Cirsium arvense (creeping thistle)

Donald (1990) summarized the management of C. arvense using non-chemical methods and herbicides in various crops. Edwards et al. (2000) provides more recent insights to promote interspecific competition to manage C. arvense through timing of crop sowing, grazing regimes, and nitrogen fertilization. Another useful review proposing integrated strategies for control of C. arvense in crops involving tillage, herbicide use and cultural control is provided by Pollack and Bailey (2001).

C. arvense has been named on most state and federal seed and weed noxious weed laws in the USA. Canadian legislation is similar (Moore, 1975). It appeared on more noxious weeds lists (33) than any other weed in North America (Skinner et al., 2000). C. arvense was introduced into French Canada from Europe (Anon., 1918) before being spread into Vermont and New York in the USA (Stevens, 1846). Detmers (1927) concluded that it must have been introduced before 1795 because a Vermont law was enacted that year to halt its spread. By 1844, Ohio law limited sale of seed contaminated with C. arvense and required landowners to mow infested land and adjacent roadsides (Detmers, 1927). Judging by its current distribution in North America, state and federal legislation has been somewhat ineffective in limiting the spread of the weed (Wilson, 1981a; Skinner et al., 2000). It is also regulated in the UK under the 1959 Weeds Act (Bond and Turner, 2003).

Combining herbicides with cultivation, mowing or grazing, and competitive crops is more effective for controlling C. arvense than herbicides alone. Various combinations have been tested and reviewed for selective control of C. arvense in the major field crops (Donald, 1990; Edwards et al., 2000). Cultural practices used alone are frequently ineffective. Even when combinations of control practices are used, repeated control measures over several years are required to reduce the severity of the problem (Donald, 1990, 1992, 1994; Donald and Prato, 1992a, 1992b). However, more recent work has identified more specific approaches that may provide effective control. Cover crops show some potential (Moyer et al., 2000). Mowing or grazing needs to be adjusted to levels appropriate for a given system (Wilson and Kachman, 1999; Eerens et al., 2002). Edwards et al. (2000) recommended sowing crops of competing species as soon as possible after cultivation. Populations of C. arvense in field margins were also reduced by sowing other species (West et al., 1997; Denys and Tschamtke, 2002). For example, Ominski et al. (1999) found that Medicago sativa effectively suppressed C. arvense, resulting in more patchy populations. Fertilization, particularly with N may also reduce C. arvense populations, particularly in the absence of grazing (Edwards et al., 2000). Repeated mowing in combination with sowing of perennial grasses has been shown to virtually eliminate C. arvense (Wilson and Kachman, 1999). Cormack (2002) achieved 75% reduction in shoots following mowing in a legume crop. In New Zealand pastures, mowing later in the season resulted in greater reduction in autumnal root biomass (Bourdot et al., 1998). Mitchell et al. (2002) observed that grazing by sheep 3-4 times approximately 3 weeks between grazing depleted root reserves of C. arvense sufficiently to prevent survival. Van Toor and Popay (1995) found that wounding C. arvense plants in advance of grazing increased grazing pressure by sheep by improving palatability. Recent advances in knowledge of the biology of C. arvense, such as the development of shoot emergence models (Donald, 2000; Jensen et al., 2002) should aid in devising management strategies.

