Influence of straw management, nitrogen fertilization and dosage rates on the dissipation of five sulfonylureas in soil

The dissipation behaviour of metsulfuron-methyl, tribenuron-methyl, thifensulfuron-methyl, triasulfuron and amidosulfuron were studied in soil following post-emergence spring applications to cereal crops. Incorporation or removal of straw and different fertilizer applications had no influence on the disappearance time (DT₅₀) of the herbicides. However, in laboratory trials dissipation of metsulfuron-methyl, triasulfuron and amidosulfuron at higher application rates was accelerated after incorporation of straw into the soil. The addition of straw decreases soil pH thereby causing faster hydrolysis of the compounds. Addition of nitrogen fertilizer increased the half-lives (t_1/2) of the herbicides. Combination of both straw and N fertilizer, however, had no effect. Neither straw nor N influenced the degradation of tribenuron-methyl and thifensulfuron-methyl.
The DT₅₀ of all herbicides investigated varied from 6 to 17 days in the field trials. Except for amidosulfuron, herbicides could not be detected 38–68 days after application. In the laboratory, t_1/2 values were higher than those in the field at 75 days (amidosulfuron), 63 days (metsulfuron-methyl), 35 days (triasulfuron) and 13 days (tribenuron-methyl). Only the t_1/2 of thifensulfuron-methyl was at 29 h faster under laboratory conditions. T_1/2 values of all herbicides were influenced by their initial concentration. The t_1/2 values of amidosulfuron and tribenuron-methyl at 1 mg a.i. kg⁻¹ soil were 2–2.5 times higher than at the recommended field application rate of 42 μg a.i. kg⁻¹ soil.     

AN AUTORADIOGRAPHIC STUDY OF THE DISTRIBUTION OF 14C LABELLED ASSIMILATES AT DIFFERENT STAGES OF DEVELOPMENT OF OXALIS LATIFOLIA H.B.K. AND O. PES-CAPRAE L.

Summary. An autoradiographic study of the distribution of ¹⁴C-labelled assimilates, derived from ¹⁴CO₂ fed to the aerial parts, was made at ten different developmental stages of Oxalis latifolia H.B.K. and Oxalis pes-caprae L. Changing patterns of distribution of labelled assimilates to the underground parts of both species were revealed. These patterns are discussed in the context of the timing of control by defoliation and by foliageapplied herbicides.
Etude autoradiographique de la distribution des métabolites marqués an ¹⁴C à différents stades de développement de l'Oxalis latifolia H.B.K. et de l' O. pes-caprae L.     

Root exudation of diclofop-methyl and triasulfuron from foliar-treated durum wheat and ryegrass

In this study, we evaluated the release of diclofop-methyl and triasulfuron from the roots of foliar-treated ryegrass and wheat. The study with ¹⁴C-diclofop-methyl indicated a basipetal translocation of foliar-applied herbicide in wheat and ryegrass. No root exudation from ¹⁴C-diclofop-methyl-treated wheat plants was observed, while 20 days after treatment (DAT) 0.2–0.9% of radioactivity absorbed by ryegrass was found exuded in the growing medium. Root exudation was stimulated three to six times by the presence of untreated wheat or ryegrass sharing the growing medium with diclofop-methyl-treated ryegrass. No subsequent uptake of exuded radiolabel by untreated plants (ryegrass or wheat) in the same pot with ¹⁴C-diclofop-methyl-treated ryegrass was observed. The study with ¹⁴C-triasulfuron indicated a basipetal translocation of foliar-applied herbicide in wheat and ryegrass and also into the growing medium. By 20 DAT, 0.5–4.2% of radioactivity absorbed by wheat or ryegrass was found exuded in the growing medium. The presence of untreated plants (wheat or ryegrass) in the same pot as triasulfuron-treated ryegrass or wheat induced exudation seven to 32 times more. The study also revealed a subsequent uptake of exuded compounds by untreated wheat or ryegrass sharing the medium of ¹⁴C-triasulfuron-treated plants. This study has demonstrated for the first time that the root exudation of exogenous compounds can be related to plant arrangement in pots. The implication is that herbicide root exudation and transfer, a form of allelopathy, could be significant in the field. A precise estimation of environmental fate, unexpected ecological side effects and residual activity of herbicides may require quantification of such exudation.     

