A set of universal and degenerate primers has been designed to clone (by polymerase chain reaction [PCR]) the conserved domains of the acetolactate synthase (ALS) gene where mutations confer resistance to ALS herbicides in plants. These primers were successful in cloning conserved domains of ALS in all monocotyledonous and dicotyledonous plants tested to date, as well as that of bacteria. Total genomic DNA was used as the source of target DNA because no introns were found in the sequences to be amplified. The design of the universal primers was performed after subtle modifications of the consensus degenerate hybrid oligonucleotide primers approach, which implies the synthesis of hybrid oligonucleotide primers containing fixed clamp 5′ and degenerate core 3′ sequences. Optimizations of PCR reactions were done according to a Taguchy approach described for the first time with degenerate oligonucleotides. This method optimizes a PCR reaction using four variables (deoxynucleoside triphosphate, DNA, primers, and Mg2+) under three different concentrations per variable using just nine reactions. The ALS herbicide-binding domains from many susceptible and resistant plants can be cloned and sequenced in a few hours by using only 100 mg of starting plant material, like one leaf or several small seedlings or seeds.

Leafy spurge can be detected during flowering with either aerial photography or hyperspectral remote sensing because of the distinctive yellow-green color of the flower bracts. The spectral characteristics of flower bracts and leaves were compared with pigment concentrations to determine the physiological basis of the remote sensing signature. Compared with leaves of leafy spurge, flower bracts had lower reflectance at blue wavelengths (400 to 500 nm), greater reflectance at green, yellow, and orange wavelengths (525 to 650 nm), and approximately equal reflectances at 680 nm (red) and at near-infrared wavelengths (725 to 850 nm). Pigments from leaves and flower bracts were extracted in dimethyl sulfoxide, and the pigment concentrations were determined spectrophotometrically. Carotenoid pigments were identified using high-performance liquid chromatography. Flower bracts had 84% less chlorophyll a, 82% less chlorophyll b, and 44% less total carotenoids than leaves, thus absorptance by the flower bracts should be less and the reflectance should be greater at blue and red wavelengths. The carotenoid to chlorophyll ratio of the flower bracts was approximately 1:1, explaining the hue of the flower bracts but not the value of reflectance. The primary carotenoids were lutein, β-carotene, and β-cryptoxanthin in a 3.7:1.5:1 ratio for flower bracts and in a 4.8:1.3:1 ratio for leaves, respectively. There was 10.2 μg g−1 fresh weight of colorless phytofluene present in the flower bracts and none in the leaves. The fluorescence spectrum indicated high blue, red, and far-red emission for leaves compared with flower bracts. Fluorescent emissions from leaves may contribute to the higher apparent leaf reflectance in the blue and red wavelength regions. The spectral characteristics of leafy spurge are important for constructing a well-documented spectral library that could be used with hyperspectral remote sensing.

The mechanism of glyphosate resistance in horseweed was investigated. Eleven biotypes of putative sensitive (S) and resistant (R) horseweed were obtained from regions across the United States and examined for foliar retention, absorption, translocation, and metabolism of glyphosate. Initial studies used spray application of 14C-glyphosate to simulate field application. When S and R biotypes were compared in the absence of toxicity at a sublethal dose, we observed comparable retention and absorption but reduced root translocation in the R biotypes. S and R biotypes from Delaware were further examined at field use rates and results confirmed similar retention and absorption but reduced root translocation in the R biotypes. Application of 14C-glyphosate to a single leaf demonstrated reduced export out of the treated leaf and lower glyphosate import into other leaves, the roots, and the crown in R relative to S biotypes. Examination of the treated leaf by autoradiography showed that glyphosate loading into the apoplast and phloem was delayed and reduced in the R biotype. Our results consistently showed a strong correlation between impaired glyphosate translocation and resistance. Tissues from both S and R biotypes showed elevated levels of shikimate suggesting that 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) remained sensitive to glyphosate. Analysis of tissue shikimate levels demonstrated reduced efficiency in EPSPS inhibition in the R biotypes. Our results suggest that resistance is likely due to altered cellular distribution that impaired phloem loading and plastidic import of glyphosate resulting in reduced overall translocation as well as inhibition of EPSPS.

