Threespike goosegrass (Eleusine tristachya) is a difficult-to-control perennial grass of increasing concern for orchard production systems in the Central Valley of California, USA. This grass has a bunch-type growth habit when tillered, which can interfere with orchard operations, particularly nut pickup from the ground at harvest. From 2016 to 2019, herbicide efficacy on threespike goosegrass was evaluated in a walnut (Juglans regia) orchard in Chico, CA, USA; an almond (Prunus dulcis) orchard in Livingston, CA, USA; and a prune (Prunus domestica) orchard in Orland, CA, USA. At each location, two independent experiments were conducted to evaluate 12 preemergent (PRE) herbicide treatments and eight postemergent (POST) treatments over several years, for a total of 16 trials. PRE herbicides were applied in January according to the region’s typical winter orchard management practices. One treatment included an additional sequential application in March to extend residual activity later into the warm season when threespike goosegrass germinates or resumes growth. In separate studies, POST control of established stands of threespike goosegrass was evaluated in May and June of each year. Each trial was conducted in a randomized complete block design with four replications. Threespike goosegrass control was visually estimated monthly for 5 months after the PRE treatments or at weekly intervals for 5 weeks following POST treatments. The most effective PRE treatment was a sequential application of indaziflam in January, followed by a March application of pendimethalin, providing 90% or greater control of threespike goosegrass 5 months after treatment across all sites and all years. Of the POST treatments, the three graminicides outperformed the other treatments with 73% to 91% control overall sites and years at 5 weeks after treatment. Fluazifop had the highest control ratings (85% to 91%) among the graminicide herbicides but was not always statistically better than clethodim or sethoxydim (74% to 83% control). Glyphosate alone resulted in unacceptable control (33% to 51%) regardless of rates tested, experimental sites, or years. Together, these results confirm grower reports of poor glyphosate performance on threespike goosegrass but suggest that effective herbicide programs can be developed to manage threespike goosegrass using PRE herbicides and POST graminicides registered in California orchard crops.

Weed management in cantaloupe and other melon crops is important to maximize fruit yield; however, there are few registered herbicides available in California. Several independent herbicide trials were conducted at University of California field stations in Davis (Yolo County), Five Points (Fresno County), and Holtville (Imperial County) from 2013 to 2019 to evaluate both registered and unregistered herbicides and incorporation methods (sprinklers, cultivation, or none) for crop safety and weed control in melons. Although specific treatments varied among locations depending on local practice and research objectives, ethalfluralin and halosulfuron were used in all experiments, and bensulide and S -metolachlor were evaluated in 4 of 6 site-years. Additional herbicides included clethodim, clomazone, DCPA, napropamide, pendimethalin, sethoxydim, and sulfentrazone. Among registered herbicides, halosulfuron, halosulfuron + ethalfluralin, and ethalfluralin + bensulide combinations provided consistently beneficial weed control across all site-years compared to the nontreated control. S -metolachlor performed as well as the best of the registered herbicides tested at each site-year; although moderate injury was noted at the Davis location, this did not reduce melon yield. The method used to incorporate preplant herbicides had a significant impact on weed control efficacy but varied by location. Mechanical incorporation of preplant herbicides resulted in improved weed control and yield compared to sprinklers. Early-season weed control, whether by herbicides or hand weeding, resulted in significant yield increase in most site-years.

