Two relatively new methods for analyzing herbicide efficacy data are described. Weighted multiple regression using the logit transformation for plant mortality data is illustrated and compared with the more accurate maximum likelihood logistic regression procedure. A partial data set evaluating the effects of increasing application rates of picloram (0, 1.1, 2.2 and 4.5 kg ae ha–1) for control of tall larkspur is used to illustrate the methods. Suggestions are made for using logistic regression to monitor herbicide efficacy over several years.

Weed science has been and will continue to be an evolving discipline. The evolution and future of weed science will depend greatly on our visions of the future and our willingness to effect and implement change. Thus, the principal question we must ask ourselves is, what do we want weed science to be in the future? This paper, which is only one part of the symposium on the Future of Weed Science, will address the issues involved with teaching. This requires identification of future students and employers, their educational needs based on professional goals, and packaging and presenting the information. Most future agriculture undergraduates and weed science graduate students will need to be recruited, will have minimal background in agriculture, and will include more minorities and women than in the past. We will offer a more interdisciplinary and internationalized curriculum to future students. The curriculum will include traditional courses with greater emphasis on communication skills, business and economics, computer science, agricultural ethics, and interpersonal skills. Courses will be taught to future students by a regional teaching concept that uses electronic media. Teaching and learning about weed science in the future will be exciting, challenging, and unquestionably different from today.

At the 1990 meeting of the Weed Science Society of America (WSSA) in Montreal a Symposium titled “Pesticide Transport to Groundwater: Perspectives and Field Methodology” was Chaired by Thanh H. Dao. That symposium made it clear that simulation modeling is being used more and more to integrate the field test data generated by pesticide companies as part of the environmental impact assessment required for pesticide registration with the Environmental Protection Agency (EPA). At the Soil Aspects Section Business Meeting it was decided that a Symposium on the state of the art of this technology would be timely and the authors agreed to organize it.

This study evaluated the response of sweet cherry to different herbicides applied at rates simulating drift. Chlorsulfuron, thifensulfuron, bromoxynil, 2,4-D, glyphosate, and a combination of 2,4-D and glyphosate were applied on one side of one- and two-year-old established cherry trees at 1/3, 1/10, 1/33, and 1/100 of the maximum rate for small grain production. The order of herbicide phytotoxicity was chlorsulfuron > 2,4-D > glyphosate > 2,4-D + glyphosate > thifensulfuron > bromoxynil. Trees recovered from injury caused by all treatments except higher rates of chlorsulfuron, 2,4-D, and glyphosate. The herbicides caused characteristic symptoms, but some resembled disease, mineral deficiency, and environmental stress symptoms. Therefore, any allegations about herbicide drift based on chronic symptoms should be supported by analysis of plant tissue.

The effect of two dates of initial manual weeding, i.e., before flowering and after flowering, and of no weeding on yield of a local groundnut variety was studied during 1988 and 1989 at Minkoameyos, Yaounde, Cameroon, in comparison with chemical weed control method or control. The use of herbicides ametryn plus promotryn for weed control gave the best mean yield, followed by handweeding after flowering, and handweeding before flowering, in that order. The no weeding treatment had the lowest yield throughout the trial period.

Computer based simulation models are increasingly being used to predict the environmental fate of crop protection chemicals. Some considerations that need to be given in selecting appropriate models for regulatory purposes include model applicability, validation, capability, ease of use, and documentation. Problems commonly encountered in modeling include limited accuracy, lack of defined objectives and standard modeling practices, and misuse of models and results. Models will continue to play an important role in the regulation of crop protection chemicals. It is important that regulators and industry agree on appropriate models and practices, and that regulatory decisions are not based solely on model results but take into account all available data.

Weed Science Extension is in a period of change. Though there is decreased funding for extension in the United States and Europe, the number of programs including extension activities is increasing in developing countries as shown by World Bank funding. There is a perception problem with the name weed science; we need a strong, recognizable science name. Decreases in one-on-one contact method for teaching, fewer extension weed specialists, and the increased need for interdisciplinary projects to support Integrated Pest Management and Sustainable Agriculture will increase stresses on our positions. These factors also increase the challenge for weed specialists and extension advisors (agents) to become leaders in current programs, initiate new multidisciplinary programs, and choose those programs that will benefit our discipline. Our clientele and the method of communicating with them will change, presenting yet another challenge.