A biotype of green foxtail found in Spain exhibited cross-resistance among acetyl-CoA carboxylase (ACCase)–inhibiting herbicides. Field doses that totally inhibited shoot fresh weight in the susceptible (S) biotype were determined for six aryloxyphenoxypropionates (clodinafop, diclofop, fenoxaprop-P, fluazifop-P, haloxyfop-P, and propaquizafop) and six cyclohexanediones (clefoxydim, clethodim, cycloxydim, sethoxydim, tepraloxydim, and tralkoxydim). The resistant (R) biotype showed cross-resistance to all herbicides except fenoxaprop-P, propaquizafop, clefoxydim, and tepraloxydim. There were no differences in the absorption, translocation, and metabolism of [14C]diclofop between the S and R biotypes. On the basis of herbicide dose that inhibited ACCase activity by 50% (I50 values), ACCase of the R biotype was 5.8-, 13.9-, 20.0-, 102.4-, 416.7-, and 625.0-fold less sensitive to clethodim, haloxyfop, diclofop, fluazifop, cycloxydim, and sethoxydim, respectively, than that of the S biotype. Two multifunctional ACCase isoforms (ACCase I and ACCase II) were purified partially and separated. ACCase II was highly resistant to diclofop acid in both biotypes, with I50 values ranging between 92 and 95 μM. However, the I50 values observed for ACCase I revealed that the R biotype was 30.8-fold less sensitive to diclofop than the S biotype. These results suggest the mechanism of resistance in green foxtail to diclofop relates to an altered ACCase I isoform.

Greenhouse and laboratory studies were conducted to examine the activity and foliar absorption of foramsulfuron in giant foxtail and woolly cupgrass with various adjuvants. Adjuvant selection was important for giant foxtail control. Foramsulfuron provided 90% or greater giant foxtail control with the addition of methylated seed oil (MSO) or MSO plus 28% urea ammonium nitrate (UAN). When a crop oil concentrate (COC) or a nonionic surfactant (NIS) was added to foramsulfuron, giant foxtail control was only 20%. However, when 28% UAN was added to COC or NIS, control was increased to 90 and 85%, respectively. Foramsulfuron absorption and control were closely related in giant foxtail. Foliar absorption of 14C-foramsulfuron in giant foxtail ranged between 35 and 90% 24 h after treatment (HAT) depending on adjuvant selection. The rate of absorption was greatest when MSO plus 28% UAN was added to foramsulfuron and absorption was maximized 4 HAT. Foramsulfuron absorption in woolly cupgrass reached its maximum levels 2 HAT with all adjuvant combinations. Although the rate of foramsulfuron absorption was quicker in woolly cupgrass, absorption trends by adjuvants were similar to those in giant foxtail. Maximum absorption of 14C-foramsulfuron in woolly cupgrass was 84% with the addition of MSO plus 28% UAN. However, even with high levels of absorption, woolly cupgrass control with foramsulfuron was poor and may be related to rapid metabolism to nonphytotoxic compounds.

Downy brome in dryland winter wheat presents a major constraint to the adoption of reduced tillage cropping systems in the Pacific Northwest of the United States. Effective suppression of downy brome during fallow periods depletes seed in the soil and reduces infestations in subsequent winter wheat crops. Delayed tillage operations or delayed herbicide applications in the spring increase the risk for production of viable downy brome seed during fallow periods. In a series of studies, downy brome panicles were sequentially sampled at Pendleton, OR, and Pullman, WA, in 1996 and 1997, and at nine locations around the winter wheat growing region of the western United States in 1999 and 2001. Cumulative growing degree days (GDD) were calculated using local, daily maximum, and minimum air temperature data. A simple GDD model based on the formula GDD = (daily maximum temperature [C] + daily minimum temperature [C])/2, with a base temperature of 0 C and a starting point of January 1, was used to calculate cumulative GDD values for panicle sampling dates. Number of seed germinating per collected panicle was recorded in greenhouse germination tests. Estimations of degree days required for production of viable downy brome seed were made using nonlinear regression of germination on GDD. The GDD value at which viable seed can be found on plants (i.e., when seed germination > 0) was of interest. Estimates of the GDD values at which viable seed could be found in the three studies ranged from 582 GDD at Bozeman, MT, to 1,287 GDD at Stillwater, OK, with a group of GDD values for Pendleton and Pullman around 1,000. Variation in seed-set GDD among locations may be attributed to differing climatic conditions that control vernalization at the various locations or to differences in vernalization requirements among downy brome biotypes (or both).

