Community assembly is a branch of ecology that looks at how communities are assembled as they follow trajectories through time. A trajectory is controlled by biotic and abiotic constraints (filters) that act at multiple scales. From a total species pool, environmental and dispersal constraints control which species enter an ecological species pool. Within this pool, internal dynamics determine which of these species becomes part of the extant community. Environmental filters act by removing species that lack specific traits. Thus, traits are filtered, and with them, species. In this paper, we present the basic ecological theory of community assembly and address how it can be used in conjunction with a trait-based approach to understand and possibly predict how weed community structure changes in response to imposed filters such as tillage or crop rotation. Weed ecologists have struggled with the need to place our practical knowledge of agriculture and weeds into a broader theory, and there have been many calls to integrate ecology with agronomy and weed science. Community assembly might be one way to do so.

This paper is a review of and a commentary on the published addresses of WSSA presidents from the founding of the Society through 2000. The paper's assumptions are that the presidential remarks will reveal what the Society has emphasized, what its major concerns and goals have been, and how the presidents have addressed what they considered to be important issues. Seven issues have been addressed by many presidents. These include the importance of agricultural production and profit, the necessity of herbicides, weed science and the environment, the regulation of herbicides, the need for education, comparison of weed science and other plant protection disciplines, and the problem of herbicide resistance.

Field experiments were conducted from 1997 through 1999 to evaluate interspecific interference between pitted morningglory at 0, 10, 16, and 62 plants m−2 and drill-seeded, glyphosate-resistant soybean as influenced by soybean population and a single glyphosate application of 1.12 kg ai ha−1. Photosynthetic rate of soybean was not influenced by pitted morningglory density or glyphosate use 2 wk after treatment (WAT). Photosynthetic rate of soybean 12 WAT was reduced by 21 and 91% with 62 treated and untreated pitted morningglory plants m−2, respectively, whereas 10 treated and untreated pitted morningglory plants m−2 had no effect on the soybean photosynthetic rate. Pitted morningglory photosynthetic rate 2 and 12 WAT was reduced by 64 and 80%, respectively, when treated with glyphosate. The reduction in the photosynthetic rate of glyphosate-treated pitted morningglory was partially attributed to shading by soybean, whereas untreated plants were fully exposed to sunlight. Glyphosate-treated pitted morningglory at 10 and 16 plants m−2 did not reduce the rate of soybean leaf area index (LAI) accumulation; however, when the density was increased to 62 pitted morningglory plants m−2, soybean LAI decreased from 1.19 to 0.88 for each accumulated 100 growing degree days. Pitted morningglory produced a maximum of 24 million seeds ha−1 in the absence of glyphosate with 217,000 soybean plants ha−1. Pitted morningglory seed production declined with increasing soybean seeding rate in the absence of glyphosate, with a 41% reduction occurring when soybean population increased from 217,000 to 521,000 plants ha−1. Seed production of treated pitted morningglory ranged from 380,000 to 700,000 seeds ha−1. Soybean seed yield was not influenced by pitted morningglory density when treated with glyphosate. Untreated pitted morningglory at 10, 16, and 62 plants m−2 reduced soybean seed yield by 47, 62, and 81%, respectively. Competitiveness of untreated soybean increased with soybean seeding rate, resulting in 22% less yield loss with 521,000 than with 217,000 plants ha−1. Soybean seed yield was not reduced by 10 and 16 glyphosate-treated pitted morningglory plants m−2, but a 9% loss in yield occurred with 62 pitted morningglory plants m−2. Averaged over all pitted morningglory densities, glyphosate-treated pitted morningglory failed to reduce soybean seed yield at each of the three soybean densities. Following a single application of glyphosate, no apparent benefit was noted from increasing the soybean population above 217,000 plants ha−1.