The nonnative Brazilian egeria (Egeria densa Planch.) is the dominant submersed plant in the Sacramento-San Joaquin River Delta, displacing native plant species and degrading habitat for endangered fish species. A mesocosm study was conducted at the U.S. Department of Agriculture (USDA) facility in Davis, CA to determine efficacy of aquatic herbicides on egeria. Fifty mesocosm tanks of 167 L capacity were planted with four 4.2-L pots of egeria. Four tanks each were treated with bispyribac sodium (45 mu g L-1), carfentrazone-ethyl (200 mu g L-1), ethylenediamine chelate of copper (1,000 mu g L-1), diquat (390 mu g L-1), dipotassium salt of endothall (5,000 mu g L-1), dimethylalkylamine salt of endothall (5,000 mu g L-1), florpyrauxifen-benzyl (50 mu g L-1), flumioxazin (400 mu g L-1), fluridone (60 mu g L-1), imazamox (500 mu g L-1), penoxsulam (60 mu g L-1), and four tanks were left as an untreated reference. All were single treatments, static exposures for 10 wk. At the end of 10 wk, all pots were harvested, and the shoots were dried at 70 C for 48 h. All herbicides produced some statistically significant reduction in biomass. Copper, diquat, endothall dimethylalkylamine, and fluridone produced > 90% control. Carfentrazone (69%) and the potassium salt of endothall (62%) provided greater than 50% control, with other herbicides producing somewhat less than 50% control. Field demonstration has substantiated some of these findings. A study of three treatment plots in 2016 found an 85% reduction in biomass in fluridone-treated plots, compared to a 26% increase in biomass in untreated plots. A field trial on two plots treated with diquat found 98% and 80% control, respectively. A field trial with the dipotassium salt of endothall resulted in 43% control after one treatment.

The invasive annual grasses barb goatgrass (Aegilops triuncialis L.) and medusahead (Elymus caput-medusae L.) are widespread in western states and present management challenges on grasslands. To develop an integrated management strategy for these species, we treated sites in five pastures in Mendocino County, comparing combinations of intensive sheep grazing, glyphosate herbicide (low and high), and application timings (tillering, boot and heading stage). We found that grazing alone reduced barb goatgrass spikelet densities by 68% and the number of seeds per spikelet by 35%. Both rates of glyphosate application without grazing had similar effects on seed production. High and low glyphosate application at tillering resulted in almost complete control of both target species. Boot- and heading-stage applications reduced barb goatgrass density by 39% and 32%, respectively. Application at the boot stage also resulted in an 82% reduction in number of seeds per barb goatgrass spikelet. Our results suggest that intensive grazing may be a useful management strategy to reduce barb goatgrass and medusahead spikelet densities and barb goatgrass seed numbers, especially when integrated with a boot- or heading-stage glyphosate application.

The Sacramento-San Joaquin Delta is the largest freshwater estuary on the West Coast of the United States. Delta habitat and economic utility are compromised by waterhyacinth, a floating aquatic weed. Standard control measures for waterhyacinth include foliar treatment with glyphosate or 2,4-D. We have conducted trials over three seasons to evaluate efficacy of newly registered, low-userate aquatic herbicides. In 2017, we evaluated water-hyacinth control using carfentrazone (133 g ai ha(-1)), flumioxazin (322 g ai ha(-1)), imazamox (280 g ae ha(-1)), and florpyrauxifen-benzyl (29.4 and 58.8 g ai ha(-1)), as well as various tank mixes, compared with a standard rate of glyphosate (1,681 g ae ha(-1)). Plots were established in floating 1-m(2) quadrats, and treatments were replicated four times. All treatments were applied in 935 L ha(-1) solution with 3.5 L ha(-1) nonionic surfactant. We also included treatments with glyphosate (1,681 g ae ha(-1)) in lower spray volumes of 234 and 468 L ha(-1). We collected biomass samples at 8 wk after treatment (WAT). Three treatments reduced waterhyacinth biomass by > 95%: florpyrauxifen-benzyl (58.8 g ai ha(-1)), flumioxazin + imazamox (322 + 280 g ai/ae ha(-1)), and the 468 L ha(-1) application of glyphosate (1,681 g ae ha(-1)). Tank mixes (flumioxazin + imazamox, carfentrazone + imazamox, carfentrazone + glyphosate, and flumioxazin + glyphosate) gave approximately additive control. Florpyrauxifen-benzyl and flumioxazin + imazamox may be effective alternatives to glyphosate for controlling waterhyacinth with reduced rates of active ingredient. Glyphosate applied in a spray volume of 468 L ha(-1) produced better control than the same rate of glyphosate in 935 L ha(-1), suggesting that spray-volume optimization may be a useful topic for future research.

