Source: California Invasive Plant Council

URL of this page:

Invasive Plants of California's Wildland

Centaurea solstitialis
Scientific name   Centaurea solstitialis
Additional name information:   L.
Common name   yellow starthistle
Synonymous scientific names   none known
Closely related California natives   0
Closely related California non-natives:   11
Listed   CalEPPC List A-1,CDFA C
By:   John D. Gerlach Jr.,Joseph M. DiTomaso

Distinctive features:  

In California, yellow starthistle (Centaurea solstitialis) grows as a deep-taprooted winter annual, or rarely as a short-lived perennial. It produces one to many solitary, spiny, yellow flower-heads during late spring, summer, and fall. Seeds begin to germinate soon after fall rains, and young plants grow as prostrate to ascending taprooted rosettes until bolting occurs in late spring or early summer. Stem leaves of bolted plants extend downward, giving the stems a winged appearance. Flowering plants range from ankle to shoulder height and change color from green to bluish green in summer. Flowerheads are generally produced from June through September. The heads are initially produced on branch tips, but robust plants may produce heads in the branch axils later in the season. The main phyllaries (flowerhead bracts) are palmately spined with a single stout, apical spine and a few much smaller, lateral spines. Some individuals produce shorter apical spines. The heads contain two types of fruits or achenes. Most are cream to tan with a white pappus or plume; achenes in the outer ring are darker and lack a pappus.

Asteraceae. Annual. Stems: 6-72 in (15-200 cm) in height. Leaves: basal, earliest, entire to slightly toothed; subsequent, lobed to deeply lobed; bright green and scabrous-bristly in seedling and rosette stages; 2-6 in (5-15 cm) long; cauline leaves long, entire, narrow, decurrent; initially green, becoming bluish green and densely covered with cobwebby hairs later in the season; leaf blades 0.4-1.2 in (1-3 cm) long. Inflorescence: produced late May-December; heads 1 to many, always solitary; involucre 0.5-0.7 in (13-17 mm) tall, ovoid; outer phyllaries with apical appendages palmately spiny, central spine 0.4-1 in (10-25 mm) long, generally stout; tips of inner phyllaries with membranous winged tips about 1 mm wide. Flowers: many; corollas 0.5-0.8 in (13-20 mm) tall, unusually equal, yellow; marginal florets sterile, corollas 2-4 lobed, spreading to ascending; inner florets fertile, 5 lobed. Fruits: achenes 0.08-0.12 in (2-3 mm) long; those produced by outer ring of flowers dull, dark brown to blackish, without pappus; those produced by interior flowers glossy, grayish to mottled light brown; pappus white with bristles 0.08-0.166 in (2-4 mm) long, pappus bristles covered with rows of minute barbs; achene attachment scar obtuse, achene base broad (Hickman 1993, Gerlach unpubl. data).


Yellow starthistle is most widely distributed in the Sacramento and northern San Joaquin valleys, Inner North Coast Ranges, northern Sierra Nevada foothills, Cascade and Klamath ranges, and the central-western regions of the state (Hickman 1993). There are many small to large relict populations in the southwestern region of California. It is currently spreading in mountain regions of the state below 7,500 feet (2,250 m) and in the central-western region. It is uncommon in deserts and at moist coastal sites. Primarily it is a problem in moderately warm, exposed areas on fertile, drier soils, including disturbed sites, grasslands, rangeland, hay fields, pastures, roadsides, and recreational areas (DiTomaso et al. 1999).

WHERE DID IT COME FROM AND HOW IS IT SPREAD?   Yellow starthistle is native to southern Europe and western Eurasia and was first collected in Oakland, California, in 1869. It was most likely introduced after 1848 as a contaminant of alfalfa seed. Introductions prior to 1899 were most likely from Chile, while introductions from 1899 to 1927 appear to be from Turkestan, Argentina, Italy, France, and Spain (Gerlach in prep., Hillman and Henry 1928). By 1917 it had become a serious weed in the Sacramento Valley and was spreading rapidly along roads, trails, streams, ditches, overflow lands, and railroad rights-of-way (Newman 1917). In 1919 Willis Jepson observed its distribution near Vacaville and stated: “It is 1,000 times as common as ten years ago, perhaps even six years ago” (Jepson 1919).
     Yellow starthistle had spread to over a million acres of California by the late 1950s and nearly two million acres by 1965. In 1985 it was estimated to cover eight million acres in California (Maddox and Mayfield 1985) and perhaps ten to twelve million acres a decade later. It is equally problematic around Medford in southwestern Oregon and in Hell’s Canyon in Oregon and Idaho (Maddox et al. 1985). It also infests, to a lesser degree, areas in eastern Oregon, eastern Washington, and Idaho (Roché and Roché 1988).
     Human activities are the primary mechanisms for the long-distance movement of yellow starthistle seed. Seed is transported in large amounts by road maintenance equipment and on the undercarriage of vehicles. The movement of contaminated hay and uncertified seed is also an important long-distance transportation mechanism. Once at a new location, seed is transported in lesser amounts and over short to medium distances by animals and humans. The short, stiff, pappus bristles are covered with microscopic, stiff, appressed, hair-like barbs that readily adhere to clothing and to hair and fur (Gerlach unpubl. data). The pappus is not an effective long-distance wind-dispersal mechanism as wind moves seeds only short distances, with maximum wind dispersal being sixteen feet (<5 m) over bare ground with wind gusts of twenty-five miles per hour (40 km/hr) (Roché 1992).