Modelling different cultivation and herbicide strategies for their effect on herbicide resistance in Alopecurus myosuroides

A single dominant mutation conferring resistance to aryloxyphenoxypropionate (AOPP) and cyclohexanedione (CHD) herbicides was incorporated into a quantitative model for the population development of Alopecurus myosuroide s Huds. The model predicts that from an initial seedbank of 100 seed m^–2, 10^–6 of which mutate to resistance each generation, and annual use of AOPP/CHD herbicides which kill 90% of susceptible but no resistant plants, a threshold of 10 plants m^–2 surviving herbicides ('field resistance') will develop: in 9–10 years if all tillage is by tine cultivation to 10 cm deep; after 28–30 years of annual ploughing; in 12 years if tine cultivations are interspersed with ploughing once every 4 years. If AOPP/CHD herbicides are alternated with herbicides with different modes of action, outcomes depend on the annual kill rate: with 95% kill (of susceptible plants by AOPP/CHDs and all plants by alternative herbicides) and tine cultivation, field resistance develops in 22 years; however, resistance can be delayed for 45 years if AOPP/CHDs are rotated with two additional herbicides, each with a different mode of action. The model predictions on the number of years required for field resistance to develop are not highly sensitive to the density of the seedbank or the initial frequency of resistance.     

Conyza albida: a new biotype with ALS inhibitor resistance

Summary A biotype of Conyza albida resistant to imazapyr was discovered on a farm in the province of Seville, Spain, on land that had been continuously treated with this herbicide. This is the first reported occurrence of target site resistance to acetolactate synthase (ALS)-inhibiting herbicides in C. albida . In order to characterize this resistant biotype, dose–response experiments, absorption and translocation assays, metabolism studies, ALS activity assays and control with alternative herbicides were performed. Dose–response experiments revealed a marked difference between resistant (R) and susceptible (S) biotypes with a resistance factor [ED₅₀(R)/ED₅₀(S)] of 300. Cross-resistance existed with amidosulfuron, imazethapyr and nicosulfuron. Control of both biotypes using alternative herbicides was good using chlorsulfuron, triasulfuron, diuron, simazine, glyphosate and glufosinate. The rest of the herbicides tested did not provide good control for either biotype. There were no differences in absorption and translocation between the two biotypes, the maximum absorption reached about 15%, and most of the radioactivity taken up remained in the treated leaf. The metabolism pattern was similar and revealed that both biotypes may form polar metabolites with similar retention time (R_f). The effect of several ALS inhibitors on ALS (target site) activity measured in leaf extracts from both biotypes was investigated. Only with imazapyr and imazethapyr did the R biotype show a higher level of resistance than the S biotype [I₅₀ (R)/I₅₀(S) value of 4.0 and 3.7 respectively]. These data suggest that the resistance to imazapyr found in the R biotype of C. albida results primarily from an altered target site.     

THE ADSORPTION OF LINURON, ATRAZINE AND EPTC BY MODEL ALIPHATIC ADSORBENTS AND SOIL ORGANIC PREPARATIONS

Summary. Model adsorbents were prepared by treating cellulose phosphate powder with a series of alkyltrimethylammonium compounds in which the size of the alkyl group was varied from C₈to C₁₈. The adsorption of linuron, atrazine and EPTC by these materials was found to increase logarithmically with increasing chain length. The extent of the adsorption was large compared with the adsorption of these herbicides by a humic acid and by a preparation made by removing the bulk of the inorganic constituents of a peat soil with a mixture of HCl and HF. Since soil organic matter is thought to contain alkyl groups, it is concluded that the possible influence of such groups should be considered in any discussion of the mechanisms involved in the adsorption of organic molecules by soils.
Adsorption du linuron, de I'atrazine de l'EPTC par des adsorbants de la série aliphatique et des préparations organiques de sol     