Waterhyacinth (Eichhornia crassipes (Mart.) Solms) is a global aquatic weed. Although a number of herbicides such as 2,4-D and glyphosate effectively control this plant, additional herbicides need to be evaluated to address concerns for herbicide stewardship and environmental restrictions on the use of herbicides in particular areas. Waterhyacinth has become a significant nuisance in the Sacramento-San Joaquin River Delta. The predominant herbicides for management of waterhyacinth in the Delta have been 2,4-D and glyphosate. However, environmental restrictions related to irrigation water residues and restrictions for preservation of endangered species are prompting consideration of the new reduced-risk herbicides imazamox and penoxsulam. Two trials were performed in floating quadrats in the Delta during the summer of 2016. In the first trial, two rates each of 2,4-D, glyphosate, imazamox, and penoxsulam were treated in four replicate quadrats. In this trial, the highest rates of all four herbicides provided greater than 80% control (2,4-D, 82%; glyphosate, 87%; imazamox, 93%; and penoxsulam, 94%). In the second trial, the lower rate of glyphosate (1,681 g a.e. ha(-1)) was compared to four rates each of imazamox (187 to 1,494 g a.i. ha(-1)) and penoxsulam (12 to 98 g a.i. ha(-1)). In this trial, the highest rates of imazamox and penoxsulam provided 96 and 95% control, respectively, compared to the untreated reference. Imazamox and penoxsulam will provide suitable control of waterhyacinth as part of an operational program and may be used as part of an integrated pest management program with considerations of herbicide resistance management. In addition, incorporating these reduced-risk herbicides into the management program can reduce the amount of pesticides applied per acre to achieve waterhyacinth control.

Exotic annual grasses such as medusahead [Taeniatherum caput-medusae (L.) Nevski] and downy brome (Bromus tectorum L.) dominate millions of hectares of grasslands in the western United States. Applying picloram, aminopyralid, and other growth regulator herbicides at late growth stages reduces seed production of most exotic annual grasses. In this study, we applied aminopyralid to T. caput-medusae to determine how reducing seed production in the current growing season influenced cover in the subsequent growing season. At eight annual grassland sites, we applied aminopyralid at 55, 123, and 245 g ae ha(-1) in spring just before T. caput-medusae heading. The two higher rates were also applied pre-emergence (PRE) in fall to allow comparisons with this previously tested timing. When applied in spring during the roughly 10-d period between the flag leaf and inflorescence first becoming visible, just 55 g ae ha(-1) of aminopyralid greatly limited seed production and subsequently reduced T. caput-medusae cover to nearly zero. Fall aminopyralid applications were less effective against T. caput-medusae, even at a rate of 245 g ae ha(-1). The growing season of application, fall treatments, but not spring treatments, sometimes reduced cover of desirable winter annual forage grasses. The growing season after application, both spring and fall treatments tended to increase forage grasses, though spring treatments generally caused larger increases. Compared with other herbicide treatment options, preheading aminopyralid treatments are a relatively inexpensive, effective approach for controlling T. caput-medusae and increasing forage production.

Woolly distaff thistle is a long-lived winter annual that threatens the ranching and dairy industries within the North Coast counties of California, particularly the organic producers. No peer-reviewed publications have documented effective control options or integrated management approaches for this species. We conducted two experiments, each replicated, in Marin County, California. The first compared several conventional herbicides at two timings and rates, while the second compared a conventional herbicide treatment with organic and integrated organic control methods, including an organic herbicide (mixture of capric and caprylic acids). Results of the conventional herbicide treatments showed most spring applications (March or April) of aminopyralid, aminocyclopyrachlor, clopyralid, and combinations of aminopyralid + triclopyr, or aminocyclopyrachlor + chlorsulfuron had greater than 99% control of woolly distaff thistle with fewer than 1.5 seedlings per 27-m2 plot by the end of the growing season. Higher rates were generally necessary to achieve the same level of control with winter (January) applications. In the organic herbicide treatments, the most consistent treatment was a combination of mowing followed by 9% (v/v) or the organic herbicide. This treatment was slightly less effective compared with aminopyralid but did have better than 95% control of woolly distaff thistle. The results of this study provide control options for both conventional and organic ranching practices where woolly distaff thistle is a problem.

