Previous PageTable Of ContentsNext Page

Evaluation of the allelopathic potential of a mustard cover crop

Jeffrey D. Weidenhamer1, Jeannette Durkalski2 and Warren A. Dick2

1Department of Chemistry, Ashland University, Ashland, OH 44805 USA Email
School of Natural Resources, Ohio Agricultural Research and Development Center – Ohio State University, Wooster, OH 44691 USA


No-tillage agricultural systems provide environmental benefits by greatly reducing soil erosion. Weed control, however, depends heavily on synthetic herbicides that may degrade environmental quality. Cover crops can provide an alternative method for weed suppression through allelopathic and competitive effects. Mustard (Sinapis alba) can be planted after the fall harvest, and killed in the winter by freezing in the northern United States. An experiment was conducted to determine whether soil is allelopathic after the growth of winter-killed mustard. The experimental design took account of important methodological issues for the design of allelopathy bioassays. Experimental designs that can address such methodological problems will be reviewed. In this study, mustard was grown for six weeks in the greenhouse and killed by freezing. Tops were removed prior to establishing green foxtail (Setaria viridis) at densities of 1, 3, 10, 30 and 100 plants per 25 cm diameter pot. Other treatment variables included amending half of the pots with activated carbon to reduce the activity of allelopathic compounds in soil, and two fertility levels. Dry shoot biomass of green foxtail was measured. No clear evidence of density-dependent allelopathic effects, or of allelopathic effects reduced by use of activated carbon, was observed in this study. Germination of green foxtail was reduced by mustard residues.

Media summary

Experiments have been conducted to evaluate the allelopathic potential of mustard as a cover crop in reduced tillage systems.

Key Words

Cover crops, allelopathy, Sinapis alba, Setaria viridis, methodology, no-tillage.


Reduced tillage systems offer many advantages from an ecological perspective. Increased infiltration of water into the soil reduces runoff and a residue cover conserves soil moisture. Organic matter is increased, and soil erosion is greatly reduced. In one study in Coshocton, Ohio, soil loss over a four-year period averaged 7 kg/ha on no-till fields with a 9% slope, and greater than 5000 kg/ha on conventionally tilled fields with a 6% slope (Edwards et al. 1993). However, a disadvantage of reduced tillage systems is the requirement that herbicides be used to control weeds. These herbicides can be transported offsite by surface runoff or be leached into groundwater (Logan et al. 1991). Herbicides comprise 85% of all pesticides applied to crops in the United States, and actual amounts used have increased five-fold since 1964.

One approach to reducing herbicide use while preserving the advantages of reduced tillage is to use cover crops to provide biological control of weeds. The approach has been successfully developed in Brazil using black oat (Avena strigosa) as a cover crop prior to planting corn (Santi et al. 2003). Another group of plants with potential as allelopathic cover crops are those in the Brassicaceae family that contain high concentrations of glucosinolates. Glucosinolate hydrolysis products, particularly isothiocyanates, have broad biocidal activity, including herbicidal activity (Brown and Morra 1997). Recent greenhouse studies (Holtz 2001) showed that certain Brassicas release sufficient quantities of allelochemicals to completely inhibit green foxtail (Setaria viridis) and redroot pigweed (Amaranthus retroflexus) germination. One possible approach to incorporating a Brassica cover crop in a temperate climate would be to plant it following the harvest of the summer crop (corn or soybeans). The cover crop would then grow for several weeks before being killed by the winter freeze. In the spring, the next crop would be planted into the cover crop residue.

The objective of this experiment was to determine whether soil is allelopathic following growth of winter-killed mustard (Sinapis alba L.) based on density-dependent growth responses of assay species. The use of activated carbon was compared to density-dependent growth responses as a means of distinguishing allelopathic effects on plant growth.


Variables in this experiment included:

(a) Presence of mustard residues. S. alba was grown in half of the pots, while half of the pots were initially unplanted.

(b) Presence of activated carbon. Half of the pots were amended with activated carbon and half were left untreated. Activated carbon has been shown to reduce the activity of allelopathic compounds in soil (Nilsson, 1994; Ridenour and Callaway, 2001).

(c) Addition of nutrients. Half of the pots were fertilized after the assay species was planted. This treatment was included because of the potential interaction of soil nutrient levels with microbial degradation of compounds from mustard (Blum et al. 1999).

(d) Assay species density. Green foxtail (Setaria viridis) was chosen as the assay species based on previous work in our laboratory (Holtz, 2001) which demonstrated sensitivity of this species to Brassica spp. In order to detect any density-dependent phytotoxic effects that would be indicative of the presence of allelopathic toxins in the soil (Weidenhamer et al. 1989), S. viridis was seeded at rates of 10-150 seeds per pot in order to achieve target densities of 1, 3, 10, 30 and 100 seedlings per pot.

Growth of cover crop

Plants were grown in 11.4-L plastic nursery pots containing 11 kg of Wooster silt loam soil. For the pots containing activated carbon, enough was added to give 2% activated carbon by volume. Twenty-five Sinapis alba cv. UI 7012 seeds were planted at a depth of approximately 1cm. Pots were thinned to four plants per pot after emergence. Plants were maintained in a water-cooled greenhouse, and watered evenly as needed. Within blocks, pots were rotated weekly. After 40 days, mustard plants had reached a height of 100-125 cm, and the growth of plants was terminated by moving pots to a large walk-in freezer for several days. Unplanted pots were included in the experimental design during the initial phase of the experiment, were watered as needed and were taken to the freezer with the mustard pots.