SELECTIVITY AND PHYSIOLOGICAL INACTIVATION OF SOME HERBICIDES INHIBITING PHOTOSYNTHESIS

Summary. The effects of several herbicides on the rates of CO₂ uptake and transpiration of intact plants were measured. A simple technique is described which reveals whether photosynthesis-inhibiting herbicides are inactivated by plants. By this method, the inactivation of simazine by maize could be confirmed. Evidence is presented for physiological inactivation of monuron by Plantago lanceolata , of cycluron (OMU) and pyrazon by sugar beet, of N'-(4-butoxyphenyl)-NN-dimethylurea by of 5-bromo-6-methyl-3-phenyluracil by flax. Certain plant species showed different reactions to some other new herbicides. The effect of herbicides in the culture solution on CO₂ uptake by the leaves is probably largely determined by the rate of transpiration of the plants. Assuming that uptake in proportional to transpiration rate, estimates of the concentration of the herbicides inside the plant were made.
L'inactivation physiologique de quelques herbicides produisant une inhibition de la photosynthèse     

Studies on herbicidal carbamoyltetrazolinone derivatives: Selection of fentrazamide

1-Phenyl-5(4 H )-4-dialkylcarbamoyltetrazolinones represent a new class of herbicides that are active against barnyardgrass ( Echinochloa spp.) in paddy fields. To select the most favorable compound as a paddy herbicide, herbicidal activity and selectivity of the carbamoyltetrazolinone compounds to transplanted rice were determined in greenhouse tests. Higher herbicidal activities were associated with the compounds having a 2-Cl- or 2-Cl-6-Me-substituted or an unsubstituted phenyl ring, and a combination of C_2–3 linear alkyl group and a C₄ branched-chain alkyl or a C_5–6 cycloalkyl group on the nitrogen atom in the carbamoyl structure. In addition, the compounds carrying a total of seven to nine carbon atoms in the substituents on the carbamoyl nitrogen exhibited a good selectivity between barnyardgrass and transplanted rice plants. From the results of soil thick-layer chromatography, the good crop compatibility of carbamoyltetrazolinones was correlated to the mobility of the compounds in paddy soil. The investigation resulted in the selection of an excellent compound, 4-(2-chlorophenyl)- N -cyclohexyl- N -ethyl-4,5-dihydro-5-oxo-1 H -tetrazole-1-carboxamide (common name: fentrazamide), as a rice paddy herbicide for controlling barnyardgrass.     

Effect of soil application of chloroxuron,metobromuron and fluometuron on nodulation,growth and nitrogen fixation by Centrosema pubescens and Vigna sinensis

The effect of soil applications of three phenylurea herbicides on the growth, nodulation and nitrogen fixation by Centrosema pubescens and Vigna sinensis under two plantings in pots were investigated under glasshouse conditions. Toxicity increased in the order chloroxuron < metobromuron < fluometuron with both legumes. C. pubescens was more susceptible to the effects of the herbicides than V. sinensis. The effects during cropping of C. pubescens and V. sinensis immediately after herbicide application and 9 months later was observed on nodule formation, plant growth and the proportion and quantity of nitrogen in the soil and in dry matter in the plant roots and tops. The levels at which the applied herbicides controlled weeds without any injurious effects on the legumes were for chloroxuron 0.5–2.0 mg/kg and for metobromuron 0.5–1.0 mg/kg but fluometuron killed weeds and was toxic to the legumes at the lowest concentration used (0.25 mg/kg).     

Control of the invasive liana, Hiptage benghalensis

The liana, hiptage ( Hiptage benghalensis ), is currently invading the wet tropics of northern Queensland and remnant bushland in south-eastern Queensland, Australia. Trials using seven herbicides and three application methods (foliar, basal bark, and cut stump) were undertaken at a site in north Queensland (158 700 hiptage plants ha⁻¹). The foliar-applied herbicides were only effective in controlling the hiptage seedlings. Of the foliar herbicides trialed, dicamba, fluroxypyr, and triclopyr/picloram controlled >75% of the treated seedlings. On the larger plants, the cut stump applications were more effective than the basal bark treatments. Kills of >95% were obtained when the plants were cut close to ground level (5 cm) and treated with herbicides that were mixed with diesel (fluroxypyr and triclopyr/picloram), with water (glyphosate), or were applied neat (picloram). The costings for the cut stump treatment of a hiptage infestation (85 000 plants ha⁻¹), excluding labor, would be $A14 324 ha⁻¹ using picloram and $A5294 ha⁻¹ and $A2676 ha⁻¹, respectively, using glyphosate and fluroxypyr. Foliar application using dicamba for seedling control would cost $A1830 ha⁻¹. The costs range from 2–17 cents per plant depending on the treatment. A lack of hiptage seeds below the soil surface, a high germinability (>98%) of the viable seeds, a low viability (0%) of 2 year old, laboratory-stored fruit, and a seedling density of 0.1 seedlings m⁻² 12 months after a control program indicate that hiptage might have a short-term seed bank. Protracted recolonization from the seed bank would therefore be unlikely after established seed-producing plants have been controlled.     