Yellowflag iris, native to Europe, is a rhizomatous, emergent, invasive plant found in pond margins, ditches, and other wetland sites in much of the United States. In water depths up to approximately 50 cm, it forms dense stands, which displace native sedges and rushes, reducing waterfowl habitat and water flow. The rhizomes can reach 6 m in lateral spread, making it very difficult to control by mechanical methods. In addition, conventional boom-sprayer applications are often impractical in most aquatic systems. Drizzle application is a technique for directed treatment of hard-to-reach invasive plants. It uses low volumes (26 to 104 L ha—1) of concentrated herbicide solution, applied using a spray gun emitting a thin stream of solution with an effective range of 6 m. In this study, conducted along the margins of two ponds at the University of California, Davis, we compared drizzle applications of glyphosate, imazapyr, and triclopyr to applications using a conventional boom sprayer. Although both glyphosate and imazapyr provided excellent control (> 96 %) of yellowflag iris with either treatment technique, only the drizzle treatments of imazapyr at 2.26 and 4.52% ae (10 and 20% product) at spray volumes of 52 and 26 L ha—1, respectively, were below the maximum labeled rate and still gave > 98 % control. Furthermore, a cost analysis indicated that the most economical application for effective control of yellowflag iris was a drizzle application of imazapyr at 4.52 % ae (20 % product) at 26 L ha—1. This study demonstrates that drizzle application with imazapyr can be a practical application method for yellowflag iris control in aquatic systems in which broadcast treatments with conventional boom sprayers may be difficult.

Aminopyralid is the most commonly used herbicide for the control of yellow starthistle and other invasive thistles in annual grasslands of California. Although the effects of aminopyralid on native plant communities over a 2-yr period have been evaluated in prairies dominated by perennial species in the northern central states, similar evaluations have not been conducted in grassland communities of California, which are generally composed of a high diversity of native and nonnative annual species. In this study we monitored the effects of 53 and 123 g ae aminopyralid ha(-1) on individual species cover and species richness over three growing seasons in two locations on California annual grassland. Treated plots were compared to untreated plots in randomized complete-block designs. Results were largely consistent between the two trials. In the first season after treatment, both rates of aminopyralid reduced dicot cover significantly, particularly members of the Asteraceae and Fabaceae. Treated plots also showed reduced species richness. However, these differences were less pronounced in the second season after treatment, particularly at the low rate. By the third season after treatment in both sites, there were no longer any significant effects on cover or species richness at the low herbicide rate. On California annual grasslands, winter applications of low rates of aminopyralid have been shown to give excellent control of yellow starthistle, providing long-term benefits to grassland ecosystems. Results of the current study suggest that negative impacts of aminopyralid on the desirable native forb community are transitory.

