Field experiments were conducted during 3 yr in Thessaloniki, northern Greece, to determine the efficacy of various herbicides applied alone or in mixtures POST on hoary cress grown in winter wheat. Also, the efficacy of these herbicides on hoary cress generated from seed or root fragments was investigated in pot experiments. All herbicides except for ioxynil plus bromoxynil provided 90 to 100% control of hoary cress generated from seed and root fragments grown in pots. Ioxynil plus bromoxynil gave 84% control of hoary cress generated from seed, but only 66% control of plants generated from root fragments. In the field, mecoprop plus dicamba, imazamethabenz plus mecoprop plus dicamba, and triclopyr applied at the four- to eight-leaf stage provided, respectively, 87, 76, and 83% control of hoary cress (averaged over year and assessment time). Thifensulfuron plus tribenuron provided only 29% control. Chlorsulfuron plus dicamba, clopyralid plus MCPA, 2,4-D plus metosulam, and fluroxypyr provided intermediate hoary cress control. Furthermore, the least hoary cress emergence 52 wk after treatment appeared in plots treated with triclopyr. In addition, wheat grown in plots treated with mecoprop plus dicamba yielded more than wheat treated with the other herbicide treatments, and 55% more than wheat grown in nontreated plots. The results of this study indicate that very good control of hoary cress and high wheat yield can be obtained by the application of mecoprop plus dicamba at the four- to eight-leaf weed growth stage.

Today, the aim of weed management is to keep the weed community at an acceptable level rather than to keep the crop totally free of weeds. Satisfactory control of weeds may often be obtained when herbicides are used at lower doses than normally recommended. To facilitate the decision of what is an adequate dose in a specific field, the farmer needs support. In 1988 and 1989, a total of 10 field trials in spring cereals were initiated in Sweden with the objective of studying long-term effects of herbicide application according to recommendations from guidelines: the guidelines were developed at the Swedish University of Agricultural Sciences and consisted of printed cards designed for in-field use. Treatments also included a full and a half dose and an untreated control. As an average over the experimental time, i.e., until 1997, the dose used in the guideline treatment varied at different sites between 20 and 70% of a full dose. In 1998, i.e., 1 yr after the last herbicide application, the plant densities of annual weeds in the guideline treatment, the half and the full doses were 51, 57, and 67% lower, respectively, than in the untreated control when averaged over sites. At two and four sites, the half and full doses resulted in significantly lower weed densities than where guidelines had been used. Compared with the control, the full and half doses increased the proportion of difficult-to-control weed species significantly at five and four sites by 21 and 24%, respectively. In the guideline treatment the proportion of difficult-to-control weeds was increased at one site. In 1998, weed counts were higher where guidelines had been used than in the full dose for common lambsquarters and common chickweed at three sites each and for wallflower mustard, catchweed bedstraw, field violet, Galeopsis spp., and Lamium spp. at one site each. At three sites, no significant treatment effects on crop yields were found, whereas yields at the remaining seven sites were higher where guidelines had been used than in other treatments in several years. It is concluded that application of dose rates according to recommendations from guidelines can be a fruitful way to reduce herbicide use.

There are various reasons for using statistics, but perhaps the most important is that the biological sciences are empirical sciences. There is always an element of variability that can only be dealt with by applying statistics. Essentially, statistics is a way to summarize the variability of data so that we can confidently say whether there is a difference among treatments or among regression parameters and tell others about the variability of the results. To that end, we must use the most appropriate statistics to get a “correct” picture of the experimental variability, and the best way of doing that is to report the size of the parameters or the means and their associated standard errors or confidence intervals. Simply declaring that the yields were 1 or 2 ton ha−1 does not mean anything without associated standard errors for those yields. Another driving force is that no journal will accept publications without the data having been subjected to some kind of statistical analysis.

Different herbicide-application strategies may serve as part of an integrated weed management (IWM) system. In 1987 and 1988, ten field trials were initiated in the south-central part of Sweden, with the objective of studying the long-term effects of herbicide-application strategies on the development of weed populations and productivity of spring-sown cereals. Each year until 1997, herbicides were applied at 25, 50, 75, or 100% of the full recommended dose. Treatments included herbicide exclusion every second year and herbicide application at 25 or 50% of a full dose during two out of three years, with a full dose applied in the third year. An untreated control was included. In 1998, weed densities in 25, 50, 75, and 100% of a full dose were reduced 43, 58, 64, and 67% compared with the control. At seven sites, the exclusion of herbicides every second year resulted in 43 to 178% higher weed densities than in 50% of a full dose, although the same amount of herbicide was applied when summarized over 2 yr. At four sites, herbicide application each year at 50, 75, or 100% of a full dose increased the density of difficult-to-control species by 24% compared with the untreated control. Averaged over sites, the untreated control contained 30% more weed species than herbicide application at 75 or 100% of a full dose. At three sites in 1996, the full dose resulted in 520 kg ha–1 (pooled over sites) higher barley yields than in the control. Neither in 1996 nor in 1997, was there any difference in crop yield between herbicide application at 25 and 100% of a full dose. This study demonstrates the potential of reducing the input of agrochemicals for weed control by using herbicides at reduced rates in competitive crops.

Information on chemical weed control in lily bulb production in South America is scarce. Greenhouse and field studies were conducted to evaluate the phytotoxic effect and weed control of herbicides applied PRE and POST in lily bulb production in Argentina. In greenhouse studies, bromoxynil, 415 g ai ha−1; fluroxypyr, 200 g ai ha−1; metsulfuron, 3 g ai ha−1; iodosulfuron-methyl-sodium, 3 g ai ha−1 + metsulfuron, 3 g ai ha−1; oxyfluorfen, 240 g ai ha−1; ioxynil, 529 g ai ha−1; and linuron, 750 g ai ha−1, produced severe phytotoxicity or death of bulbs. Glyphosate at 720 g ai ha−1 and aclonifen at 720 g ai ha−1 produced little to no symptoms and were considered safe to apply to lilies. In field conditions, PRE herbicides metolachlor, 960 g ai ha−1 + atrazine, 1,500 g ai ha−1, and metolachlor, 960 g ai ha−1 + flumetsulam, 80 g ai ha−1, provided good weed control but were phytotoxic for lily plants, with chlorosis as the main symptom. Metolachlor plus linuron resulted in little or no symptoms of injury and no reduction in bulb yield. Diuron, 800 g ai ha−1 POST was the most effective treatment without phytotoxicity, and, in combination with metolachlor, 960 g ai ha−1 + linuron, 750 g ai ha−1 PRE, controlled weeds until 40 d after diuron application without yield reduction. Results obtained with glyphosate indicate that the Lilium genus presents some tolerance to this herbicide, which justifies further evaluation for weed control in lily bulb production.