Field experiments were conducted at Fayetteville, AR, in 1997, 1998, and 1999 to evaluate the degree of interspecific interference between drill-seeded, glyphosate-resistant soybean and hemp sesbania as influenced by soybean population, hemp sesbania density, and a single glyphosate application. Soybean was planted at 247,000, 430,000, and 618,000 seed ha−1 with early-season emergence of 217,000, 371,000, and 521,000 plants ha−1. Hemp sesbania densities were 0, 4, 10, and 16 plants m−2 in combination with 0 and 1.12 kg ai ha−1 glyphosate applied at the V4 to V6 soybean growth stage. Untreated hemp sesbania produced a maximum of 49 million seed ha−1 with 217,000 soybean ha−1, whereas seed production of glyphosate-treated hemp sesbania ranged from 2.8 to 0.6 million seed ha−1 with 217,000 and 521,000 soybean ha−1, respectively. Soybean seed yield was reduced 43% by 16 untreated hemp sesbania m−2, while glyphosate-treated hemp sesbania did not reduce seed yield. Averaged over all untreated hemp sesbania densities, soybean yield loss was reduced from 44 to 22% by increasing the soybean population from 217,000 to 521,000 plants ha−1. Because of the absence of soybean yield loss at 521,000 soybean ha−1, this density appears beneficial when using a single application of glyphosate to provide season-long weed control. However, a single glyphosate application in combination with a high soybean density did not completely prevent hemp sesbania seed production, and the seeding rate required to achieve a stand of 521,000 plants ha−1 would not be affordable; thus, alternate management methods targeting hemp sesbania must be employed to prevent an increase in the number of seed in the soil seedbank and to prevent a potential shift in the weed spectrum.

Velvetleaf growth and canopy architecture were compared under a range of light conditions representative of competitive and noncompetitive environments typical of irrigated Mediterranean-type agroecosystems. Velvetleaf biomass and seed production exceeded those reported in the literature. Plants grown in full light produced 1,370 g dry weight and 44,200 seeds per plant and showed low relative variability. Velvetleaf grown with corn was reduced to 21 g dry weight and 349 seeds per plant, and had high relative variability for biomass and seed numbers. Velvetleaf grown with kidney bean, intraspecific neighbors, or under shadecloth had dry weights and seed numbers that were intermediate to plants grown in full light or with corn. Relative growth rate (RGR), net assimilation rate (NAR), and leaf area ratio (LAR) were assessed utilizing Richards functions, which were fitted to the primary biomass and leaf area data by weighted regression. RGR was highest for all plants early in the season, but declined later. Dynamics of NAR and LAR appeared to be correlated with increased self-shading, shading by neighbors, leaf age, and shedding of lower canopy leaves. Dynamics of specific leaf area corresponded with light availability such that the leaves exposed to full light were thicker than those exposed to shade. The branches of plants in all treatments had random azimuths and the foliage area density was concentrated along the perimeter of the plant's canopy. Velvetleaf increased the canopy radius through extensive branching when exposed to full sunlight. Leaf area distribution and branching patterns resulted in leaf area indices of less than 1.0. Leaves maintained a perpendicular angle to the sun throughout the day, but this depended on whether leaves received a consistent directional signal from the sun and not necessarily on whether they received a high-intensity signal. When shaded, the allocation of dry matter went primarily to the stem tissue, which increased the height rather than the girth of the plants. There was a 10- to 20-d delay for allocations to seed in the case of shaded plants relative to those grown in full sunlight. In brief, velvetleaf had a wide range of growth and canopy responses to a variety of light availabilities and it should have little difficulty in becoming fully established in the irrigated agroecosystems of Mediterranean-type regions.