Increasing lease and purchase costs for irrigated pasture are encouraging producers to use herbicides to control broadleaf weeds and maximise forage production on their current land base, as opposed to seeking out additional pastureland. There are two challenges to using herbicides to control broadleaf weeds in irrigated pasture. One challenge is that producers often encounter weed species for which herbicide information is limited or non-existent. Another challenge is that broadleaf herbicides tend to damage desirable legumes. We evaluated the efficacy of several broadleaf herbicide options for Symphyotrichum divaricatum, a species not previously the subject of research in irrigated pasture. Additionally, we monitored herbicide impacts on desirable legumes, principally Trifolium repens. We also evaluated control of Convolvulus arvensis and Plantago lanceolata, subjects of considerable herbicide research. Our goal was to identify herbicide options that controlled all three weeds while causing the least damage to legumes. Several rates and combinations of dicamba, diflufenzopyr, triclopyr, MCPA, and 2,4-D provided statistically indistinguishable control of C. arvensis and P. lanceolata, with point estimates suggesting 66–88% control in 2011 and 43–86% control in 2012. 2,4-D + triclopyr appeared to provide the best control of P. lanceolata and C. arvensis, with point estimates for high rates indicating nearly 100% control in both years. However, this combination was particularly damaging to legumes. Based on these findings, we recommend 2,4-D + triclopyr where desirable legumes are absent. Where desirable legumes are present, we recommend 2,4-D alone. Whereas 2,4-D alone appeared not to perform as well as 2,4-D + triclopyr at controlling S. divaricatum in 2012, it was less damaging to desirable legumes.

Dittrichia graveolens is a rapidly spreading invasive plant in California. While populations are observed primarily in disturbed areas, there is concern it may expand into adjacent undisturbed areas, particularly grasslands and riparian corridors. In a field experiment conducted in two successive years, we compared plant growth and phenological development of fall, winter, and spring sown seeds. Plants establish equally well in disturbed upland sites in both above and below average precipitation years but the absence of late spring rainfall negatively affected total plant biomass. In a greenhouse experiment, we compared growth in four light environments (100, 50, 27 and 9 % available light). Total plant growth decreased exponentially with decreasing light. This suggests that D. graveolens is not competitive in low light environments, such as woodlands and riparian forests. All plants flowered in early- to mid-September, coinciding with flowering in field grown plants, suggesting that photoperiod is the primary signal for reproductive growth. Using a minirhizotron system, we measured root growth over time in D. graveolens and three common California annual grassland species, two non-natives, Centaurea solstitialis and Bromus hordeaceus, and the native forb Holocarpha virgata. Root growth of D. graveolens began later in the season than the other species, reaching depths > 1 m by late May. Roots of C. solstitialis and H. virgata reached > 1 m earlier in the season. The temporal difference in root growth suggests that D. graveolens may be less competitive for soil moisture with other early season annuals than other deep-rooted broadleaf species found in grasslands.

Crops grown for bioenergy production are a mandated component of the United States energy portfolio. Giant miscanthus (Miscanthusxgiganteus) is a leading bioenergy crop similar in habit to the invasive plant giant reed (Arundo donax). To characterize the environmental tolerance of giant miscanthus, we compared the soil moisture stress tolerance of giant miscanthus and giant reed under glasshouse conditions. We subjected both species to soil moisture conditions of severe drought (-4.2MPa), mild drought (-0.5MPa), field-capacity (control), and flooded soils. These conditions were applied to two cohorts: one in which soil moisture conditions were imposed on newly planted rhizome fragments, and one in which conditions were imposed on established plants after 8weeks of growth in field-capacity soil. After 16weeks, we harvested all plants, measured above- and belowground biomass, and evaluated the reproductive viability of rhizome fragments. The total biomass of each species under flooded conditions was not different from the field-capacity control groups regardless of cohort. However, drought did affect the two cohorts differently. In the cohort treated after 8weeks of growth, mild and severe drought conditions resulted in 56% and 66% reductions in biomass, averaged over both species, compared with the controls. In the cohort treated for the entire 16weeks, mild and severe drought conditions resulted in 92% and 94% reductions in biomass. Rhizome fragments from both species and both cohorts showed 100% viability following flooded and control treatments; drought treatments reduced rhizome viability in both species, with a greater impact on giant miscanthus. Although giant miscanthus does not appear to have the potential to escape and establish in relatively dry upland ecosystems, it does show tolerance to flooded conditions similar to giant reed.

