Greenhouse and growth chamber experiments were conducted to study glyphosate efficacy, rainfastness, absorption, and translocation in redvine. Glyphosate at 0, 0.56, 1.12, 2.24, and 4.48 kg ai/ha was applied to redvine plants raised from rootstocks at the five- to seven-leaf stage (about 25 cm tall). Redvine control ranged from 55% at 0.56 kg/ha glyphosate to 98% at 4.48 kg/ha. Glyphosate at rates above 1.12 kg/ha, greatly reduced regrowth from rootstocks of treated plants. A simulated rainfall of 2.5 cm (7.5 cm/h intensity) within 24 h of glyphosate application reduced efficacy by 23% compared with no simulated rainfall. Absorption of 14C-glyphosate in redvine increased from 1.8 to 21.9%, and translocation increased from 0.1 to 8.1% from 6 to 192 h after application, respectively. Translocation was both acropetal and basipetal, and by 96 h of exposure, the 14C radioactivity was widely distributed throughout the plant. Absorption and translocation was greatly affected by posttreatment temperature. Absorption and translocation were highest (34.9 and 10.6%, respectively) in plants maintained at 35/30 C (14/10 h, day/night), followed by 15/10 C (21.2 and 4.9%, respectively), and was lowest (7.8 and 1.6%, respectively) in plants maintained at 25/20 C. Results suggest that longer periods of leaf exposure to the herbicide and high temperatures could increase glyphosate absorption, translocation to redvine rootstocks, and subsequent control. These data also suggest that effective control of redvine in the field will require glyphosate rates higher than those recommended for use in glyphosate-resistant crops.

Three experiments were conducted in the field using common cocklebur as a test species to identify weed growth differences due to propagation method and weed age when transplanted (with no crop and with soybean). Two experiments involved weed establishment in noncrop areas using four propagation methods (direct seeded–transplanted, peat tablets, peat pots, and plastic inserts) in a factorial arrangement with three transplant timings (at the cotyledon stage, at two true leaves, and at four true leaves). A direct-seeded method was also included as a check. Propagation methods and transplant timings were contrasted to the direct-seeded method because it was the “most natural” method of those tested. Based on results from the first year, the peat tablet method at the three transplant timings plus the direct-seeded method were used to establish the weed in a field experiment with soybean. In 1996 at 4 wk after emergence (WAE), 8 WAE, and senescence, weeds planted in peat tablets and transplanted at the cotyledon stage were most similar to the direct-seeded method. Peat pots and plastic inserts displayed an intermediate number of differences, whereas direct-seeded–transplanted weeds exhibited the most differences of any propagation method. Transplanting at the two-leaf stage showed more differences, with the four-leaf stage showing even more. In 1997 at 4 WAE, few propagation methods or transplant timings were different from the direct-seeded method. Results from harvests at 8 WAE and at senescence were comparable to those in the previous year. Differences between years at 4 WAE were attributed to greater heat stress and less moisture early in 1996. When common cocklebur was grown in soybean, no differences were detected at 4 WAE between peat tablets transplanted at the cotyledon stage vs. direct-seeded weeds and only one difference when transplanted at the two- and four-leaf stages. At 8 WAE, the early transplant was again the most comparable treatment to direct-seeded weeds, with many differences at the later transplant timings. At senescence, weed differences were no longer apparent. Patterns established for peat tablets with the weed growing alone largely held true when it was grown with soybean.

A field study evaluated the effects of green kyllinga establishment method (seed vs. stolon), two mowing heights (2.5 and 5.0 cm), and three nitrogen (N) rates (0, 24, and 49 kg/ha/mo) on green kyllinga infestation in ‘Tifway’ bermudagrass turf. The study was initiated in vigorous and newly established or “weak” bermudagrass turf in May 1997 and continued until December 1998. The green kyllinga area was measured periodically each year and plant dry weight (g/500 cm2) was calculated in December 1997 and 1998. In 1997, stolon established green kyllinga plots were twice as large as seeded plots in vigorous turf and four times larger in weak turf. Method of establishment, however, was less important in 1998 as seedling populations became more established. In weak turf, increasing N rate to 49 kg/ha/mo decreased green kyllinga spread by 50% in 1997 and by 40% in 1998 compared to no N. In vigorous turf, mowing height influenced green kyllinga infestation more than N. Low mowing height (2.5 cm) increased green kyllinga infestations nearly twofold in vigorous turf in 1997 and more than fivefold in 1998. Golf course fairways are often maintained at clipping heights shorter than 2.5 cm, and green kyllinga is a prevalent weed at these sites. Green kyllinga may gain a competitive advantage in bermudagrass turf at lower mowing heights.