Jubatagrass is one of the most invasive nonnative species along sensitive natural coastal sites of California. This study was designed to understand the biology of reproduction and seed longevity under field conditions. Jubatagrass can produce over 100,000 wind-dispersed seeds from a single inflorescence. Seeds are produced apomictically, and germination is directly related to seed size. Of the total seeds produced, only 20 to 30% were of ample size to readily germinate when exposed to light and under a temperature range similar to coastal environments. Seeds not exposed to light also germinated but at about 30% the level of light-exposed seeds. This suggests that exposed disturbed coastal sites with moderate temperatures have high potential for germination and establishment of jubatagrass. The percentages of germinable and viable seeds were not significantly different, indicating that jubatagrass does not have a primary dormancy. This was supported by field experiments demonstrating that seeds do not persist under natural conditions for more than 6 mo. These results indicate that an intensive 1-yr control program targeting established seedlings and mature plants should sufficiently manage existing populations. However, effective long-term management of jubatagrass must focus on anticipating environments susceptible to invasion, reducing new seed recruitment, and preventing subsequent seed germination and seedling establishment.

Artificial weed seed banks are practical for studying seed bank depletion and weed seedling emergence because the number, depth, and species composition of seed banks can be managed. However, no studies have determined whether artificial seed banks are representative of natural seed banks. We compared the emergence of velvetleaf, giant foxtail, and common waterhemp in a natural seed bank, an artificial seed bank with stratified seeds, and an artificial seed bank with nonstratified seeds. Velvetleaf seedling emergence was higher in the nonstratified seed bank in 2001, but no differences were observed in 2002. The number of viable velvetleaf seeds at the end of the experiment was lower in the natural seed bank than in the artificial seed banks in 2002. Velvetleaf emergence occurred earlier in the natural seed bank than in the artificial seed banks. Giant foxtail emergence was higher in the artificial seed banks (58 to 82%) than in the natural seed bank (5 to 23%). Common waterhemp emergence ranged from 7 to 65% in the artificial seed banks and from 1 to 5% in the natural seed bank. In general, the distribution of emergence with time differed in the natural seed bank compared with the artificial seed banks. These differences were attributed to differences in soil temperature and soil bulk density between the natural and artificial seed banks. Artificial seed banks showed lower soil bulk density and greater temperature fluctuation than the natural seed bank. However, there was no consistent relationship between growing degree days and emergence timing in the three treatments for any of the species studied.

Weed biotypes putatively resistant to acetolactate synthase (ALS)–inhibiting herbicides were reported by Iowa farmers from 1997 to 2001. Greenhouse studies confirmed cross-resistance to triazolopyrimidine sulfonanilide and sulfonylurea (SU) herbicides in giant ragweed from Scott County, IA (Werner Farm), which corresponded to resistance to susceptibility (R:S) GR50 (50% growth reduction) ratios of 21 and 28 to cloransulam and primisulfuron + prosulfuron, respectively. At the enzyme level, this represented a 49- and 20-fold I 50 (50% enzyme inhibition) increase. Cross-resistance to imidazolinone (IMI) and SU herbicides was also observed in common sunflower from Cherokee, IA. Compared with a susceptible biotype, the resistant common sunflower biotype demonstrated GR50 R:S ratios of 36 and 43 to imazethapyr and chlorimuron, respectively. Shattercane from Malvern, IA, was susceptible to nicosulfuron but was resistant to imazethapyr (GR50 R:S ratio = 29). The woolly cupgrass biotypes from Union County, IA (Pettit Farm and Travis Farm), were reportedly resistant but were identified susceptible to both IMI and SU herbicides. Using an in vivo ALS assay, extractable endogenous 2,3-diketone concentrations ranged from 25 to 71 nmol g−1 fresh weight for all species. Compared with susceptible biotypes, 2,3-diketone levels accumulated to at least twofold higher levels in treated resistant plants 120 h after herbicide application. Field history data suggested that resistance evolved independently in three environments where ALS-inhibiting herbicides represented an important component of the selection pressure.

Horseweed is a winter annual weed that has evolved resistance to multiple herbicide modes of action in 11 countries worldwide. A paraquat-resistant horseweed population in an Ontario orchard that was being managed by a rotation of herbicides began to show increased tolerance to the herbicide linuron. Experiments were conducted in the greenhouse to compare the response of this population and several paraquat-susceptible populations to linuron, diquat, and oxyfluorfen. Plants were sprayed with a range of doses of each herbicide when they were from 5 to 10 wk old, and the ED50, or dose at which shoot dry weight was reduced by 50%, was estimated. There was a sevenfold difference in the ED50 values of the paraquat-resistant and -susceptible populations in response to diquat and a threefold difference in response to linuron. The response to oxyfluorfen was age dependent. The ratio of resistance to susceptible ED50 values was estimated as 57 for 5-wk-old plants and 11 for 8-wk-old plants in response to oxyfluorfen. Ten-week-old plants from both populations showed no response to oxyfluorfen at rates up to 4.8 kg ai ha−1.

