Proso millet has become a major weed in North America during the past 20 yr. Experiments were conducted to assess the role of the soil seedbank in proso millet's success as a weed. Proso millet has at least eight distinct biotypes, all weeds, which exhibit a wide range of seedbank behavior. At one extreme, weed biotypes resembling crop varieties form transient seedbanks where all viable seed germinate or die before newly ripened seed enter the seedbank. In contrast, the black- and dark-red-seeded biotypes form persistent seedbanks that last several years in the soil. These latter seedbanks differ from those previously described in that there is always a fraction of the seedbank that is not capable of immediate germination. The black-seeded biotype of proso millet can establish a persistent seedbank quickly and that makes the weed very difficult to eradicate.

Seedbanks and seedling emergence of annual weeds were examined in arable fields at eight locations in the Corn Belt. Seed densities were estimated by direct seed extraction from each of several soil cores in each sampled plot. Average total seedbank densities ranged from 600 to 162 000 viable seed m-2 among locations. Coefficients of variation (CV) typically exceeded 50%. CV for seed densities of individual species usually exceeded 100%, indicating strongly aggregated distributions. CV were lower for species with dense seed populations than those with sparse seed populations. Variance of total seedbank densities was unstable when < 10 cores were examined per plot, but stabilized at all locations when ≥ 15 cores were analyzed, despite a 12-fold difference in plot size and 270-fold difference in seed density among locations. Percentage viable seed that emerged as seedlings in field plots ranged from < 1% for yellow rocket to 30% for giant foxtail. Redroot pigweed and common lambsquarters were the most frequently encountered species. Emergence percentages of these species were related inversely to rainfall or air temperatures in April or May, presumably because anoxia and/or high temperatures induced secondary dormancy in nondormant seed. From 50 to 90% of total seed in the seedbank were dead. This information can be employed by bioeconomic weed management models, which currently use coarse estimates of emergence percentages to customize recommendations for weed control.

A long-term integrated pest management study initiated in 1980 and continued through 1991 was conducted to determine interactions of tillage, crop rotation, and herbicide use levels on weed seed populations, weed populations, and crop yield. This paper presents giant foxtail seed population and stand along with corn yield in continuous corn, corn rotated with soybean, or corn following wheat in a soybean-wheat-corn rotation. Increasing herbicide use levels above the minimum reduced giant foxtail seed in the 0- to 2.5-cm depth of soil. Reducing tillage from conventional moldboard plowing to chiseling to no-tilling increased giant foxtail seed in only the top 0 to 2.5 cm of soil. No-tilling increased giant foxtail seed over conventional tillage in each year data were collected. Growing corn in a soybean-corn or soybean-wheat-corn rotation reduced giant foxtail seed from corn grown continuously in all three soil depths sampled: 0 to 2.5 cm, 2.5 to 10 cm, and 10 to 20 cm. Although stands of giant foxtail tended to follow soil weed seed counts, crop rotation significantly reduced giant foxtail stand with maximum reduction in the soybean-wheat-corn rotation in all tillage systems. Giant foxtail stands were reduced following wheat in no-tilling, probably because of the allelopathic influence of wheat straw. Corn yields showed weed management levels above minimum control are not justified regardless of tillage and crop rotation.

Changes in the weed seedbank due to crop production practices are an important determinant of subsequent weed problems. Research was conducted to evaluate effects of primary tillage (moldboard plowing and chisel plowing), secondary tillage (row cultivation), and herbicides on weed species changes in the soil seedbank in three irrigated row crop rotational sequences over a 3-yr period. The cropping sequences consisted of continuous corn for 3 yr, continuous pinto beans for 3 yr, or sugarbeets for 2 yr followed by corn in the third year. Cropping sequence was the most dominant factor influencing species composition in the seedbank. This was partly due to herbicide use in each cropping sequence producing a shift in the weed seedbank in favor of species less susceptible to applied herbicides. A comparison between moldboard and chisel plowing indicated that weed seed of predominant species were more prevalent near the soil surface after chisel plowing. The number of predominant annual weed seed over the 3-yr period increased more rapidly in the seedbank after chisel plowing compared to moldboard plowing unless effective weed control could be maintained to produce a decline in seedbank number. In this case, seedbank decline was generally more rapid after moldboard plowing. Row cultivation generally reduced seedbanks of most species compared to uncultivated plots in the pinto bean and sugarbeet sequences. A simple model was developed to validate the observation that rate of change in the weed seedbank is influenced by type of tillage and weed control effectiveness.

Seedbank studies often suffer from major methodological inadequacies such as absence of appropriate statistical data analysis and low sampling intensity. Multivariate analysis and computer mapping are innovative ways to treat seedbank data. Computer contour mapping was used to visualize spatial patterns of a population of common lambsquarters at three intervals during a growing season. At one site, high spring seed density of 600 000 seed m-2 was decreased to 18.3% of its original size by July, while at another site, low spring seedbank of common lambsquarters of 25 000 seed m-2 increased to 40 000 seed m-2 by autumn. Seedbank studies usually report results on total seed density or on densities of the most abundant species because of difficulties in analyzing large species matrices using parametric statistics. Multivariate analysis and specifically canonical discriminant analysis (CDA) are well suited for seedbank populations. The seedbanks of six agricultural habitats were demonstrated to be floristically different based on the analysis of the relative abundance of weed species in each site using CDA. Organic soils either under grassland or cultivated had significantly larger total seedbanks than mineral soils. If seedbanks are to be used in predictive population models, quantitative data that are reliable, rapidly obtained with limited resources, and logistically feasible for large sampling protocols are needed. Image analysis may be a potential rapid technique for weed seed recognition of washed soil samples.

The DNA sequence of a 196 base pair (bp) region of the acetolactate synthase (ALS) genes of three weed species, kochia, prickly lettuce, and Russian thistle, was determined. This region encompasses the coding sequence for Domain A, a region of the amino acid sequence previously demonstrated to play a pivotal role in conferring resistance to herbicides that inhibit ALS. The Domain A DNA sequence from a chlorsulfuron-resistant (R) prickly lettuce biotype from Idaho differed from that of a chlorsulfuron-susceptible (S) biotype by a single point mutation, which substituted a histidine for a proline. The Domain A DNA sequence from an R kochia biotype from Kansas also differed from that of an S biotype by a single point mutation in the same proline codon. This point mutation, however, conferred substitution of threonine for proline. Two different ALS-homologous sequences were isolated from an R biotype of Russian thistle. Neither sequence encoded amino acid substitutions in Domain A that differed from the consensus S sequence. The DNA sequence variation among the R and S kochia biotypes was used to characterize six Ada County, Idaho, kochia collections for correlation between phenotypic chlorsulfuron susceptibility and restriction digest patterns (RFLPs) of polymerase chain reaction amplification products. Most collections showed excellent correspondence between the RFLP patterns and the phenotypic response to chlorsulfuron application. However, one entirely R collection had the RFLP pattern of the S biotype, suggesting that resistance was not due to mutation in the proline codon.

Harold Alley, Professor of Merit at the University of Wyoming, died in the summer 1991. He was an extension weed scientist, but he also conducted extensive research in range weed control, especially revegetation of range areas with herbicide and cultural management practices in grazing situations. He was very active in the Western Weed Science Society and was a fellow of WSSA. He also received the WSSA Outstanding Extension Specialists Award. He is survived by his wife Jean and three children.