A practical expert system is needed to handle POST weed control in herbicide-intensive crops such as wheat, barley, oat, rye, sugarbeet, corn, and sorghum. SELOMA is an expert system having a step-by-step problem-solving procedure closely resembling what a weed management expert would follow. It is based on field surveys of weed density, and crop and weed growth stage and height. SELOMA evaluates weed competitiveness and provides weed management advice. It suggests whether or not to intervene, chemical and mechanical weed control treatments, and selects the best herbicides, including commercial formulations, costs, and optimal dosages.

We present a germination and emergence model that can be used as a sub-model in an individual-based model of population dynamics. Seed germination and seedling elongation rates were measured in petri dishes in the laboratory for ivyleaf morningglory, velvetleaf, corn, and soybean seeds, as functions of temperature and water potential. The analysis yielded a set of indices: the germination temperature index (GTI), the germination water index (GWI), the emergence temperature index (ETI), and the emergence water index (EWI). The seed populations were divided into 100 discrete cohorts, with each cohort (i), having its own germination rate (GR) as the product of a reference rate and the germination indices, or GRi = GRiREF × GTI × GWI. After germination, the emergence rate (ER) was the product of a reference rate and emergence indices, or ERi = ERREF × ETI × EWI. The model was tested against the timing of emergence in the field for seeds planted 1, 2, 4, and 8 cm deep in natural or rain-excluded environments. The predictions were more accurate for all depths combined than for any particular depth.

Field experiments in 1988 and 1989 evaluated five POST herbicides alone or combined for horse purslane control in peanut. Acifluorfen, acifluorfen plus bentazon, pyridate plus 2,4-DB, and lactofen controlled more than 75% when rated up to 26 days after treatment (DAT). However at 67 DAT, acifluorfen alone, acifluorfen plus bentazon or 2,4-DB, or lactofen applied EPOST controlled horse purslane greater than 60% in 1988. In 1989, only 2,4-DB or pyridate plus 2,4-DB controlled horse purslane at least 70% at 116 DAT. Pyridate alone did not control horse purslane adequately.

Carryover potential of imazaquin (0.14 kg ai ha−1), chlorimuron plus metribuzin (0.06 + 0.36 kg ha−1), and fomesafen (0.28 kg ha−1) to snap bean, watermelon, cucumber, mustard, and sunflower was evaluated in Arkansas in 1987 to 1989. Crops were direct-seeded into treated plots at various times after application. Imazaquin and chlorimuron plus metribuzin injured sunflower, watermelon, cucumber, and mustard when planted 16 wk following application in one of two years. Fomesafen injured all crops initially but did not injure snap bean, sunflower, watermelon, and cucumber planted 16 wk after application. In experiments in which herbicides were applied to soybean, fall-planted spinach and mustard were injured in one of two years.

The effect of annual spring PRE applications of fenac, metribuzin, and terbacil on the competitiveness of sugarcane with bermudagrass during a 3-yr crop cycle was evaluated. For each herbicide treatment, bermudagrass was allowed to interfere with sugarcane during the growing season for 0, 1 (plant-cane crop only), 2 (plant-cane and first-ratoon crops), or 3 (plant-cane through second-ratoon crops) yr. Predicted total yields of cane from the three harvested crops was 3600 kg ha−1 higher on average in plots maintained free of bermudagrass and treated with metribuzin than where fenac and terbacil were applied. Bermudagrass biomass increased with each growing season. Greatest biomass production occurred in the fenac plots and the smallest in terbacil plots. Bermudagrass did not reduce the predicted total yields of cane where metribuzin or terbacil was applied. With fenac, predicted total cane yields were 4, 7, and 11% lower than the weed-free control where bermudagrass interference was allowed for 1, 2, and 3 yr, respectively. Season-long bermudagrass interference in the fenac plots reduced actual cane yields 10% (plant-cane crop), 6% (first-ratoon crop), and 14% (second-ratoon crop) when compared with plots maintained free of bermudagrass for the entire growing season.

Oxyfluorfen (0.28, 0.42, 0.56, and 0.84 kg ai ha−1) under clear polyethylene film was evaluated for weed control, crop injury, and effects on yields in transplanted muskmelon, cucumber, and summer squash. Numerous narrowleaf and broadleaf weeds were effectively suppressed by 0.42 ha−1 of oxyfluorfen. Crop injury, occurring soon after transplanting, was reported in New York and North Carolina. Injury was usually transient, and injured crops frequently grew more vigorously than those grown on untreated black polyethylene mulch. Muskmelons were consistently the most tolerant of the three crops. At high rates, yields of squash and cucumber in 1988 were reduced in New York and North Carolina, respectively. In greenhouse studies, positioning the cotyledons under the polyethylene film caused greater injury in all three crops than when cotyledons remained above the plastic.