Three Italian Lolium weed populations, one susceptible and two resistant to diclofop, were characterized by the technique of inter simple sequence repeats (ISSR). The goal of this study was to taxonomically identify these Lolium populations as well as to evaluate evidence for introgression of ISSR fragments from Festuca and the potential role of this introgression in the diclofop response. ISSR analysis confirmed the genomic background of the weed populations to be consistent with that of Lolium. However, the great range of variation in ISSR banding patterns highlighted that the three ryegrass accessions are mixed populations made up of individuals resulting presumably from intrageneric and intergeneric hybridization in the Lolium–Festuca complex. Two Festuca genus-discriminating and 20 Festuca species-discriminating ISSR markers were screened among all the three ryegrass populations. The resistant Tuscania population carried the highest percentage of Festuca genome (16.8%) followed by the resistant Roma (13.6%) and susceptible Vetralla (7.6%) populations. On the basis of these data some influence of Festuca genome in diclofop resistance levels of studied ryegrass populations could be hypothesized.

The inheritance of glyphosate-resistant goosegrass was studied by making reciprocal crosses between resistant (R) and susceptible (S) biotypes. Eighty-four F1 hybrids were confirmed using isozyme analyses. Reciprocal F1 hybrids displayed uniform levels of resistance intermediate (I) to that of the parental types, with no indication of maternal inheritance. The F1 hybrids were selfed to produce F2 populations. F3 populations were produced by selfing resistant and intermediate F2 phenotypes. A segregation ratio of 1:2:1 (S:I:R) was observed in the F2 and subsequently in the F3 generations derived from selfing intermediate F2 phenotypes. Individuals in F3 families derived from selfing resistant F2 phenotypes were resistant. Results from the present study show that glyphosate resistance in goosegrass is inherited as a single, nuclear, and incompletely dominant gene.

The emergence of annual species depends on the number of seeds present and the biotic and abiotic conditions directly surrounding those seeds (the microsite). A field experiment was conducted to study the relative importance of seed presence vs. microsite conditions in determining the emergence of four annual species. Green foxtail, wild mustard, wild oat, and canola were seeded at 200, 400, and 1,200 seeds m−2 in separate plots in a coarse, loamy, mixed Typic Haplocryoll and a fine, mixed Typic Haplocryoll soil. Five microsite modification treatments (control, irrigation, soil compaction, soil compaction plus irrigation, and no crop) were applied to all weed seed density treatments for each weed species. All plots were seeded to spring wheat. Irrigation or soil compaction increased percent emergence of wild oat. Green foxtail emergence tended to increase with soil compaction in 2001 but not in 2002. Wild mustard and canola emergence were largely unaffected by microsite modification treatments. Weed emergence increased with increasing seed density for all species, but the proportion of the total number of seeds emerging decreased with increasing seed density for all species. We suggest that the emergence of the four weed species in this experiment was both seed and microsite limited. Increasing the number of seeds in the soil increased the probability of seeds landing within an appropriate microsite. For these four species, therefore, weed spread and weed patch formation may be determined both by seed production and dispersal and by variability in soil microsite conditions. Results suggest that weed management practices should limit seed dispersal of all species and discourage weed emergence of hard-to-control species during critical establishment periods.

The nitrogen (N) response of competing plants may be affected by photosynthetically active radiation (PAR) availability and maximum potential growth rate, which determine N requirements. The responses of two crop (corn and soybean) and six weed species (common lambsquarters, common waterhemp, giant foxtail, velvetleaf, wild mustard, and woolly cupgrass) in low and high (150 and 450 μmol m−2 s−1) PAR levels to daily fertilization with either low or high (0.2 or 7.5 mM) NH4NO3 levels were studied. Leaf area of all species responded positively to N by 8 d after emergence (DAE) when grown in high PAR; in low PAR, most species did not respond until 11 DAE. Dry weight and leaf area of all species at 18 DAE were greater with high than with low N. These responses to high N were also greater in high than in low PAR for all species. Dry weights with high N were up to 100% greater in low PAR and up to 700% greater in high PAR than dry weights with low N. These responses suggest that low PAR reduced the benefit of N to the plants. The regression of relative growth rate (RGR) with high N to RGR with low N had a slope that was less than unity (β = 0.79), indicating that species with a higher RGR with high N experienced greater decreases in RGR with low N. Similarly, the sensitivity (change in RGR) of plants grown with high and low N was positively related to RGR with high N. RGR differences among crop and weed species may be related to differences in N requirement that could be exploited for weed management. RGR and seed size were negatively correlated, which may explain previous observations that small-seeded weeds were more sensitive to environmental stress.