Knowing the distribution of weed seedlings in farmer-managed fields could help researchers develop reliable distribution maps for site-specific weed management. With a knowledge of the spatial arrangement of a weed population, cost effective sampling programs and management strategies can be designed, so inputs can be selected and applied to specific field areas where management is warranted. In 1997 and 1998, weeds were sampled at 612 to 682 sites in two center pivot irrigated corn fields (71 and 53 ha) in eastern Colorado. Weeds were enumerated when corn reached the two-leaf, four-leaf, and physiological maturity stages in a 76.2- by 76.2-m grid, a random-directed grid where sites were established at intervals of 76.2 m, and a star configuration based on a 7.62- by 7.62-m grid within three 23,225 m2 areas. Directional correlograms were calculated for 0, 30, 60, 90, 120, and 150° from the crop row. Fifteen weed species were observed across fields. Spatial dependence occurred in 7 of the 93 samples (a collection of sampling units for a particular weed species that was detected within a field at a particular sampling time and year) for populations of field sandbur, pigweed species, nightshade species, and common lambsquarters. Correlogram analysis indicated that 18 to 72% of the variation in sample density was a result of spatial dependence over a geographic distance not exceeding 5 to 363 m among the examined data. Because of the lack of spatial correlation for weed seedling distributions in these eastern Colorado corn fields, interpolated density maps should be based on grid sizes (separation distances) less than 7.62 m for weed seedling infestations.

Sprouting percentage was estimated for purple nutsedge tubers in the field from daily fluctuating soil temperatures. Tuber sprouting under alternating temperatures ranging from 20 to 45 C for 14 d responded quadratically to alternations of high and low temperature. A response surface regression of the cumulative sprouting percentage accounted for 88% of the variation. The cumulative sprouting percentage curves were sigmoidal, and the Richards function satisfactorily regressed the characteristics of the curves. A simulation model was developed for the cumulative sprouting percentage by estimating sprouting from daily high and low temperatures and accumulating daily increments of tuber sprouting. Five weeks of soil solarization with clear polyethylene film at Waimanalo, Hawaii raised the mean soil temperature at 15-cm depth by 5.8 C in spring and by 7.2 C in summer. Solarization also increased the mean daily temperature difference from 1.5 to 3.7 C in spring and from 2.3 to 3.8 C in summer. Solarization increased the final sprouting percentage in the field from 74 to 97% in spring and from 97 to 100% in summer. The simulation model estimated the final field sprouting of tubers within 95% confidence intervals of the observed means.

Understanding weed–crop interactions is critical in predicting crop yield loss, but it is also important to understand how these interactions affect weed productivity. Therefore, research was conducted to characterize the weed relative leaf area and weed relative volume of several giant foxtail cohorts in soybean, and to assess weed density and cohort emergence time, weed relative leaf area, and weed relative volume as predictors of giant foxtail shoot biomass and fecundity. Giant foxtail cohorts emerged at VE (emergence), VC (cotyledon), V1 (first node), and V3 (third node) soybean growth stages and were thinned to densities of 0, 4, 16, 36, and 64 plants m−2. Based on weed density and cohort emergence time, the maximum shoot biomass per square meter or the maximum fecundity per square meter differed between years. In contrast, shoot biomass or fecundity per plant, as weed density approached zero, and the rate at which shoot biomass or fecundity decreased exponentially, as time increased, were similar between years. Based on the weed relative leaf area, the cohort effect on giant foxtail shoot biomass differed between years, whereas the cohort effect on giant foxtail fecundity was similar between years. Maximum giant foxtail shoot biomass per square meter or fecundity per square meter differed between years when estimated from weed relative leaf area. Based on the weed relative volume, the cohort effect on giant foxtail shoot biomass per square meter or fecundity per square meter was similar between years, as was the maximum giant foxtail shoot biomass per square meter or fecundity per square meter. The temporal stability of weed relative volume, used to describe giant foxtail shoot biomass or fecundity, may aid in improving bioeconomic weed management models.

A new strain of the fungus Plectosporium tabacinum was isolated from naturally-infected false cleavers plants and evaluated as a bioherbicide for the control of both herbicide-resistant and herbicide-susceptible false cleavers in canola. Plectosporium tabacinum was nonpathogenic to both Argentine canola and Polish canola and to both herbicide-tolerant and conventional cultivars. Plectosporium tabacinum killed false cleavers seedlings when applied at a concentration of 1 × 107 conidia ml−1 and provided with 16 h dew. The efficacy of P. tabacinum on herbicide-resistant false cleavers was identical to that on herbicide-susceptible false cleavers. Catchweed bedstraw was also highly susceptible to P. tabacinum. Further host-range tests on 34 plant species in 26 genera and 12 families demonstrated that P. tabacinum is sufficiently host specific for false cleavers control in western Canada. This fungus may provide a novel approach for managing herbicide-resistant false cleavers.