Maw (1976) and Moore (1975) summarized information on insects found on shoots and roots of C. arvense. For additional information on insects and nematodes found on C. arvense, see Natural Enemies. Survey work has identified numerous potential native biological control agents for C. arvense (Watson and Keogh, 1980; Perju et al., 1995). Biological agents for controlling C. arvense have been reviewed (Andres, 1980; Peschken et al., 1980; Trumble and Kok, 1982; Monnig, 1987). Widespread adoption of foreign biological control agents is unlikely because of public concern for native thistles (Peschken and Beecher, 1973) and the general lack of effectiveness of currently available biological control agents. Unfortunately, many of the insects and nematodes listed (see Natural Enemies) are often widespread, persistent pests of important crop species, limiting their use on commercial farms. The weevil Ceutorhynchus litura severely reduced overwintering survival of below-ground adventitious shoots of C. arvense to as little as 3% of that of uninfested shoots in Canada (Peschken and Beecher, 1973). In a 3-year study in Montana, 8 to 12% of weevil-infested shoots survived from one year to the next compared with 94 to 99% of uninfested shoots (Rees, 1990). This stem feeder was not nearly as devastating in spring as in autumn. Weevil damage also promoted the invasion of damaged shoots by other arthropods (mites, spiders, springtails), nematodes, and fungi, although the role of these organisms in plant death was not determined. The insect may have assisted in spreading the rust fungus, Puccinia punctiformis (Peschken and Beecher, 1973), although this assertion was not substantiated later (Peschken and Wilkinson, 1981). C. litura was released 18 times in the USA in California, Colorado, Idaho, Montana, New Jersey, South Dakota, and Washington between 1971 and 1975. In Montana, C. litura spread 9 km in 10 years and the proportion of infested plants increased from 11 to 29% in 1977 to over 80% after 10 years. In Canada, this insect did not greatly or consistently increase mortality of C. arvense shoots (Peschken and Wilkinson, 1981). A weevil, Larinus planus, that feeds on seed heads of C. arvense was accidently introduced into the USA, and may be useful for controlling seed production to prevent large areas of infestation from expanding (Drlik et al., 2000). However, it has been shown to attack a native thistle, Cirsium undulatum var. tracyi in Colorado (Louda and O'Brien, 2002). Rhiocyllus conicus, a weevil that plays a similar role is no longer favoured for biological control because it also attacks native thistles (Drlik et al., 2000). Predispersal seed predation by Dasyneura gibsoni and Orellia ruficanda can significantly reduce seed output (Forsyth and Watson, 1985; Heimann and Cussans, 1996). Cassida rubiginosa is a fairly effective control agent, and may work well in combination with the effects of competition on C. arvense using plants such as Festuca arundinacea and Coronilla varia (Ang et al., 1995).Fungi and higher plant parasites found on C. arvense have been reviewed (Moore, 1975; see also Natural Enemies). Most pathogenic viruses or bacteria reported on C. arvense are diseases of crops, such as the tobacco rattle tobravirus (Cooper and Harrison, 1973), limiting their potential for biological control of C. arvense. C. arvense is an alternative host for diseases and nematodes of crops. Drlik et al. (2000) list Pseudomonas syringae pv. tagetis, Puccinia punctiformis, and Sclerotinia scloerotiorum as the three main pathogens under investigation for use against C. arvense in North America. None of these were yet available commercially. Pseudomonas syringae pv. tagetis showed limited ability to affect Canada thistle survival in one USA study (Gronwald et al., 2002), but in Minnesota it reduced C. arvense biomass significantly in conservation tillage systems (Hoeft et al., 2001). Work in Germany has suggested that control with P. punctiformis could prevent flowering and hinder several years' growth (Kluth et al., 2003), and shows some potential for development into a mycoherbicide (Bond and Turner, 2003). The fungus Sclerotinia sclerotiorum applied to C. arvense patches as a biological control agent killed 20 to 80% of shoots in Montana (Brosten and Sands, 1986). Shoot emergence was also severely reduced in the growing season following treatment. Defoliation of shoots has a debilitating effect on the root systems as well (Bourdot and Harvey, 1994). S. sclerotiorum is an aggressive, persistent pathogen on many broadleaved crop species, limiting its use on commercial farms. A study in the Netherlands showed that the risks associated with S. sclerotinium may be manageable, however (Bourdot et al., 2001). Harvey et al. (1998) investigated using auxotrophic strains of S. sclerotinium to decrease the risk of pathogenic effects on crops, but these strains were less effective against C. arvense. Infestation of C. arvense by the bacterium P. syringae tagetis results in stunted plants that are unable to flower and are less able to overwinter (Drlik et al., 2000). Three insects that feed on C. arvense, (Aphis fabae spp. Cirsiiacanthoidis, Uroleucon cirsii and the beetle Cassida rubiginosa) were found to transmit the fungal pathogen P. punctiformis (Kluth et al., 2002), indicating the possibility of using synergism in biological control efforts. Furthermore, Bacher et al. (2002) showed that development of the beetle Apion onopordi was improved in plants infested with P. punctiformis, which is in turn is promoted by A. onopordi (Friedli and Bacher, 2001). However, Kruess (2002) found that the combination of Cassida rubiginosa and the pathogen Phoma destructive provided less efficient control of C. arvense. Green et al. (2001) observed high disease ratings for infection by Alternaria cirsinoxia in Saskatchewan, Canada.

Chemical control of C. arvense has been reviewed (Moore, 1975; Donald, 1990). Different growth stages differ greatly in susceptibility to herbicides. Plants in the rosette stage are more susceptible than plants that have already bolted (Hunter, 1996; Miller and Lym, 1998); seedlings are more susceptible and sensitive to a greater variety of herbicides than mature plants (Vangessel, 1999). With the advent of herbicide-resistant crops, new possibilities for control of C. arvense within crops have appeared, including use of glyphosate, which does provide effective control in these systems (May, 2000; Sarpe et al., 2001).Herbicides that have been used in different systems:Temperate cerealsBromoxynil, chlorsulfuron, clopyralid, 2,4-D, dicamba, MCPA, metsulfuron, flurasolum, Iodosulfuron-methyl-sodiumforamsulfuron.Maize and/or sorghumAtrazine, bentazone, bromoxynil, clopyralid, 2,4-D, dicamba.SoyabeansAcifluorfen, bentazone.SugarbeetsClopyralid.Dry BeansBentazone.PeasBentazone, MCPA.PastureBromoxynil, chlorsulfuron, 2,4-D, dicamba, metsulfuron, picloram, hexazinone.Chemical fallowAtrazine, chlorsulfuron, 2,4-D, dicamba, glyphosate, metsulfuron, picloram.Non-CroplandAmitrole, atrazine, bromoxynil, chlorsulfuron, 2,4-D, dicamba, dichlobenil, glyphosate, hexazinone, imazapyr, metsulfuron, picloram, sulfometuron, tebuthiuron.