Canada thistle was grown under field conditions in 2000 and 2003 at ambient and elevated (∼ 350 μmol mol−1 above ambient) carbon dioxide [CO2] to assess how rising [CO2] alters growth, biomass allocation, and efficacy of the postemergent herbicide glyphosate. By the time of glyphosate application, approximately 2 mo after emergence, elevated CO2 had resulted in significant increases in both root and shoot biomass. However, the relative positive effect of [CO2] was much larger for root, relative to shoot growth, during this period (2.5- to 3.3-fold vs. 1.2- to 1.4-fold, respectively) with a subsequent increase in root to shoot ratio. Glyphosate was applied at 2.24 kg ae ha−1 in 2000 and 2003. Subjective classification of leaf damage in shoots after spraying indicated no significant difference in the extent of necrosis in aboveground tissue as a function of CO2 concentration. After a 6-wk regrowth period, significant reductions in shoot and root biomass relative to unsprayed plots were observed under ambient [CO2]. However, the decrease in the ratio of sprayed to unsprayed biomass was significantly less at elevated relative to ambient [CO2] conditions for roots in both years, and no difference in shoot biomass was observed between sprayed and unsprayed plots for Canada thistle grown at elevated [CO2] in either year. The observed reduction in glyphosate efficacy at the enriched [CO2] treatment did not appear to be associated with differential herbicide uptake, suggesting that tolerance was simply a dilution effect, related to the large stimulation of root relative to shoot biomass at elevated [CO2]. Overall, the study indicates that carbon dioxide–induced increases in root biomass could make Canada thistle and other perennial weeds that reproduce asexually from belowground organs harder to control in a higher [CO2] world.

Laboratory experiments were conducted to determine the effects of oxygen concentration (21, 10, 5, and 2.5%), exposure to light, and osmotic potential on the germination of wheat, canola, and various weed species. Germination of most species increased as osmotic potential was increased. Seed germination for some species like barnyardgrass was inhibited by the combination of exposure to normoxic (21% oxygen) conditions and light. This combination of conditions may function as a signal to prevent soil surface germination. Wild mustard germination increased with increasing oxygen concentration when seeds were not exposed to light, whereas green foxtail germination was relatively insensitive to oxygen concentration. Wild oat germination increased with increasing osmotic potential, and osmotic potential had a greater influence when the seeds were exposed to light. Dandelion, foxtail barley, curly dock, and perennial sowthistle germination was affected more by osmotic potential and light exposure than by oxygen concentration. A better understanding of the mechanisms of depth detection for specific species will lead to a better understanding of their recruitment biology. This information may help model the potential for invasion and proliferation of each species as well as devise improved management strategies.

No-tillage offers potential for improved soil quality, reduced erosion, and equal or increased crop yields. We hypothesized that, compared with conservation tillage (CT), no-tillage (NT) offers conditions more conducive to microbial decay of weed seed. In NT systems seed remain at or near the soil surface where crop residues, moisture, and lack of disturbance create an environment with greater soil microbial diversity. In late fall of 1998 and 1999, dormant seed of wild oat, either individually glued to plastic toothpicks or mixed with soil and placed in mesh bags, were buried (mean seed depth of 2.5 cm) in replicated field plots managed by NT or CT since 1982. Treatments including fungicide seed treatment (thiram + metalaxyl + captan) and soil fumigation (propylene oxide) provided estimates of the contribution of microorganisms to observed mortality. Seed were retrieved in May and August, 1999 and 2000. Contrary to our original hypothesis, the proportion of dead seed was generally similar in NT and CT systems. Lack of tillage system by seed or soil treatments affecting the proportion of dead or decayed seed suggests that the contribution of microorganisms to seed fate is similar in these tillage environments. However, the proportion of dormant seed was consistently lower in the NT compared with CT treatments; there was a corresponding increase in the proportion of germinated seed. Overall, more than half of the wild oat seed bank losses could be directly attributed to germination whereas losses due to decay were relatively minor by comparison. Despite favorable distribution of seed and improved quality of the surface-strata of soil in NT systems, this study fails to provide evidence that enhanced microbial decay will contribute to a “weed-suppressive” capacity in such cropping systems.