Annual bluegrass control was reduced following 7 yr of continuous fall application of dinitroaniline (DNA) herbicides. Annual bluegrass control was < 40% on two fairways in year eight following prodiamine applied at 1.1 kg ai ha−1. In dose–response studies conducted in growth chambers, this annual bluegrass population exhibited 105-fold resistance in shoot growth to prodiamine compared with a known susceptible population. A 6.4-fold resistance to prodiamine was found when comparing annual bluegrass root growth to the known susceptible biotype. Spring-applied oxadiazon did not affect shoot or root growth between annual bluegrass biotypes. Equivalent levels of control were attained with pronamide. The presence of DNA-resistant annual bluegrass, in addition to previously confirmed triazine-resistant biotypes on North Carolina golf courses, indicates a need for resistance management strategies to be integrated into golf turf management practices.

Field studies were conducted at four locations in North Carolina in 1998 and 1999 to evaluate a computer program, Herbicide Application Decision Support System (HADSS™), for weed management in peanut (Arachis hypogaea). Weed management systems included metolachlor or ethalfluralin preplant-incorporated (PPI) used alone or in combination with diclosulam preemergence (PRE) or flumioxazin PRE. These herbicide combinations were used alone, followed by (fb) postemergence (POST) herbicides recommended by HADSS™ or fb a standard POST program of paraquat plus bentazon early postemergence (EPOST) fb acifluorfen plus bentazon POST. The standard POST herbicide system and HADSS™ POST recommendations were also used without soil-applied herbicides. Ethalfluralin PPI alone controlled large crabgrass (Digitaria sanguinalis) better than metolachlor PPI. Combinations of metolachlor or ethalfluralin PPI with either diclosulam or flumioxazin PRE provided equivalent control of all weeds evaluated except yellow nutsedge (Cyperus esculentus). The addition of diclosulam or flumioxazin PRE to systems containing metolachlor or ethalfluralin PPI always improved control of ivyleaf morningglory (Ipomoea hederacea) and yellow nutsedge and improved yield and net returns in 15 of 16 comparisons where no POST herbicides were used. For systems that used diclosulam or flumioxazin PRE, the HADSS™ POST and standard POST herbicide systems improved yield in 4 of 12 and 2 of 12 comparisons, respectively, compared with similar systems that did not use diclosulam or flumioxazin. However, in systems using either HADSS™ POST or the standard POST system, yield was always improved when compared with metolachlor or ethalfluralin PPI alone. HADSS™ POST provided equal or higher weed control, peanut yield, and net returns when compared with the standard POST herbicide system.

Crop rotations, particularly those that include legumes, often result in improved soil quality and crop yields. A study was conducted to confirm the presence and persistence of the residual effects of crop rotation and weed management on a test crop and weeds in three tillage systems (moldboard plow [MP]; chisel plow [CP]; no tillage [NT]). Rotation (spring barley monoculture and spring barley–red clover rotation) and weed management (intensive, moderate, minimum) treatments, initiated in 1987, were terminated, and a test crop of spring wheat was grown in 1995 and 1996. Tillage treatments were maintained throughout. Multivariate analysis showed that weed communities were more affected by treatment termination in the rotation NT treatment with minimum weed management than in all other treatments. The former treatment was dominated by perennial broadleaf weeds but sustained adequate wheat yields (3.3 Mg ha−1) compared with the monoculture (1.0 Mg ha−1) one year after termination. Weed communities in CP and MP plots were less affected by treatment termination. Yet, changes in herbicide use at termination caused the virtual elimination of quackgrass from tilled plots and allowed field pennycress to become ubiquitous across treatments. Residual effects from crop rotation were more important than those from weed management in increasing wheat yields in tilled systems. Differences in wheat yield in NT systems 2 yr after treatment termination were attributed to residual effects from previous weed management rather than from crop rotation. Beneficial effects of crop rotation and weed management may persist for 2 yr but will vary according to tillage system.