Sorption and leaching potential of herbicides on Brazilian soils

Sorption of the herbicides alachlor, atrazine, dicamba, hexazinone, imazethapyr, metsulfuron-methyl, nicosulfuron, simazine and sulfometuron-methyl was characterized on six Brazilian soils, using the batch equilibration method. In general, weak acid herbicides (dicamba, imazethapyr, metsulfuron-methyl, nicosulfuron and sulfometuron-methyl) were the least sorbed, whereas weak bases such as triazines and nonionic herbicides (alachlor) were the most sorbed. The Kd values found showed a significant correlation with soil organic carbon content (OC) for all herbicides except imazethapyr and nicosulfuron. Koc values showed a smaller variation among soils than Kd . To estimate the leaching potential, Koc and the ground-water ubiquity score (GUS) were used to calculate half-lives ( t_1/2 ) that would rank these herbicides as leachers or non-leachers. Comparison of calculated values to published values for t_1/2 demonstrated that sulfonylureas and hexazinone are leachers in all soils, alachlor is transitional, and atrazine, simazine and dicamba are leachers or transitional, depending on soil type. Results discussed in this paper provide background to prioritize herbicides or chemical groups that should be evaluated in field conditions with regard to their leaching potential to ground-water in tropical soils.     

Current status of biological control of paddy weeds in Vietnam

Rice is a staple food in Vietnam and accounts for > 7.7 × 10⁶ cultivated ha, which provide 35.5 × 10⁶ t of rice, of which 4.2 × 10⁶ t were exported in 2004. The enlargement of the cropping area and the enhancement of rice yield have rapidly increased the amount of agrochemicals, including herbicides, in crop production in Vietnam. From 1990–2003, the percentage of herbicides in total pesticides has increased ≈ 10-fold to 30.2%. In addition, the improper use of herbicides caused environmental hazards, unsafe agricultural products, and human health problems. Biological management integrated with traditional weed control techniques might help to reduce the dependence on synthetic herbicides and build eco-friendly, sustainable agricultural production in Vietnam. This paper reviews the efforts in establishing a strategy for biological management of weeds that was conducted in recent years by Vietnamese weed scientists. This has included cropping system management, water and soil management, integrated pest management, and utilization of plant allelopathy as major components of the strategy. Many plants with strong allelopathic potential can be a source for biological weed suppression and soil fertility improvement. The utilization of allelopathic properties in rice might also help to provide new rice cultivars with weed-suppressing characteristics.     

Weed infestation and tea growth under various weed management methods in a young tea (Camellia sinensis[L.] Kuntze) plantation

A field experiment was conducted in the low country of Sri Lanka, during the period 1994–1995 to investigate the severity of weed infestation and tea growth in relation to weed management methods in newly established tea ( Camellia sinensis [L.] Kuntze). Manual weeding (hand and slash weeding) at various intervals was compared with various herbicides, with or without mulching. Weed control with herbicides was superior to that of hand weeding at 6-week intervals or more. Weed control with oxyfluorfen at 0.29 kg ai ha⁻¹ + paraquat at 0.17 kg ai ha⁻¹ or glyphosate at 0.99 kg ai ha⁻¹ + kaolin at 3.42 kg ha⁻¹ were superior. Plots unweeded for 12 weeks or more produced significantly greater ( P  < 0.05) weed biomass than plots unweeded for 6 weeks. Although the least weed dry weight ( P  < 0.05) and the greatest number of weed species were recorded with hand weeding at 2 week intervals, there was no particular benefit on tea growth when compared with hand weeding at 6 and 12 week intervals. Inter row mulching in chemically treated plots was more favorable for tea growth than no mulching, while living weed cover in unmulched slash weeded plots suppressed tea growth. A combination of mulching and herbicides, particularly oxyfluorfen and paraquat, followed by hand weeding at least every 6–8 weeks was considered the most appropriate weed management system for young tea.     