WHAT PROBLEMS DOES IT CAUSE?   Dense infestations of yellow starthistle displace native plants and animals, threatening natural ecosystems and nature reserves. Yellow starthistle also significantly depletes soil moisture reserves in annual grasslands in California (Gerlach unpubl. data) and in perennial grasslands in Oregon (Borman et al. 1992). Long-term ingestion by horses causes a neurological disorder known as chewing disease, a lethal lesion of the nigropallidal region of the brain. This disease is expressed as a twitching of the lips, tongue flicking, and involuntary chewing. Permanent brain damage is possible, and affected horses may starve to death (Kingsbury 1964). Yellow starthistle interferes with grazing and lowers yield and forage quality of rangelands, thus increasing the cost of managing livestock (Roché and Roché 1988). It can also reduce land value and limit access to recreational areas.


Plants reproduce only by seed and generally flower from May to September. When adequate moisture is available, yellow starthistle can survive as a short-lived perennial and flower throughout fall, winter, and spring. However, the flowers produced during winter are often killed by frost (Gerlach unpubl. data). Almost all plants are self-incompatible and require pollen from a genetically compatible plant to produce seed (Maddox et al. 1996).

(click on photos to view larger image)



Centaurea solstitialis on left; C. melitensis on right

European honeybees are an important pollinator, and in some populations are responsible for 57 percent of seed set (Barthell unpubl. data). Seeds produced per head (30-80) and flowerhead production per plant (1-1,000) are variable, depending on soil moisture levels and intensity of competition (DiTomaso, unpubl. data). Large plants can produce nearly 75,000 seeds. Seed production in heavily infested areas varies between fifty to 200 million seeds per acre. Studies of seed survival in soil have found significant survival to ten years (Callihan et al. 1993). Seeds typically germinate in late fall or early winter, when soil moisture is present (Maddox 1981) and overwinter as basal rosettes.
     Germination responses in yellow starthistle are greatly reduced in dark environments and by exposure to light enriched in the far-red portion of the spectrum (Joley 1995). The two types of achenes also differ in response to light (Joley 1995). During early seedling establishment, root growth is vigorous and can extend deeper than one meter (3.3 ft) (Roché et al. 1994, DiTomaso unpubl. data), providing plants with access to deep soil moisture reserves during dry summer months. Reduced light levels cause the rosettes to produce fewer but larger leaves and to assume a more upright growth form (Roché et al. 1994). Reduced light levels also significantly reduce root growth and flower production (Roché et al. 1994). Consequently, survival and reproduction are significantly reduced in shaded areas, and the plant is probably less competitive in dense stands of established perennials. Bolting occurs from late spring to early summer, and spiny flowerheads generally are produced from early summer to late summer or fall. The spines on the flowerheads may protect them from herbivory by large animals, but they do not prevent significant herbivory by grasshoppers or seed predation by birds (Gerlach unpubl. data).



It is important to prevent large-scale infestations by controlling new invasions. Spot eradication is the least expensive and most effective method of preventing establishment of yellow starthistle. In established stands, any successful control strategy will require dramatic reduction or, preferably, elimination of new seed production, multiple years of management, and follow-up treatment or restoration to prevent rapid reestablishment.
     Effective control using any of the available techniques depends on proper timing. Combinations of techniques may prove more effective than any single technique. For example, prescribed burning followed by spot application of post-emergence herbicides to surviving plants can prevent the rapid reinfestation of the treated area. Similarly, combining mowing and grazing, revegetation and mowing (Thomsen et al. 1996a, Thomsen et al. 1996b), or herbicides and biological control may provide better control than any of these strategies used alone. Effective combinations may depend on location or on the objectives and restrictions imposed on land managers.