Absorption and translocation of nicosulfuron and dicamba applied alone and combined was studied in 25-cm-high hemp dogbane originating from lateral roots that had overwintered. Absorption of 14C-nicosulfuron by hemp dogbane was not affected by applications of 70 g ai/ha unlabeled dicamba. Upward translocation of 14C-nicosulfuron in hemp dogbane was 86% greater 6 d after treatment (DAT) when dicamba was tank mixed with nicosulfuron, compared to nicosulfuron applied alone. Combinations of nicosulfuron plus dicamba increased translocation of nicosulfuron to hemp dogbane crown and roots 237 and 130%, 1 and 6 DAT, respectively, compared to nicosulfuron applied alone. Absorption of 14C-dicamba by hemp dogbane was not affected by the addition of 31 g ai/ha nicosulfuron. Upward translocation of dicamba in hemp dogbane was 42% greater 6 DAT for nicosulfuron plus dicamba compared to dicamba applied alone. Increased hemp dogbane control with nicosulfuron plus dicamba compared to either herbicide applied alone can be explained by enhanced translocation of both herbicides.

Glyphosate is often mixed with residual herbicides to control emerged weeds in no-till crop production systems. Field studies were conducted in Ohio from 1992 to 1994 to evaluate the weed control provided by residual herbicides and reduced rates of glyphosate in full-season, no-till soybean. Herbicide treatments were applied at two timings to examine the effect of weed size. At 4 wk after treatment, greater than 85% Pennsylvania smartweed control was obtained with metribuzin plus chlorimuron and linuron plus chlorimuron applied with 280 g ai/ha glyphosate and with imazethapyr and imazaquin applied with 560 g/ha glyphosate. All residual herbicides provided at least 85% common lambsquarters control when applied with 560 g/ha glyphosate. In 1992, the early application of residual herbicides provided this level of common lambsquarters control without glyphosate due to the small weed size at the time of application. Residual herbicides applied with 280 g/ha glyphosate controlled giant foxtail 85% or greater. Higher glyphosate rates were needed to control barnyardgrass. The performance of reduced glyphosate rates was dependent on weed species and weed size.

Glyphosate was applied to pepper as single or sequential applications to assess the consequences of drift or other accidental exposures. Foliar injury increased and plant vigor declined with increased rates of glyphosate and were exacerbated by a second application. Single applications at flowering (stage 1) were more damaging than single applications after fruit set (stage 2). Decline in marketable yield with increased glyphosate rate was greater with stage 1 applications, except in spring 1987 when marketable yields with single applications of glyphosate at stage 1 or stage 2 were statistically similar. Sequential applications resulted in the lowest marketable yields. Total yields declined with increased glyphosate rate and decline was more pronounced with sequential applications than with single applications. Total yield was affected to a greater extent when glyphosate was applied at stage 1 than at stage 2. Yield was more sensitive to two successive exposures than to single applications. Mean fruit weight was reduced by glyphosate in two of the four experiments.

Field experiments were conducted in 1997 and 1998 near Columbia and Novelty, MO, and at Urbana, IL, to evaluate corn injury, weed control, corn yield, and estimated economic returns with weed management programs in glufosinate-resistant corn. Herbicide programs included acetochlor preemergence (PRE) followed by glufosinate alone or with atrazine postemergence (POST) and total POST programs consisting of single and sequential applications of glufosinate alone or tank mixed with acetochlor, atrazine, or acetochlor plus atrazine. Metolachlor PRE followed by dicamba plus atrazine early POST (EPOST) and metolachlor plus atrazine PRE were included for comparison. In the total POST treatments, mid-POST applications controlled shattercane and common cocklebur better than EPOST applications. However, yield reductions as high as 23% occurred because of early-season weed interference, although weeds were controlled later in the season. Applying atrazine with glufosinate generally increased control of giant foxtail, common cocklebur, morningglory species, and common waterhemp compared to glufosinate alone, but did not increase control of common lambsquarters, velvetleaf, or Pennsylvania smartweed. Corn yield was positively correlated with weed control (r = 0.88) and more strongly dependent on grass (r = 0.82) than broadleaf (r = 0.70) weed control. Net incomes were positively correlated to corn yield (r = 0.73). Four of the top six net income-producing treatments included two herbicide applications. Three of the treatments were PRE followed by POST programs, and the fourth was a sequential POST treatment of glufosinate.