Downy brome or cheatgrass (Bromus tectorum) and medusahead (Taeniatherum caput-medusae) are the most problematic invasive annual grasses in rangelands of the western United States, including sagebrush communities that provide habitat to sage grouse. Rehabilitation of infested sites requires effective weed control strategies combined with seeding of native plants or desirable competitive species. In this study, we evaluated the effect of three fall-applied pre-emergence herbicides (imazapic, rimsulfuron, and chlorsulfuron+sulfometuron), and one spring-applied postemergence herbicide (glyphosate) on the control of downy brome and medusahead and the response of seeded perennial species and resident vegetation in two sagebrush communities in northeastern California. All pre-emergence treatments gave >93% control of both invasive species at both sites in the first year. Glyphosate was less consistent, giving > 94% control at one site and only 61% control of both species at the other site. Imazapic was the only herbicide to maintain good control (78-88%) of both species 2 yr after treatment. No herbicide caused detectible long-term damage to either perennial grasses or annual forbs, and imazapic treatment resulted in an increase in resident native forb cover 3 yr after treatment. Broadcast seeding with or without soil incorporation did not result in successful establishment of perennial species, probably due to below-average precipitation in the year of seeding. These results indicate that several chemical options can give short-term control of downy brome and medusahead. Over the course of the study, imazapic provided the best management of both invasive annual grasses while increasing native forb cover.

Giant miscanthus (Miscanthus × giganteus) and giant reed (Arundo donax) are leading bioenergy crops. Both exhibit many invasive characteristics, though only giant reed is known to be invasive. Despite this, neither produces viable seed, limiting movement to vegetative propagules. Therefore, to assess vegetative fragments as potential propagules, we quantified seasonal changes in culm node viability and performance in giant miscanthus and giant reed under greenhouse conditions. Giant miscanthus culms were collected in spring, summer, fall, and winter from established fields, while giant reed culms were collected in summer, fall, and winter from feral stands. Treatments at each timing consisted of whole culms and single-node culm fragments planted in soil or placed in standing water for an 8-week period. Giant miscanthus whole culms and fragments produced shoots and roots in both soil and standing water immediately following cutting from spring to summer, but failed to produce shoots and roots after fall and winter cutting dates. All rhizome fragments survived and generated shoots and roots after burial. By comparison, giant reed produced shoots and roots in both soil and standing water throughout the year, regardless of cutting date. With giant miscanthus, precautions should be taken when living culms or rhizome fragments are harvested and transported through riparian habitats during the summer months. By comparison, giant reed showed a remarkable increase in propagule generation and productivity throughout the year and, thus, escaped propagules present a far greater risk of unintentional establishment compared to giant miscanthus.

Understanding seed characteristics and seedling establishment patterns is essential for the development of effective management strategies for invasive annual species. Dittrichia graveolens (stinkwort) has increased its range rapidly within California since 1995, yet its biology is not well understood, which has led to poorly timed management. In this study, seed viability, germination, longevity, and dormancy, as well as seedling emergence characteristics of D. graveolens were evaluated in field, greenhouse, and laboratory experiments in Davis, CA, over a 2-yr period (fall 2010 to summer 2012). In the laboratory, seed germination of D. graveolens occurred at a wide range of constant temperatures (12 to 34 C). Cumulative germination was comparable to total seed viability (80 to 95%) at optimal germination temperatures, indicating that primary (innate) dormancy is likely absent. The base temperature for germination was identified using a thermal time model: 6.5 C and 4 C for 2010 and 2011 seed populations, respectively. In the field, seedlings emerged from fall through spring following precipitation events. A very low percentage of seedlings (2.5%) emerged in the second year after planting. Equivalent seedling emergence was observed over a wide range of light conditions (100, 50, 27, and 9% of available sunlight) in a greenhouse experiment, indicating that seed germination is not limited by high or low light. Results from these seed experiments improve our understanding of the reproductive biology of this rapidly expanding exotic annual and provide valuable information for developing effective timing and longevity of management programs.

