid and octoploid progeny indicate somatic instability. Meiotic analysis suggest pairing is primarily as bivalents, but 1-3 trivalents, 1-3 quadrivalents, and up to seven univalents were noted at metaphase I. Lagging chromosomes and micronuclei are common in microsporocytes. Replicated yield trials indicated that 2x/6x octoploids yielded significantly more grain than 4x/6x derived octoploids, but yielded less than 6x checks. The octoploids, A. longiglumis x A. sativa and A. canariensis x A. sativa, yielded 12% and 25% below the mean of the hexaploids. A. longiglumis x A. sativa yielded significantly higher than Orbit in 1983, and Astro and Sun II in 1984. Highly significant G, E, and G x E interactions were observed over years. Mass selection for seed size and density indicated significant differences between treatments for individual plant yields and chromosome number. This selection technique may be useful for population improvement. |