Field studies were conducted in 1990 and 1991 to determine the effects of corn planting date and hairy vetch control method on the efficacy of fall-planted hairy vetch as a weedsuppressive cover crop for no-till corn. Glyphosate controlled hairy vetch when applied at the early bud growth stage (April), but hairy vetch residue provided no weed control compared to the weedy check. Mowing was not an effective means of suppressing hairy vetch at the early bud stage. Untreated hairy vetch reduced weed biomass 96% in 1990 and 58% in 1991 but reduced yield over 76% in April-planted corn. There was no competition of untreated hairy vetch with corn when corn planting was delayed until May or June (mid- or late-bloom growth stages of hairy vetch). Corn planted in May into untreated hairy vetch yielded similarly to corn planted in a no-cover weed-free check.

Field studies were conducted from 1988 to 1990 on a Sharkey clay to evaluate residual weed control in rice with quinclorac applied PPI, PRE to dry soil, and PRE to moist soil. Quinclorac applied at 0.4 or 0.6 kg ai ha−1 PPI or PRE to dry or moist soil controlled more than 80% of barnyardgrass, pitted morningglory, and hemp sesbania without rice injury. Quinclorac applied at 0.3 kg ha−1 controlled these three weed species substantially but inconsistently. No rice injury was observed from any quinclorac treatment. Except for one of three years when irrigation was delayed for 7 d after PRE application to dry soil, application timing did not consistently affect weed control or rice yield.

Interactions of fluazifop-P, sethoxydim, clethodim, and quizalofop-P with DPX-PE350 were evaluated. DPX-PE350 at 110 g ae ha−1 reduced control of large crabgrass when mixed with all graminicides and control of broadleaf signalgrass when mixed with fluazifop-P or quizalofop-P. Increasing the rate of graminicide reduced or eliminated antagonism. Applying sethoxydim and DPX-PE350 sequentially eliminated antagonism on large crabgrass. Control of seedling johnsongrass by the graminicides was not reduced by DPX-PE350. DPX-PE350 reduced control of rhizomatous johnsongrass by sethoxydim. Graminicides had no effect on entireleaf morningglory control with DPX-PE350.

Three levels of weed control (0, 70, and 100%) were maintained in 1989 during corn production at two sites. Weed control in 1990 with a premix formulation of metolachlor and atrazine was directly related to 1989 control levels. Herbicides generally were more effective in conventional tillage than in no-tillage. In no-tillage, giant foxtail control in 1990 averaged 59% following 100% control in 1989 compared to 15% following 0% weed control. At one location, tillage minimized the impact of the prior year's weed control with herbicides providing greater than 85% control in conventional tillage regardless of prior history. In no-tillage, weed control of 85% or better was obtained only following 100% control the previous year.

Research was conducted to determine an effective morningglory control system for cotton grown in central Texas, Prometryn applied PRE or directed-POST applications of methazole or methazole + MSMA did not always adequately control ivyleaf, entireleaf, and seedling sharppod morningglories. However, adequate control of the above species plus pitted morningglory was achieved with a management system that used a dinitroaniline applied PPI or PRE, prometryn applied PRE (0.9 kg ai ha−1), followed by methazole (0.56 kg ha−1) or methazole + MSMA (0.56 + 1.80 kg ha−1) directed POST to cotton 10 to 15 cm high.

Experiments were conducted to compare control of entireleaf morningglory with POST applications of DPX-PE350 applied with no adjuvant, a nonionic surfactant, or crop oil concentrate in spray volumes of 94, 235, and 375 L ha−1. Control was greater with a nonionic surfactant or crop oil concentrate than with no adjuvant and control with crop oil concentrate exceeded control with the nonionic surfactant. Spray volume had no effect on efficacy. DPX-PE350 was more effective when applied to foliage only or to soil plus foliage than when applied to soil only after emergence of entireleaf morningglory. Control from soil-only applications was lower on a silty clay soil than on a silt loam soil. Except for sicklepod, mixing DPX-PE350 with fluometuron or MSMA had little effect on weed control.

Experiments were conducted in 1988, 1989, and 1990 at two locations in Missouri to study the influence of fall tillage on the soil persistence of several soybean herbicides and subsequent injury to the rotational crops winter wheat, corn, cotton, and grain sorghum. Chlorimuron, clomazone, imazaquin, imazethapyr, and metribuzin plus chlorimuron were applied at their label and double label rate (2X) in soybean. Fall chisel plowing did not influence the carryover potential of these herbicides on any of the crops, in any year, or at either location of the research. However, herbicides injured several of these rotational crops. The 2X-label rate of clomazone reduced winter wheat grain yield at both locations. None of the herbicide treatments at either location reduced corn yield. Imazaquin applied at the 2X rate caused greater than 30% visible injury to cotton in 1989; however, cotton yield was not affected. The 2X rate of chlorimuron caused a reduction of grain sorghum yield in 1989, but not in 1990 or 1991 at Novelty.