Yellow starthistle (Centaurea solstitialis) is among the most invasive plant species in the western United States. The long-term management of yellow starthisde should include an integrated approach that incorporates establishment of competitive vegetation. In this study, conducted in two locations at Fort Hunter Liggett in Monterey County, California, we evaluated the combination of prescribed burning, the herbicide aminopyralid, and reseeding of native broadleaf and grass species on both yellow starthistle control and native plant restoration. Both study sites were burned in late October 2009. Over the following season, aminopyralid was applied at three timings and native plant species were seeded at three timings, using both a drill-seeder and broadcast spreader. Evaluations over the next 3 yr showed that aminopyralid provided complete to nearly complete control of yellow starthistle when applied between January and March, and this level of control was maintained for two seasons. Native plants failed to establish when broadcast seeded, regardless of the timing. December and January drill seeding timings were the most successful in establishing native species. There was a strong herbicide and drill seed timing interaction effect on native grass cover at both study sites. Over the course of the study the native perennial grass Stipa cernua was the most successful seeded species to establish, but establishment was slow and required 3 yr. Our results indicate that a January or March aminopyralid treatment integrated with a native perennial grass drill seeding program in January offers the greatest probability of both successful yellow starthisde control and perennial grass establishment.

The expected production of biomass-derived liquid fuels in the United States may require cultivation of millions of acres of bioenergy crops, including perennial grasses such as switchgrass. Switchgrass is not native to California and possesses many qualities in common with other perennial grasses that are invasive. To evaluate the potential invasiveness of switchgrass in California, we conducted risk analysis and climate-matching models as well as greenhouse and field evaluations of switchgrass, looking at its environmental tolerance and competitive ability against resident riparian vegetation. We concluded that dryland regions of California are not suitable to vigorous establishment and invasion of switchgrass. However, riparian areas appear to be far more likely to support switchgrass populations. With effective mitigation practices in place throughout the development, growth, harvest, transport and storage processes, it should be possible to minimize or eliminate the movement of seeds and vegetative propagules to sensitive habitats. Consequently, we believe that switchgrass is unlikely to become a significant problem in California, even with widescale production.

Dalmatian toadflax is listed as a noxious weed in most of the western United States, but control of this species has not been extensively studied in California. Studies in other states show effective control of Dalmatian toadflax with piclorarn, but this herbicide is not registered for use in California. In addition, reports vary as to the optimal timing for herbicide applications. In this study we evaluated several herbicides with combined foliar and soil-residual activity at two times of application: postsenescence (fall) and rosette (winter to early spring). We applied two series of treatments (2008 and 2009 to 2010) on adjacent sites in high desert scrub of southern California. In the year of treatment and the following year, we evaluated Dalmatian toadflax cover and presence/absence of associated dominant species (>= 5% cover). Although time of application, treatment, and timing by treatment interaction all produced significant differences in Dalmatian toadflax cover in the 2008 trial, only the high rate of aminocyclopyrachlor (280 g ae ha(-1)) applied to dormant plants in fall consistently reduced cover through the second year. No treatments at the rosette stage consistently produced 2 yr of control. In 2009 to 2010, treatments were more effective, probably owing to higher precipitation in spring. In both dormant and rosette applications made in 2009 to 2010, aminocydopyrachlor (140 and 280 g ae ha(-1)) and aminocyclopyrachlor + chlorsulfuron (140 g ae ha(-1) + 53 g ai ha(-1)) gave second year control; chlorsulfuron at the dormant stage (105 and 158 gal ha(-1)) and aminopyralid at the rosette stage (245 g ae ha(-1)) also gave 2 yr of control. The treatments had only minor effects on grass species. The response of broadleaf species varied among treatments, with aminocydopyrachlor at the high rate increasing Eriogonum spp., but greatly reducing Asteraceae species. These results provide options for the management of Dalmatian toadflax in California and other western states.