Field experiments were conducted in southern Wisconsin from 1996 to 1998 to evaluate the effects of acetochlor applied preemergence (PRE) followed by nicosulfuron or sethoxydim applied postemergence (POST) and of cultivation on woolly cupgrass control, corn grain yield, and woolly cupgrass seed production. Sethoxydim treatments provided greater woolly cupgrass control than nicosulfuron treatments in 1997 and similar control to nicosulfuron treatments in 1996 and 1998. However, neither herbicide killed all emerged woolly cupgrass plants, regardless of herbicide rate or POST application timing. Late postemergence treatments provided greater season-long control than early postemergence treatments in 1996 and 1998, but there was no difference in control between timings in 1997. Acetochlor applied PRE at 1,800 g ai/ha (1×) followed by nicosulfuron (35 g ai/ha) or sethoxydim (213 g ai/ha) at the registered rate (1×) or at one half of the registered rate (0.5×) consistently provided sufficient woolly cupgrass control to maximize corn yield, regardless of cultivation or POST herbicide application timing. Acetochlor PRE treatments, cultivated and noncultivated, followed by sethoxydim POST and cultivated acetochlor PRE treatments followed by nicosulfuron POST treatments provided sufficient control in each year to limit woolly cupgrass seed production to a level that decreased the predicted future seedbank density, regardless of herbicide rate or POST herbicide application timing. However, noncultivated acetochlor treatments followed by nicosulfuron applied at 1× followed by 0.5× or at 0.5× followed by 0.5× did not consistently reduce seed production to a level that decreased the predicted future seedbank density. These results suggest that the rate of sethoxydim can be reduced to 0.5× if a full rate of acetochlor is applied PRE with little effect on corn yield or woolly cupgrass seedbank density. However, reducing the rate of nicosulfuron to 0.5× following an application of acetochlor at the full rate or reducing both the acetochlor and nicosulfuron or sethoxydim rate to 0.5× is recommended only if a cultivation is planned.

Eurasian watermilfoil, an aquatic invasive weed, has been identified recently at a number of sites in western Nevada and northeastern California, including Lake Tahoe. Because Eurasian watermilfoil is easily spread by fragments, transport on boats and boating equipment plays a key role in contaminating new water bodies. This is an important means of the potential spread of this weed throughout key recreational and agricultural areas surrounding Lake Tahoe. Unless the weed is controlled, significant alterations of aquatic ecosystems, with associated degradation of natural resources and economic damages to human uses of those resources, may occur. This research uses the economic valuation approach known as benefits transfer to estimate the value of a portion of the recreational service flows that society currently enjoys in the Truckee River watershed below Lake Tahoe. The lower-bound estimates of baseline water-based recreation value at a subset of sites in the watershed range from $30 to $45 million/yr. Impacts from the continued spread of Eurasian watermilfoil in the watershed could be significant; for example, even a 1% decrease in recreation values would correspond to roughly $500,000/yr as a lower bound.

Studies were conducted from 1995 to 1997 near Tifton, GA, to evaluate shallow tillage implements and tillage frequency for stale seedbed weed control in peanut. Tillage implements evaluated were a power tiller, disk harrow, field conditioner, and sweep cultivator. Plots for each implement were tilled once or twice prior to planting peanut. Results from midseason weed counts and peanut yield showed similar responses among implements and frequency of tillage. These results indicate no advantage of any shallow tillage implement for stale seedbed weed control, although peanut yields were generally greater in plots with tilled stale seedbeds than in the nontilled control. The implement of choice should be based on cost of operation and compatibility with the overall peanut production system.