Use of composted swine manure produced in deep-bedded hoop structures is a promising approach for recycling farm waste products and improving soil fertility, but little is known about its effects on crop–weed interactions. A 2-yr study was conducted to evaluate the effect of compost amendments and tillage on soybean–common waterhemp competition. Experiments were conducted in no-tillage and chisel plow main plots with compost applied to one of two types of subplots. Common waterhemp and soybean growth was measured in sub-subplots accommodating weed-free soybean and soybean with common waterhemp sown at soybean planting, soybean emergence (VE), soybean second-node stage (V2), and soybean sixth-node stage (V6). Soybean heights were not influenced by compost or common waterhemp sowing time. Soybean stem diameters were influenced by year, tillage regime, and an interaction between compost and common waterhemp sowing time. In contrast, common waterhemp heights and basal diameters were greater when sown at planting and VE in compost-amended subplots than in compost-free subplots. Overall, there was a negative quadratic relationship between common waterhemp biomass and soybean yield (r 2 = 0.746). The extremely low common waterhemp emergence in V2 and V6 treatments suggested that early-season weed suppression was sufficient to protect soybean from common waterhemp competition. The sex determination of 2,557 common waterhemp plants showed a marginally higher male to female ratio in compost-amended treatments than in compost-free treatments (P = 0.0611). A linear-slope regression indicated that common waterhemp fecundity was positively related to individual plant biomass, with a change in slope occurring at 118.7 g. Under the conditions present in this experiment, compost did not enhance soybean yield but increased the competitive ability of waterhemp. Because composted swine manure can have a major influence on competition of common waterhemp with soybean, effective weed management practices should be in place when this soil amendment is used.

Managing crop fertilization may be an important component of integrated weed management systems that protect crop yield and reduce weed populations over time. A field study was conducted to determine the effects of various timings and application methods of nitrogen (N) fertilizer on weed growth and spring wheat yield. Nitrogen fertilizer was applied the previous fall (October) or at planting (May) at a dose of 50 kg ha−1. Nitrogen application treatments consisted of granular ammonium nitrate applied broadcast on the soil surface, banded 10 cm deep between every crop row, banded 10 cm deep between every second crop row, or point-injected liquid ammonium nitrate placed between every second crop row at 20-cm intervals and 10 cm deep. Treatments were applied in 4 consecutive yr to determine annual and cumulative effects over years. Density and biomass of wild oat, green foxtail, wild mustard, and common lambsquarters were sometimes lower with spring- than with fall-applied N. Spring wheat yield was never lower and was higher in 50% of the cases, when N was spring rather than fall applied. Nitrogen application method generally had larger and more consistent effects than application timing on weed growth and wheat yield. Shoot N concentration and biomass of weeds were often lower with subsurface banded or point-injected N than with surface broadcast N, and concurrent increases in spring wheat yield usually occurred with these N placement treatments. Depending on the weed species, the weed seedbank at the conclusion of the 4-yr study was reduced by 25 to 63% with point-injected compared with broadcast N. Information gained in this study will contribute to the development of more integrated and cost-effective weed management programs in wheat.

Myrothecium verrucaria, isolated from sicklepod, was investigated for bioherbicide potential against a wide range of economically important weed species from agronomic, pasture, and horticultural systems. A number of different weed species from a range of plant families were highly susceptible to sprays of crude preparations of the fungus. A small number of species, primarily monocots, were tolerant, showing no damage symptoms or insignificant effects on biomass. Symptoms developed very rapidly in susceptible hosts, suggesting the activity of toxins, several of which are known to be produced by Myrothecium spp. The activity of crude harvests of M. verrucaria was not diminished when they were filtered to remove fungal mycelium and spores. In contrast, washed conidia had relatively little impact on weed species. We conclude that the activity of M. verrucaria is primarily caused by the activity of metabolites produced by the fungus in culture and not due to infection by the fungus per se. Myrothecium verrucaria cultural preparations have extremely potent herbicidal properties, but given its ability to produce mammalian toxins, we caution against its use.