Proso millet, Panicum miliaceum L., has become a problem weed in corn-growing areas of Canada over the last 15 years. Several different biotypes of this weed exist, and one of the features by which they may be distinguished from one another is seed colour. Those plants with dark olive – bronze – black seeds that shatter readily are the most difficult to control. Part of their success as weeds is due to their ability to form a long-term viable seed bank in the soil, making quick elimination from an area impossible. Other biotypes closely resemble crop types and have little dormancy or ability to form a long-term seed bank. Between these two extremes are 'crown' and some 'golden' biotypes, some populations of which have weedy characteristics that may make them more aggressive weeds as the selection process continues. Some biotypes are more readily controlled than others by a combination of management practices, including rotation and spraying. Good chemical control is available for use in some broad-leaved crops (e.g., soybeans), but consistently effective control of this weed in corn has not yet been achieved. Thus, a knowledge of the ecology of the weed can assist in other forms of management. Each biotype that has been discovered in Canada is described and illustrated, together with an outline of ecological characteristics that may aid in its identification and control.

HurcHrNsoN, I., Cor.osr, J. aNo Lnwn, R. A. 1984

No abstract

Survival, dormancy, and germination of buried seeds of 30 populations of proso millet (Panicum miliaceum L.) were investigated. The effects of duration and depth of burial and type of soil were considered. The 30 populations belong to three agronomic groups: crop, croplike weed, and black-seeded weed. Black seeds exhibited much greater overwinter survival and dormancy than did seeds of the other two groups. Crop seeds had almost no survival through the winter. Only one croplike weed population exhibited appreciable (13–40%) survival after one winter in the soil. A combination of endogenous and enforced dormancy in black seeds effected an adaptive germination pattern of little (1%) germination at both the surface and 20 cm deep and much greater (40%) germination at 5 cm deep. Surviving crop and croplike weed seeds germinated irrespective of position in or on the soil. In a longer term experiment with only three populations both the croplike weed and black-seeded populations survived best in a well-drained soil. For the black-seeded population greater germination in the well-drained soil depleted that seed bank earlier than those in the medium-drained or poorly drained soils. The crop and croplike weed populations produced only transient seed banks in all three soil types, whereas some black seeds survived for 4 years.

Proso millet occurs both as a crop and a weed in North America. In 1970, an olive-black seeded biotype called ‘wild proso millet’ was found as an aggressive weed in row crops in Minnesota and Wisconsin and has since spread over a large area. We used Random Amplified Polymorphic DNA (RAPD) to assess genetic relationships among biotypes, measure genetic variation within wild proso millet across its range, and detect hybridization between wild proso millet and crop biotypes of proso millet. We found 97 RAPD genotypes among 398 individuals: 69 wild proso millet genotypes, 26 crop and crop-like weed genotypes, and two hybrid genotypes. Five RAPD markers consistently differentiated wild proso millet from crop cultivars and crop-like weeds. About 10% of the genotypes had at least one marker of the other type, suggesting possible hybridization between wild proso millet and crop biotypes. Most genotypes occurred in only one or two of the over 100 populations tested. The most widespread wild proso millet genotype occurred in 12 populations distributed in North Dakota, Minnesota, Illinois, and Wisconsin. More genetic variation exists among populations of wild proso millet than expected for a plant that presumably experienced a severe genetic bottleneck only 20 generations ago. Hypermutation rates and crossing between wild proso millet and crop cultivars could not account for the degree of genetic variation found in wild proso millet. The pattern of genetic variation among wild proso millet populations suggests multiple introductions of wild proso millet to North America.