A possible involvement of gibberellin in the mechanism of Avenafatua resistance to triallate and cross-resistance to difenzoquat

Avena fatua L. populations from numerous fields in Alberta have simultaneously developed resistance to recommended rates of two dissimilar herbicides, triallate and difenzoquat. We used exogenous applications of gibberellic acid (GA₃) to investigate the possibility that endogenous gibberellins were involved in the A. fatua 's resistance mechanism. For control plants, without applied GA₃, shoots of the most resistant (R) populations emerged more rapidly and elongated faster than shoots of the susceptible (S) populations. Increases in shoot elongation in response to exogenous GA₃ were significantly lower in R populations compared with S populations. This suggests that R populations may have elevated endogenous gibberellin levels, relative to S populations. Additionally, inhibition of S population shoot elongation and shoot anatomical abnormalities caused by relatively low concentrations of triallate and difenzoquat could be prevented by exogenous application of GA₃. These results suggest that there may be a phytohormonal involvement in the mechanism of triallate/difenzoquat resistance in A. fatua . That is, higher endogenous gibberellin levels in R populations may result in meristematic growth that is rapid enough to preclude phytotoxic levels of these herbicides from reaching the shoot meristem, which is the likely site of action.     

AN AUTORADIOGRAPHIC STUDY OF THE MOVEMENT OF 14C-LABELLED ASSIMILATES IN THE DEVELOPING WHEAT PLANT

Summary. ¹⁴CO₂ together with autoradiographic techniques was used to stuy the movement of labelled carbon within wheat plants at different stages in their development. Movement of ¹⁴C about the plain was found to be extensive in the early stages, gradually bccoming more localized until at a stage after ear emergence there was little transference of labelled carbon between tillers within a 24-hour period.
In the early stages of development ¹⁴C from leaves on the main shoot was transported throughout the plant but accumulated in greatest amounts in the meristematic regions. Assimilates produced by the leaves of newly formed tillers were distributed mainly to the meristematic regions of those tillers although labelled carbon was moved in into the rest of the plant. In the later stages of development of each tiller the labelled assimilate was restricted to the tiller itself and to any very young tillers associated with it. After car emergence the pattern of distribution of the labelled carbon from the youngest and second youngest leaves was found to differ. Movement was predominantly towards the ear from the flag leaf and towards the root system from the leaf below the flag.
It was noticed that mature leaves accumulated some labelled material and the possibility that this occurred via the transpiration stream following migration from phloem to xylem, as is the case with some of the case with some of the translocated herbicides, was investigated, but no indication of xylem transference was obtained.
The possible application of these studies to investigations into the relationship between the movement natural assimilates and of translocated herbicides is discussed.     

Comparison of sulfonylurea herbicide residue detection in soil by bioassay, enzyme-linked immunosorbent assay and hplc

The ability of bioassay, enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (hplc) methods to detect sulfonylurea herbicides in soil was evaluated as part of a project studying the leaching and persistence of these herbicides in the alkaline soils of south-eastern Australia. Soil samples with known concentrations between 0.1 and 10 μg a.i. kg⁻¹ chlorsulfuron, metsulfuron-methyl or triasulfuron were prepared by an independent laboratory and supplied in coded bags to separate laboratories for testing. The accuracy of the results was analysed, and the merits of each method are discussed. Bioassay was suitable for measuring biologically active residues from 0.1 to 1.0 μg a.i. kg⁻¹. ELISA accurately measured residues in the range of 0.1–10 μg a.i. kg⁻¹, making it the most widely adaptable assay tested. It will be useful for measuring residues in sodic subsoils where bioassay plants grow poorly. There was good reproducibility between the bioassay and ELISA. The hplc technique used in this study was not as accurate as bioassay or ELISA at quantifying residues of 3.0–10 μg a.i. kg⁻¹ and could not detect residues at or below 1.0 μg a.i. kg⁻¹.     