Greenhouse experiments were conducted in Greece to evaluate the level of propanil resistance in three barnyardgrass biotypes collected from rice fields where propanil had been used for around 25 yr. Suspected resistant biotypes were not controlled by propanil at 2.6, 5.2, 10.4, and 20.8 kg ai/ha, with the highest rate five times greater than the recommended rate. Control was reduced even more when propanil application was delayed until the three- to five-leaf stage. The susceptible barnyardgrass biotype collected from a nontreated propanil area was controlled with 2.6 kg/ha propanil. All barnyardgrass biotypes were effectively controlled by quinclorac (0.75 kg ai/ha) applied either alone at the two- to three-leaf stage or in mixture with propanil (3.5 kg/ha) at the three- to five-leaf stage. Only two of the resistant barnyardgrass biotypes were effectively controlled by azimsulfuron (0.02 kg ai/ha) applied alone at the two- to three-leaf stage, but all biotypes were controlled with the addition of propanil (3.5 kg/ha) at the three- to five-leaf stage. The resistant barnyardgrass biotype, which required the highest rate of propanil to reduce biomass by 50%, also tended to exhibit the most growth potential 7 to 22 d after emergence when compared with the other resistant biotypes or the susceptible biotype. Results clearly show that varying levels of barnyardgrass resistance to propanil exist but that effective control alternatives are available.

Sicklepod (Senna obtusifolia), hemp sesbania (Sesbania exaltata), and pitted morningglory (Ipomoea lacunosa) seed production and viability in early-maturing soybean (Glycine max) were evaluated following preharvest desiccation. The herbicide combinations 1.1 kg ai/ha glyphosate plus 6.7 kg ai/ha sodium chlorate, 2.2 kg/ha glyphosate plus 3.4 kg/ha sodium chlorate, and 0.3 kg ai/ha paraquat plus 6.7 kg/ha sodium chlorate reduced sicklepod germination, emergence, and seedling growth in most cases. Oxyfluorfen at 0.3 kg ai/ha plus 3.4 or 6.7 kg/ha sodium chlorate also reduced sicklepod germination and growth. Glufosinate from 0.8 to 1.4 kg ai/ha reduced sicklepod germination but did not have a consistent effect on other seedling growth parameters. Bromoxynil at 1.1 kg ai/ha plus 3.4 kg/ha sodium chlorate reduced sicklepod germination but did not have a consistent effect on most growth parameters. A wide range of treatments reduced hemp sesbania germination, emergence, and growth to very low levels. All desiccant applications reduced pitted morningglory seed production compared to the untreated check. Glyphosate at 1.1 kg/ha plus 6.7 kg/ha sodium chlorate, 0.3 kg/ha oxyfluorfen plus 3.4 kg/ha sodium chlorate, and 0.8 or 1.1 kg/ha glufosinate reduced pitted morningglory seed production and desiccated weeds effectively. Most treatments reduced emergence in 1996, when seeds were smaller and less mature than those collected in 1995. When application volume of paraquat alone or in tank mixture decreased, germination and growth of sicklepod was reduced, indicating increased paraquat efficacy.

Digital imagery from satellites and airborne remote sensing offer an opportunity to accurately detect weed infestations. Image resolution and plant growth stage are critical factors for maximum weed detection with low errors. Data analysis in traditional image assessment has relied on agreement measures, such as Cohen's kappa and asymptotic procedures, that compare what is on the image but not on the ground and what is on the ground but not on the image. Statistical comparisons of multispectral images, however, require some knowledge of the variability of the image classification results to determine significant differences among agreement measures. Bayesian methods were used to develop probability distributions for an agreement measure, conditional kappa, and were then subsequently applied to assess and compare image resolutions and plant growth stages. Results showed that images of a study site known to have yellow starthistle populations could identify the noninfested areas with greater accuracy than infested areas at spatial resolutions of 0.5, 1.0, 2.0, and 4.0 m. The detection accuracy of yellow starthistle in the images taken either prebloom or at flowering with 4.0-m spatial resolution usually was equal to or better than spatial resolutions of 0.5, 1.0, and 2.0 m for the cover classes that were not, moderately (31 to 70%), and highly (71 to 100%) infested. The 0.5-m resolution was better than 4.0-m spatial resolution when detecting the moderate cover class, but both resolutions had high omissional and commissional errors. Contrasting the best detection resolution for finding yellow starthistle colonies across flight times indicated that flying at flowering stage with the 4.0-m spatial resolution provided the best detection of the yellow starthistle cover classes considered. In the cases where different spatial resolutions resulted in equal detection accuracy, the larger spatial resolution was selected because of lower costs of acquiring and processing the data.