Physical control:  

Mechanical methods: Tillage can control this thistle; however, this will expose the soil for rapid reinfestation if subsequent rainfall occurs. Under these conditions, repeated cultivation is necessary (DiTomaso et al. 1998). During dry summer months, tillage practices designed to detach roots from shoots prior to seed production are effective. For this reason, the weed is rarely a problem in agricultural crops. Weedeaters or mowing can also be used effectively. However, mowing too early, during the bolting or spiny stage, will allow increased light penetration and more vigorous plant growth and high seed production. Mowing is best when conducted at a stage where 2 to 5 percent of the seed heads are flowering (Benefield et al. 1999). Mowing after this period will not prevent seed production, as many flowerheads will already have produced viable seed. In addition, mowing is successful only when the lowest branches of plants are above the height of the mower blades. Under this condition, recovery is minimized. Results should be repeatedly monitored, as a second or perhaps a third mowing may be necessary to ensure reduced recovery and seed production (Thomsen et al. 1996a, 1996b).

Prescribed burning: Under certain conditions, burning can provide effective control and enhance the survival of native forbs and perennial grasses (Robards, unpubl. data, DiTomaso et al. 1999a). This can be achieved most effectively by burning after native species have dispersed their seeds but before yellow starthistle produces viable seed (June-July). Dried vegetation of senesced plants will serve as fuel for the burn. At Sugarloaf Ridge State Park in Sonoma County, three consecutive burns reduced the seedbank by 99.5 percent and provided 98 percent control of this weed, while increasing native plant diversity and perennial grasses (DiTomaso et al. 1999a). No additional control method was used in the fourth year. In that year, unfortunately, the seedbank of yellow starthistle increased by thirty-fold compared to the previous year (DiTomaso unpubl. data).


Biological control:  

Insects and fungi: Six USDA approved insect species that feed on yellow starthistle have become established in California (Pitcairn 1997a and 1997b). These include three weevils, Bangasternus orientalis, Eustenopus villosus, and Larinus curtus, and three flies, Urophora sirunaseva, Chaetorellia australis, and C. succinea (Woods et al. 1995). All of these insects attack yellow starthistle flowerheads, and the larvae utilize the developing seeds as a food source. The most effective of these species are E. villosus and C. succinea (Balciunas and Villegas 1999). With the possible exception of a few sites, the insects do not appear to be significantly reducing starthistle populations, but success may require considerably more time for insect numbers to increase to sufficient levels.

Current evidence indicated a 50 to 75 percent reduction in seed production in areas with significant bioagent populations (Pitcairn and Ditomaso unpubl. data). A root-attacking flea beetle (Ceratapion brasicorne) is also being studied (Pitcairn, pers. comm.). Researchers are seeking other starthistle-specific foliar- and stem-feeding insects in Asia Minor. Research is also currently being conducted on three native or naturalized fungal pathogens, Ascochyta sp., Colletotrichum sp., and Sclerotinia sclerotiorum for the control of yellow starthistle seedlings (Woods and Popescu 1997).

Grazing: Intensive grazing by sheep, goats, or cattle before the spiny stage but after bolting can reduce biomass and seed production in yellow starthistle (Thomsen et al. 1996a, 1996b). To be effective, large numbers of animals must be used for short durations. Grazing is best between May and June, but depends on location. This can be a good forage species.

Plant competition: Revegetation with annual legumes capable of producing viable seed provides some level of control in pastures (Thomsen et al. 1996a, 1996b). In some areas subterranean clover (Trifolium subterraneum) proved to be the best of sixty-six legumes tested. In other sites rose clover (T. hirtum) and/or perennial grasses may be the preferred species. Control was enhanced when revegetation was combined with repeated mowing (Whitson et al. 1987).


Chemical control:  

Although several non-selective pre-emergence herbicides will control yellow starthistle, few of these can be used in rangeland or natural ecosystems. The exception is chlorsulfuron, which provides good control in winter when combined with a broadleaf selective post-emergence compound. However, chlorsulfuron is not registered for use in rangelands or pastures.

The primary options for control in non-crop areas are post-emergence herbicides; 2,4-D, triclopyr, dicamba, and glyphosate (DiTomaso et al. 1998). All but glyphosate are selective and preferably applied in late winter or early spring to control seedlings without harming grasses. Once plants have reached the bolting stage, the most effective control can be achieved with glyphosate (1 percent solution). The best time to treat with glyphosate is after annual grasses or forbs have senesced, but prior to yellow starthistle seed production (May-June). The most effective compound for yellow starthistle control is clopyralid (as Transline®), a broadleaf selective herbicide (DiTomaso et al. 1998). Clopyralid provides excellent control, both pre-emergence and post-emergence, at rates between 1.5-4 acid equivalent or 4-10 oz formulated product per acre. Although excellent control was achieved with applications from December through April, earlier applications led to significant increases in quantity of other forage species, particularly grasses.