Yellow nutsedge control with rust (Puccinia canaliculata (Schw.) Lagerh.) collected in Salisbury, MD., was evaluated alone or in combination with pebulate at 1.7 or 3.3 kg ai ha-1 in transplanted tomato. Foliar applications of urediniospores in mid-June did not affect the yellow nutsedge population; however, tuber biomass was reduced 32%. In July, yellow nutsedge infection was manifest as an increase of phytotoxicity from 2.3 to 4.1 (0 to 10) and of disease incidence increase from 3 to 37%. Pebulate, 3.3 kg ha-1 reduced the yellow nutsedge population and ground cover by 79 and 90%, respectively, in June, with a 48% reduction of both tuber population and biomass in September. Tomato yields were not increased by rust or pebulate treatments. Marketable tomato yield was increased 46% by handweeding. Yellow nutsedge control was not enhanced by the combination treatments of pebulate followed by rust inoculation.

In a three county area, the density of musk thistle was higher than plumeless thistle in pastures in the easternmost county, equivalent in the central county, and lower in the westernmost county. Bull thistle was detected frequently, but density was always low. Detection of Canada thistle decreased from east to west.

Experiments were conducted in 1989 and 1990 to study the effects of 220 g ai ha-1 AC 263,222, 840 g ai ha-1 glufosinate, 560 g ai ha-1 glyphosate, and 840 g ai ha-1 paraquat on sicklepod seed production and quality when applied at R5, R6, R7, and R8 growth stages of sicklepod. No seed were produced on plants treated at R5 with paraquat and glufosinate; glyphosate was the least effective herbicide at this growth stage. Although sicklepod seed germinated after AC 263,222 application at R5, no seedling emergence occurred. Normal seedlings, seedling emergence, and seedling radicle length were reduced by all herbicides applied to parent plants at R5, R6, and R7 growth stages. AC 263,222 applied at R7 reduced normal seedlings and seedling emergence more than any other herbicide. No effects were observed on any of the variables investigated when herbicides were applied at R8.

The use of models for regulatory purposes requires accurate descriptions of important chemical fate processes, techniques for applying models for specific regulatory issues, techniques for parameterizing the model, and sufficient model distribution and support. Researchers (i.e. model developers), on the other hand, want to conceive more and more complex, sophisticated models in order to better represent the important environmental processes; unfortunately this invariably tends to lead to models with greater data requirements, parameters that are difficult to evaluate, greater expertise needed for application, and greater computational requirements, all of which makes for more difficult model application and use. The development of usable models for both current needs and future regulatory requirements demands that we achieve a balance between these ‘research needs' and ‘user needs'. Software engineering and code development issues can help to establish the needed bridge between research and user needs so that future models can more quickly escape the research arena and satisfy user needs as effective regulatory tools.

The agricultural-chemical industry has been offering solutions in plant protection for several decades. The primary role of Industry in weed control has mainly been to provide chemicals with increasingly attractive profiles regarding safe handling, ecological acceptance, and economical attractiveness. These objectives have been reached by significant developments, e.g. in rates, persistence, and specificity for non-target organisms. It is therefore safe to assume that in the year 2015 and beyond chemicals will continue to maintain a major role in weed control. However, as cropping systems and criteria for desirable control levels change, industry will have to change from a re-active to a pro-active participant in the development of integrated systems. Chemical solutions will be complemented by biological and agronomical methods and will be further supported by biotechnological successes in the crop area. In addition, it is anticipated that sophisticated computer models now in development will help exploit the potential of products as well as of integrated systems. Thus, fully integrated companies active in chemical, biological, and molecular-biological research and having branches in the agro-chemical as well as in the seed business will be especially suited to be driving forces in the changing world of modern weed control. The practice will ask for services much more than for single products. Industry will not only have to offer those services but at the same time assist in the education of growers to enable efficient use of the increasingly intricate methods of future weed control.

Field studies were conducted in 1988 and 1989 to examine the interrelations of cultivation and herbicide band width in controlling pitted morningglory and sicklepod in soybean. Alachlor + imazaquin PRE followed by imazaquin + surfactant POST were not applied or were applied on bands 15, 30, 45, 60, 75 cm wide or were broadcast. Plots were cultivated zero, one, two, or three times. Without cultivation, at least a 60-cm wide band was needed to achieve maximum soybean seed yields. A slight linear increase in soybean seed yield in response to increasing band width was observed with one cultivation. With two cultivations, soybean seed yields were similar with and without herbicides, but a 15-cm wide herbicide band was needed to achieve maximum production with three cultivations. Greatest gross returns exclusive of weed management costs were realized with two or three cultivations and a 15-cm wide band.