Field trials were conducted with spring-sown rye and field pea cover crops to determine the effect of five rye–pea proportions and three seeding rates (high, medium, and low) on weed suppression during cover crop growth. Measurements on weed and cover crop growth were taken approximately 2 mo after seeding when cover crops were killed. Cover crops were killed by mowing in 1996 and by undercutting in 1997 and 1998. Cover crop biomass, averaged over rye–pea proportion, was highest in 1998 at 4.3 million tons (MT)/ha (high seeding rate) and lowest in 1997 at 1.5 MT/ha (low seeding rate). Cover crops of pure rye or rye–pea mixes suppressed weeds more effectively than did pure pea. Dominant weeds were ladysthumb, smooth pigweed, smallflower galinsoga, and common lambsquarters. Ground cover by weeds ranged from a low of 2% (rye–pea mixes) to a maximum of 73% (pure pea). Cover crop mixes of 50% or more rye seeded at the high rate gave the best weed suppression.

Competition from large crabgrass is often responsible for the establishment failure of spring-seeded cool-season turfgrasses. The objective of this study was to determine how soon after seedling emergence dithiopyr could be applied without causing unacceptable damage to Kentucky bluegrass. Dithiopyr at 0.28 and 0.56 kg/ha was applied at various times prior to seeding and after emergence of Kentucky bluegrass in the field and greenhouse. Dithiopyr damaged Kentucky bluegrass seedlings in the greenhouse when applied earlier than 10 d after emergence (DAE) at 0.28 kg/ha and 14 DAE at 0.56 kg/ha. In the field study with heavy large crabgrass infestation, no significant phytotoxicity or long-term thinning occurred with any application of dithiopyr after emergence of Kentucky bluegrass. Dithiopyr applied as early as 3 DAE improved cover of Kentucky bluegrass by reducing competition from large crabgrass. When seeding is done early in spring before large crabgrass germination, dithiopyr could be applied at 0.28 kg/ha 10 DAE or at 0.56 kg/ha 14 DAE without risk of injury. Dithiopyr could be applied as early as 3 DAE on sites with heavy large crabgrass pressure or to late spring seedings when large crabgrass is germinating because the risk of turf thinning is warranted given the benefit of reduced large crabgrass competition for the desired turf.

Greenhouse experiments evaluated azafenidin and thiazopyr applied separately, as a tank mix, or in rapid succession, for preemergence control of yellow nutsedge (Cyperus esculentus) and hairy beggarticks (Bidens pilosa). Average of yellow nutsedge shoot dry weights at 21 and 42 d after treatment (DAT), indicated that azafenidin at 0.28, 0.56, and 1.12 kg ai/ha controlled 70, 86, and 90%, respectively, and thiazopyr at 1.12 kg/ha controlled 10% of yellow nutsedge. However, when thiazopyr at 1.12 kg/ha was tank-mixed with azafenidin at 0.28, 0.56, and 1.12 kg/ha, yellow nutsedge control averaged for 21 and 42 DAT was reduced to 41, 69, and 78%, respectively. Root plus tuber dry weight of yellow nutsedge increased threefold when 1.12 kg/ha thiazopyr was mixed with 0.28 kg/ha azafenidin, compared to azafenidin applied alone at the above rate. When the herbicides were applied sequentially, within minutes of each other, the antagonistic responses were reproduced, independent of the order of herbicide application. Thiazopyr applied at 0.56 and 1.12 kg/ha stimulated the emergence and establishment of hairy beggarticks and antagonized azafenidin for the control of yellow nutsedge based on a mathematical model. Although thiazopyr reduced the preemergence herbicidal efficacy of azafenidin on hairy beggarticks control, antagonism could not be determined because of the stimulatory effects of thiazopyr on hairy beggarticks emergence.

A greenhouse study was conducted to determine the effects of sublethal dicamba concentrations in the nutrient media on hydroponically grown tomato plants. Tomato leaf area was the most sensitive vegetative growth parameter measured in response to dicamba concentrations, ranging from 0 to 22 µg/L. Leaf area was reduced 31 and 76%, and specific leaf weights, a relative measure of leaf thickness (g/cm2), increased 26 and 121% after 30-d exposure to dicamba concentrations of 2.2 and 22 µg/L, respectively. In long-term experiments conducted until plants produced first ripe fruit, regression analysis indicated leaf area reductions of 8 and 66% from initial dicamba concentrations of 1 and 10 µg/L, respectively. Reductions in total fruit fresh weight were highly correlated (r = 0.93) with leaf area reductions caused by dicamba. A hyperbolic regression model gave predicted losses in fruit fresh weight per plant of 6% at 1 µg/L dicamba and 73% at 10 µg/L dicamba (r 2 = 0.87). Results generally indicated that the level of dicamba in the nutrient media of hydroponically grown tomatoes that produced no observable effect was ≤ 1 µg/L.