A nonionic surfactant with the appropriate structure and concentration increased rimsulfuron activity 10-fold. Surfactants sharply increased rimsulfuron activity as concentration increased up to 0.05% (w/w) with slightly more enhancement up to 0.1% (w/w). Surfactants with a wide range of 14 chemical, physical, and surface properties were evaluated on giant foxtail, velvetleaf, and corn. Two properties, HLB (hydrophilic-lipophilic balance) and the physical form of the spray deposit, explained the biological variation. The most active surfactants had an HLB from 12 to 17 and formed a moist gel spray deposit on the leaf on all three species. A separate study with 13 chemically pure linear alcohol ethoxylates supported these conclusions and additionally showed that increasing molecular size increases activity. Because crop injury correlates with weed control, the best surfactant strategy is to maximize weed control at the lowest rimsulfuron rate.

Nicosulfuron (50 and 100 g ai ha−1), primisulfuron (40 and 80 g ai ha−1), imazethapyr (70 and 140 g ai ha-1), and DPX-PE350 (100 and 200 g ai ha-1) injury to corn, cotton, grain sorghum, rice, and soybean was evaluated at Fayetteville, AR, during 1989-90 and 1990-91. The crops were planted 0, 1, 2, 4, 8, 14 or 16, and 52 wk after herbicide application. Visible injury to all crops declined as the time interval increased. No visible injury was observed to any crop planted at least 14 wk after nicosulfuron or primisulfuron application, or 52 wk after DPX-PE350 application. Imazethapyr injured all crops except soybean at the 52-wk interval in 1990 but not in 1989.

Field experiments were conducted to determine carryover potential to grain sorghum and soybean of DPX-PE350 applied POST at 0.05, 0.1, and 0.2 kg ai ha−1 to cotton the previous year. DPX-PE350 did not injure soybean or affect yield adversely. Grain sorghum was injured and maturity delayed on a Sharkey silty clay but not on a Calloway silt loam. Grain sorghum yield was reduced on both soils 16 and 22%, respectively, by residues from the 0.1 and 0.2 kg ha−1 rates of DPX-PE350. In an incubation study, dissipation of DPX-PE350 was greater at 35 C than at 5 C., and did not differ between the two soils.

Three field experiments were conducted in 1987 through 1989 to evaluate hogpotato control and rotational crop response resulting from applications of five herbicides. At the end of the first year, triclopyr and imazapyr controlled hogpotato late-season as high as 87 and 94%, respectively. Following a sequential application in 1988, triclopyr and imazapyr controlled 94 and 100% of hogpotato, respectively, during the 1989 growing season. Hogpotato biomass, collected at the termination of the experiments, was reduced by all treatments at all locations except at Altus where dicamba was applied in 1987 followed by a spot application of glyphosate in 1988. Soils from these experiments were bioassayed for residual herbicide activity using three potential rotational crop species. Tebuthiuron and imazapyr caused the greatest biomass reductions with cotton and wheat being the most sensitive species to these herbicides.

Asulam applied POST alone and in mixtures with residual herbicides and with DSMA and MSMA was evaluated for johnsongrass control. Johnsongrass control following treatment with asulam at 3.7 kg ai ha−1 was not improved by tank mixtures with the residual herbicides atrazine, metribuzin, pendimethalin, and terbacil compared with asulam applied alone. Johnsongrass injury symptoms appeared sooner when asulam was applied in mixtures with either DSMA or MSMA, but johnsongrass control 4 wk after treatment was similar to asulam applied alone. As a result of the johnsongrass control obtained with early POST treatments of asulam applied alone, cane and sugar yields were increased by 56 and 58%, respectively, when compared with the weedy check. Additional cane and sugar yield increases were not obtained by any of the asulam mixtures because johnsongrass control was not improved. Asulam applied again in May to johnsongrass regrowth improved late-season control in two of the three studies over a single early POST application, but cane and sugar yields were not increased.

In 1987-89, field experiments were conducted to evaluate rates of activated carbon in transplant water to reduce injury to tobacco from imazaquin. Rates of activated carbon were 0, 2.2, 4.5, and 9 kg ha−1 in 1987 and 0, 9, 18, and 36 kg ha−1 in 1988-89. Imazaquin rates were 0, 0.3, and 0.6 kg ha−1 applied over-the-top immediately after transplanting. Results indicated that 9 kg ha−1 of activated carbon was the minimum amount to reduce injury from imazaquin. No additional protection was achieved with rates above 9 kg ha−1, whereas lower rates appeared to offer less protection based on visible stunting ratings.