Sulfonylurea herbicide-resistant prickly lettuce was discovered in Idaho in 1987. The objectives of these surveys were to determine the change with time in the proportion of sulfonylurea resistant and susceptible prickly lettuce biotypes on the farm where it originally occurred, and to determine the spread of sulfonylurea resistant prickly lettuce beyond its point of origin. On average, the proportion of resistant plants had decreased from 1988 to 1990, a period when sulfonylurea herbicide use was discontinued on the farm. Resistant prickly lettuce plants were found at seven sites away from the original infested farm. Several sites were near the farm along roadsides that had been sprayed with sulfometuron. This study shows that the proportion of resistant prickly lettuce decreased where previously found, but its range increased.

The movement and degradation of agricultural chemicals in soil and groundwater is a complex process, influenced by climatic conditions, site characteristics, chemical properties, and agricultural practices. Because of this complexity, mechanistic models are useful tools in evaluating environmental fate. Although currently available unsaturated zone models often do not provide accurate concentration profiles when compared with field measurements, they can be used to estimate the depth and extent of movement. Another application of modeling is to compare predicted movement under different simulation conditions to determine the significance of the various factors affecting the behavior of agricultural chemicals. When properly designed and interpreted, modeling studies can contribute to the understanding of the environmental fate of agricultural chemicals. When necessary, modeling can also be useful in developing management practices included on the label or in state regulations. Modeling has been a frequently used tool in developing restrictions for minimizing the movement of aldicarb residues into groundwater and preventing residues in drinking water. Examples of model use include optimizing application timing, estimating well setback distances, and identifying vulnerable soils.

An experimental directed sprayer has been improved for use in conservation-tillage and no-tillage field conditions and variable row widths. The directed sprayer consists of a rear-mounted toolbar with individually flotational spraying units on lightweight runners which move over field surfaces without the collection of loose residues. Improvements allow convenient adjustments which combined lateral, vertical, angular, and fore-and-aft repositionings to correctly aim the hydraulic nozzles for desired pesticide targeting. Such new designs are needed to deliver pesticides to targets at minimum application rates and with minimum environmental impacts.

Threshold weed management methods have recently been elaborated to consider effects of threshold management on weed population dynamics. Such economic optimum thresholds are calculated using population-dynamics models which require detailed information about weed demography, including seed production (as affected by events between germination and seed dispersal), seed dispersal, and seed survival and movement in soil. Factors affecting any of these aspects of demography appear likely to modulate the growth rate of a sub-threshold population and therefore to influence the economic optimum threshold value. To test this conjecture and evaluate weed threshold management, including associated risk, improved understanding is particularly needed of weed seed dispersal, seedbank processes, and unpredictable demographic variation.

Predicting the off-target movement of agrochemicals applied to turf is an essential component for assessing risk to aquatic environments. For surface transport, a fully validated model does not exist. Such a model is required to make cost effective risk assessments and regulatory evaluations. To address this deficiency, a turf model validation program is outlined in this paper. The effect of turf characteristics on the fate and movement of pesticides is presented followed by the discussion of the selection for a proper turf runoff model. The design and implementation of small plot runoff studies and their usefulness in model validation are discussed. The procedures of model calibration and model verification are explained for model validation purpose.

When applied correctly, atrazine can be used safely in certain areas of the United States for weed control in winter wheat, but persistence in the soil may injure wheat replanted after crop failure. Studies were conducted at Moro, OR, in 1989–90 and 1990–91 to determine the effect of September preplant applications of atrazine at 0.56, 1.12, and 2.24 kg a.i. ha-1 (1X, 2X, and 4X label rates) on wheat seeded at that time and on wheat re-seeded 2 or 6 mo later. Yield of winter wheat re-seeded in November following September atrazine applications was not reduced at any rate. The September-planted wheat was killed with glyphosate in the spring to simulate winter killing, then plots were re-seeded to spring wheat. The commercial rate of atrazine, 0.56 kg ha-1, caused only minor injury to spring wheat, but injury at the 2X and 4X rates was excessive.