Weed seeds in long-term dryland tillage and cropping system plots

Successful crop production depends on effective weed control. Weed seedbanks were determined after 12 years of dryland cropping with winter wheat and grain sorghum under different tillage methods (no- and stubble mulch) and cropping sequences. Seeds of 12 species were detected. Amaranthus retroflexus was most abundant, but seed numbers were similar under all conditions. Portulaca oleracea , Panicum capillare , Setaria viridis and Sorghum halepense seed numbers differed because of some factors, with those for P. oleracea being greatest. For others, seed numbers were low (≤0.11 kg⁻¹ soil), except for Digitaria sanguinalis with 7.8 kg⁻¹ soil and Bromus japonicus with 1.3 kg⁻¹ soil. Most seeds were near the surface with both tillage methods, with enough present under most conditions to cause a problem if proper control measures were not used, especially with continuous cropping. When a weed problem occurs under conditions as in this study, the results indicate that it could be reduced by growing winter and summer crops in rotation, which permits controlling weeds with tillage and/or herbicides when a crop is not growing. Other possibilities for controlling weeds under conditions as in this study are to alternate between dicotyledonous and monocotyledonous crops, which would permit use of a wider range of herbicides, and to use selective in-crop herbicides.     

Effect of fungal toxins on germination of Striga hermonthica seeds

Fourteen fungal toxins were assayed in vitro to evaluate their effect on seed germination of the parasitic weed Striga hermonthica . Among them, T-2 toxin proved to be the most active, being able to inhibit 100% seed germination at 10⁻⁵ M, and being still active when tested at a concentration of 10⁻⁷ M (19% inhibition). Deoxinivalenol was also very active, causing 100% and 69% reduction in germination when assayed at 10⁻⁴ and 10⁻⁵ M respectively. Cytochalasin E, tenuazonic acid, fumonisin B₁, enniatin and nivalenol were shown to have an inhibitory effect of around 50% at 10⁻⁴ M, whereas other toxins had lower or no activity. The high activity shown by some fungal toxins suggests that they may have potential for use as more natural and safe herbicides to suppress parasite seed germination.     

Optimizing the performance of diclofop-methyl, cycloxydim, and clodinafop-propargyl on littleseed canarygrass (Phalaris minor) and wild oat (Avena ludoviciana) control with adjuvants

Optimizing the herbicide dose by the addition of adjuvants is an acceptable way to reduce the risk of side-effects from herbicides. Therefore, to detect a suitable adjuvant for diclofop-methyl, cycloxydim, and clodinafop-propargyl against littleseed canarygrass ( Phalaris minor ) and wild oat ( Avena ludoviciana ), six dose–response experiments were conducted. The treatments consisted of diclofop-methyl at 0, 112, 225, 450, 675, and 900 g ai ha⁻¹, cycloxydim at 0, 15, 30, 60, 90, and 120 g ai ha⁻¹, and clodinafop-propargyl at 0, 8, 16, 32, 48, and 64 g ai ha⁻¹ with and without the adjuvants of Frigate, olive oil, and castor oil at 0.2% (v/v) in order to control both littleseed canarygrass and wild oat. Tested herbicides performance was enhanced by all adjuvants against littleseed canarygrass and wild oat. The addition of Frigate and the vegetable oils had the lowest and the highest effect on the performance of all of the herbicides on both littleseed canarygrass and wild oat, respectively, which confirms the solubilizing nature of the cuticular waxes by vegetable oils. A comparison between the two vegetable oils revealed that olive oil exerted a greater control of littleseed canarygrass than did the castor oil. In contrast, castor oil exerted a greater control of wild oat than did the olive oil, which can be related to differences in the leaf surface micromorphology of the weeds.     

A SPLIT-ROOT TETRAZOLIUM METHOD FOR EVALUATING EFFECTIVENESS AND PHYTOTOXICITY OF ROOT-ACTIVE HERBICIDES

Summary. Copper sulphate (GUSO₄.5H₂O) and sodium hydroxide (NaOH) were used to test the performance of a split-root method for evaluating the effectiveness and phytotoxicity of root-active herbicides. This method involves a vertical separation of the root system of weeping willow ( Salix babylonica L.) cuttings. One section is treated with the herbicide dissolved in half-strength Hewitt's medium while the other remains in medium without herbicide during the test period. Roots are assayed with 2,3,5-triphenyttetra-zolium chloride 24 h after treatment to determine their viability. Both herbicides satisfactorily reduced the viability of the treated roots as measured by the quantity of formazan extracted. Copper sulphate was systemic and injured untreated plant parts; sodium hydroxide affected only treated roots. The split-root tetrazolium method is relatively rapid and does not require sophisticated laboratory equipment.
Méthode au tétrazolium sur portions de racines pour l'évaluation de l'efficacité et de la phytotoxicité des herbicides agissant par les racines