Field experiments were conducted in 1996 and 1997 near Arlington, WI, to compare the efficacy of glyphosate applied below registered rates in sequential and tank-mix combinations with residual herbicides to no-till, narrow-row, glyphosate-resistant soybean. In the sequential combinations of preplant (PP) residual herbicides and postemergence (POST) glyphosate study, glyphosate followed one of eight burndown treatments. Clomazone applied PP controlled 86% of common lambsquarters in 1996 when followed by 420 g ae/ha glyphosate early POST (EPOST). All other herbicide treatments controlled 94% or greater regardless of weed species, PP treatment, glyphosate timing, or glyphosate rate. The greatest soybean yields occurred in EPOST glyphosate applications in 1996 and late POST (LPOST) glyphosate applications in 1997. The only time PP residual herbicides were beneficial was prior to the LPOST glyphosate application in 1996. In the tank-mix POST combinations of glyphosate and residual herbicides study, glyphosate was applied alone or in combination with four residual herbicides. Soybean injury did not exceed 5% except in the glyphosate and imazethapyr combination in 1997. Control of common lambsquarters, velvetleaf, and giant foxtail was 90% or greater when averaged across all residual combinations and glyphosate timings and rates. Imazethapyr alone controlled velvetleaf 99% and giant foxtail 92% in 1997. When glyphosate was applied alone, soybean yields were similar at all glyphosate rates and application timings, except the 630 g/ha glyphosate LPOST resulted in a lower yield than 420 g/ha glyphosate LPOST. Only one residual herbicide, SAN 582, combined with glyphosate produced yields equivalent to the highest yielding treatments when averaged over both glyphosate rates and timings. Cloransulam added to 420 g/ha glyphosate EPOST and chlorimuron plus thifensulfuron and imazethapyr added to 420 g/ha glyphosate LPOST resulted in lower soybean yields compared to the same rate of glyphosate applied alone at the respective timings. Thus, no herbicide combination preformed better than glyphosate applied in a timely manner alone. However, in situations where early-season weed competition is severe and a timely glyphosate application is not possible, a PP residual herbicide may be beneficial.

The development of glufosinate-resistant corn represents a new weed management system for corn growers. Field experiments were conducted from 1995 to 1997 at four locations in southwestern Ontario. The objective of this study was to determine the effect of timing of weed control relative to the growth stage of corn with glufosinate applied alone or in combination with residual herbicides. Control of all species tested improved with the addition of atrazine plus dicamba to glufosinate, applied from the two- to eight-leaf stage of corn growth. Based on a 90% weed dry matter reduction, glufosinate with atrazine plus dicamba controlled common ragweed, common lambsquarters, and pigweed species at the three-leaf stage of corn and yellow foxtail, barnyardgrass, and large crabgrass at the two-, four-, and eight-leaf stage of corn, respectively. Weed control with glufosinate alone was improved when applied at the later growth stages of corn. Glufosinate applied alone at the four-leaf stage of corn controlled common ragweed and common lambsquarters, whereas pigweed species were controlled effectively at the eight-leaf stage of corn growth. Corn grain yield was consistently higher when glufosinate was applied in combination with residual herbicides, compared to glufosinate alone. Glufosinate in combination with residual herbicides applied to corn at the three- to five-leaf stage may represent the best timing for weed control. Our data suggested that a tank mixture of glufosinate with other postemergence residual herbicides or a split application of glufosinate in combination with cultivation may be required for weed control in glufosinate-resistant corn.

The objective of this study was to compare the consistency and accuracy of visually estimated weed biomass and weed control data to data obtained through image analysis. Weed biomass and weed control were evaluated in soybean herbicide efficacy trials conducted at the University of Nebraska–Lincoln during 1992 and 1993. Measurements were based on visual estimates and on aerial photographs taken at a height of 3.5 m above the soil surface. Photographs were digitized and classified, producing pixel values for broadleaf weeds, grass weeds, soybean, and soil. Percent weed cover was calculated in relation to the crop canopy, based on the respective number of pixels per image. Visual and photographic ratings of weed biomass and of weed control were not closely correlated. In the first year the visual method discriminated between more treatments than the photographic method, but the opposite occurred in the second year. The photographic method predicted yield more closely than the visual estimates. We concluded that visual estimates were less consistent and more subject to observer bias than measurements obtained with the photographic method.