The effects of several crop rotations and herbicide programs on populations of goosegrass, stinkgrass, large crabgrass, smooth crabgrass, fall panicum, and yellow nutsedge were investigated at two sites from 1991 to 1994. Crop rotations were continuous corn, continuous soybean, corn–soybean, and corn–tomato–soybean. Herbicide programs were the split-plots and included continuous use of acetolactate synthase (ALS)-inhibitor herbicides, continuous use of non–ALS-inhibitor herbicides, annual rotations between ALS- and non–ALS-inhibitor herbicides, combinations of ALS- and non–ALS-inhibitor herbicides in the same year, and no herbicide. Grass and yellow nutsedge densities generally were affected by an interaction between crop rotations and herbicide programs by 1994. Goosegrass densities in 1994 were highest from the continuous use of ALS-inhibitor herbicides in the corn–tomato–soybean rotation and were generally high in 1994 in the continuous corn and corn–tomato–soybean rotations. Stinkgrass densities were highest by 1994 where imazethapyr was applied alone for four consecutive years. Stinkgrass densities were also high where imazethapyr was applied in combination with butylate for 4 yr and at Site 2 (designated Northampton) where imazaquin plus nicosulfuron was applied for 4 yr. Herbicide programs did not produce shifts in large crabgrass densities, except that densities were highest where butylate plus atrazine was applied for 4 yr. Smooth crabgrass was present in significant densities at Site 1 (designated Accomac) only where imazaquin plus nicosulfuron was used for 4 yr or at Northampton from continuous ALS-inhibitor programs. Fall panicum densities were highest by 1994 where the combination of butylate plus atrazine was applied continuously for 4 yr. Yellow nutsedge control was lowest and densities were highest at Northampton where the combination of fomesafen plus fluazifop-P plus fenoxaprop was applied continuously for 4 yr. Yellow nutsedge densities by 1994 at Northampton were also high where these herbicides were applied with imazaquin for 4 yr or where these herbicides were applied in rotation with imazaquin plus nicosulfuron or butylate plus atrazine.

Managing weed infestations in a spatially precise manner requires accurate and cost-effective weed identification techniques. The goal of our research was to quantify the accuracy of continuous weed presence–absence maps and assess how management based on those maps may affect producer net returns. Each continuous sampled map covered the entire field and contained vector polygons labeled as either wild oat presence or wild oat absence. The accuracy of the continuous wild oat maps at each sampling time was determined from georeferenced quadrats of wild oat densities. The accuracy of the continuous wild oat seedling maps ranged from 48.3 to 87.1% among the six site-years. The accuracy of the wild oat seedling maps improved by at least 8% when a 10-m buffer was included around areas mapped as wild oat presence. The accuracy of continuous wild oat panicle maps from the combine at harvest ranged from 65.8 to 90.9% among the six site-years. The variation in accuracy for the wild oat seedling maps among sites was greater than the accuracy of the panicle maps. Net returns ($ ha−1) for four site-years were calculated and compared for four possible weed management approaches on each field. A site-specific herbicide application to areas mapped as wild oat presence always generated higher net returns than a herbicide application over the entire field for four sites. A site-specific herbicide application to areas mapped as wild oat presence plus a surrounding 10-m buffer area only resulted in a higher net return in one of the 12 site-years compared with a site-specific herbicide application without the 10-m buffer. This site had the lowest (48.3%) wild oat seedling map accuracy, and uncontrolled wild oat had a high-yield effect. This research indicates that using a continuous weed sampling method based on presence or absence for site-specific herbicide application can be profitable over a herbicide application to the entire field, even with the associated technology cost and seedling map errors.