Upon removal of the pots from the freezer, mustard tops were removed by clipping, while roots remained undisturbed in the soil. All pots were planted with S. viridis within 24 hours of removal from the freezer. Half of the pots were fertilized eight days after planting with a soluble 20-20-20 fertilizer which included micronutrients. Lower density pots were thinned to the target densities of 1, 3 and 10 seedlings per pot. Germination of higher density pots (seeded with 45 and 150 seeds per pot respectively) was determined at harvest for a measure of germination percentage. Foxtail shoots were harvested after 21 days, and shoot dry mass was determined. There were three replicates of each treatment.



The lower densities of green foxtail (1, 3, and 10 seedlings per pot) were thinned to the target densities as seedlings emerged. At the higher target densities (30 and 100 seedlings per pot), no thinning was done and actual emergence was counted at harvest. Emergence in the mustard treatments was reduced compared to controls. For the 30 seedling per pot target density, actual emergence averaged 29.5 seedlings per pot (89.4% + 4.7% of the corresponding controls). For the 100 seedling per pot target density, actual emergence was 87.2 seedlings per pot (80.2% + 2.8% of the corresponding controls).


Density-dependent phytotoxicity is a result of plant competition for phytotoxic allelochemicals, but because the substance being competed for is not beneficial for the plant, a plant that takes up a larger quantity suffers reduced growth as a result. Density-dependent phytotoxic effects are expected to be most pronounced at the lowest plant densities, because these plants receive the highest doses of toxins. Plant size can actually increase as plant density increases, a result that is contrary to the expected effects of resource competition. Demonstration of either of the following results is contrary to the expected results of resource competition and supports the hypothesis of the presence of a toxin(s) in soil: (a) Compared to a control soil, growth reductions occur at low but are diminished at high density; and (b) Maximum individual plant weight occurs at an intermediate density, due to a reversal in slope of the predicted log yield−log density line. In some cases, reduction rather than reversal of the slope of this line will be seen (Weidenhamer et al. 1989). While analysis of the data is continuing, no clear allelopathic effects on growth of foxtail were apparent in this study. Representative data are shown in Figure 1.

Figure 1. Log of individual plant biomass plotted as a function of actual emergence density in mustard and control pots that were not fertilized and did not have activated carbon added. Bars show standard errors.


The lack of apparent effect on green foxtail growth by Sinapis alba root residues was surprising. In previous work (Dick, unpublished results), the same cultivar of S. alba inhibited subsequent growth of green foxtail in the field when it was planted in the spring and mowed to ground level at the end of June.

The experimental conditions for this experiment were chosen to mimic a somewhat different agronomic practice – planting mustard as a fall cover crop and allowing it to be killed by over-wintering. The methods were designed to conserve the allelopathic potential which would have developed in the soil, by planting foxtail immediately after thawing the pots and before substantial degradation occurred, and by not over-watering the pots and leaching any toxins from the soil. However, the tops of plants were removed, and not mulched onto the soil surface. Both shoots and roots of Brassica species are sources of isothiocyanates, the suspected allelopathic agents in S. alba, and isothiocyanates (ITCs) have been shown to have very short lifetimes once released in soil (Petersen et al. 2001). Also, the profile of different glucosinolates (the source of ITCs) is different between shoots and roots. It may be that the release of inhibitory compounds from shoots mulched on the soil surface is crucial to the allelopathic effects of this cover crop. Because strong allelopathic effects were not observed in this study, it was not possible to compare the use of activated carbon to the variation of plant density as a means of detecting allelopathic effects.


Brown PD and Morra MJ (1997). Control of soil-borne plant pests using glucosinolate-containing plants. Advances in Agronomy 61, 167-231.

Blum U, Shafer SR and Lehman ME (1999). Evidence for inhibitory allelopathic interactions involving phenolic acids in field soils: Concepts vs. an experimental model. Critical Reviews in Plant Sciences 18, 673-693.

Edwards WM, Triplett GB, Van Doren EM, Owens LB, Edwards CE and Dick WA (1993). Tillage studies with a corn/soybean rotation: Hydrology and sediment loss. Soil Science Society of America Journal 57, 1051-1055.

Holtz GP (2001). Brassica cover crops for biological weed control in no-tillage. Ph.D. diss. Ohio State Univ., Columbus.

Logan TJ, Lal R and Dick WA (1991). Tillage systems and soil properties in North America. Soil and Tillage Research 20, 241-270.

Nilsson, M-C (1994). Separation of allelopathy and resource competition by the boreal dwarf shrub Empetrum hermaphroditum Hagerup. Oecologia 98, 1-7.

Ridenour WM and Callaway RM (2001). The relative importance of allelopathy in interference: the effects of an invasive weed on a native bunchgrass. Oecologia 126, 444-450.

Santi A, Amado TJC and Acosta JAA (2003). Black oat biomass and nutrient cycling as affected by nitrogen fertilization in soil under notillage. Revista Brasiliera de Ciencia do Solo 27, 1075-1083.

Weidenhamer JD, Hartnett DC and Romeo JT (1989). Density-dependent phytotoxicity: Distinguishing resource competition and allelopathic interference in plants. Journal of Applied Ecology 26, 613-624.

Previous PageTop Of PageNext Page