Improvement of Andropogon scoparius Michx. by breeding and selection

Andropogon scoparius in nature is an extremely variable species, divided into rather definite habitat types (the ecotypes of Turesson), which, however, exhibit considerable variability within themselves. In general, northern types are earlier, less leafy, and smaller than types from farther south. Leaf area is the best single measure of quality and yield of forage. Wide variations in total leaf areas exist that are due to genetic differences in the plant. This is shown by the fact that the progeny of individual plants vary significantly from the progeny of other plants. The variability between these groups is greater than the variability within them. This indicates that selection, even in open-pollinated populations, tends to increase uniformity. No satisfactory measure of quality has yet been devised. Grazing tests are probably the best measure and shall be applied before final selection of strains is made. Basal diameter of the plants varies widely, yet there is a tendency toward uniformity within groups of progeny of individual plants. Plant height has not been an important factor in selection of superior types of plants, yet studies indicate it to be definitely influenced by genetic make-up. There is a marked uniformity within the progeny of selected plants. It is indicated that time of maturity can be changed by selection. This will make it possible to produce a strain that heads later, hence gives a longer summer grazing season. It is important that time of maturity be as late as possible, yet not so late that plants are frosted before producing seed. Seed set is reduced by selfing in Andropogon scoparius. It is not definitely known whether this reduction in seed set is due to genetic causes or to abnormal conditions within the selfing bags. There is no evidence to indicate that selfing has any effect on germination for selfed seed has been found to have as high germination percentages as that from open-pollinated plants. Differences have been observed in rate of germination. It is important that grasses should germinate quickly in order that the seedling plant become established before summer droughts. Seedlings that become established quickly are better able to compete with weeds and other grasses. Rather definite relationships have been found to exist between certain characters in Andropogon scoparius, and that these relationships follow definite trends is indicated as follows: (a) Basal diameter is correlated positively with number of culms, with a tendency for the correlation to be higher in younger plants. (b) Basal diameter is also correlated positively with plant height but no trends can be observed. Since these characters are expression of general vigor their high positive correlation is in accordance with expectations. (c) Basal diameter and leaf area are positively correlated, the highest correlation existing in young plants. (d) Height of plant and leaf area, while positively correlated, exhibit the highest correlation in older plants. Evidence indicates, however, that the low correlation in the third generation may be due to the effect of selection rather than to age of plants. (e) In general, late plants tend to be leafier than early plants. After two generations of selection, however, no correlation exists between these two characters. The limited amount of evidence on the effect of self-fertilization indicates that the vigor of Andropogon scoparius is not seriously, if at all, affected by inbreeding.