Six species of insects and one rust fungus that attack yellow starthistle (YST) have become established in California. All the insects attack the flower heads, which reduces seed production. The hairy weevil and the false peacock fly have achieved high densities over large areas in California. The latter species was unintentionally introduced, and is not permitted for release; however, it is very specific to yellow starthistle (Balciunas and Villegas 2007). Currently, the combination of these two insects provides the most impact.

Yellow starthistle populations have decreased in some areas, especially in ungrazed grasslands that have a dense cover of grasses. In a long-term field study by CDFA, the hairy weevil attacked 20-80% of flower heads (variation year-to-year), the false peacock fly 2-28%, the gall fly 1-23%, the flower weevil 0-1%, and the bud weevil 0-4% (Pitcairn et al. 2005). Some naturally-occurring generalist microbial pathogens (Ascophyta n. sp., Sclerotinia minor and Colletotrichum gloeosporioides) attack seedlings, which can cause up to 90% mortality (Pitcairn et al. 1999). The introduced rust (Puccinia jaceae var. solstitialis) was distributed widely throughout California but did not establish well at most sites and generally appears to be ineffective in California (Woods et al. 2010). The insect biological control agents feed directly on seeds in flower heads and reduce seed production, but they do little to reduce the growth or size of existing plants. To achieve population-level effects on yellow starthistle, high levels of seed destruction (e.g. >80 may be necessary. The rosette weevil, Ceratapion basicorne, was permitted for release in 2019 and was first released in April 2020 in Solano County. However, because of its slow rate of population growth and difficulty to rear in the laboratory, it will take several years before it is available for general distribution. This insect should help to reduce plant size and survivorship.

The bud weevil, Bangasternus orientalis (Coleoptera: Curculionidae) (Fig. 1), lays an egg, which looks like a black lump on the stem or leaf just below a flower head (Fig. 2). The larva tunnels up the stem into the flower head and consumes the developing seeds. A pupal chamber is formed inside the flower head, and adults emerge by the end of the summer leaving a characteristic 'bullet hole' (Fig. 3) and hide during winter hibernation.

The rosette weevil, Ceratapion basicorne(Coleoptera: Curculionidae), was permitted in 2019 for release in California. It has one generation per year. Adults start to feed on rosette leaves in spring, and lay eggs in the leaf blades and midribs (Fig. 4). Larvae tunnel down the petiole and feed inside the upper root. More than one insect can develop inside a plant. Pupation occurs inside the plant (Fig. 5), and adults emerge at the time the plant is bolting (late May - June). Adults feed and mate on the plant for a week or two then disappear until the next spring.

The peacock fly, Chaetorellia australis (Diptera: Tephritidae), lays one or more eggs in a flower bud (Fig. 6). Larvae chew on developing seeds, which leaves a mixture of seeds and loose debris inside attacked flower heads (Fig. 7). This fly has a characteristic papery 'puparium' in which it completes development to the adult stage inside the flowerhead. More than one insect can develop inside a flower head. There are two or more generations per year, and the insect passes winter as a larva inside the seed head. This species emerges earlier in the spring than C. succinea, and depends primarily on bachelor's button (Centaurea cyanus) as a host for its first generation. Because bachelor's button is not very common in California this species is rare.

The false peacock fly, Chaetorellia succinea (Diptera: Tephritidae), is very similar to the peacock fly but has an extra spot on its thorax (Fig. 6). This is the most widespread of the YST insects in California and often is seen resting on a flower bud. Its biology is very similar to that of C. australis, and it was mistakenly introduced for this species in 1991. Since then its host specificity has been studied, and it appears to not pose a risk to any nontarget native plants in California. This insect has been found to attack one variety of safflower where it was being grown to foster game birds, but it was not found in fields for commercial production (Balciunas and Villegas. 2001, 2007).

The hairy weevil, Eustenopus villosus (Coleoptera: Curculionidae), is the most effective permitted biological control agent for yellow starthistle (Fig. 8). Adults feed on the first small flower buds, which can change the architecture of the plant (Figs. 9 & 10). Flower buds normally develop at the tips of branches, but the weevil eats many of these, which causes secondary buds to develop further down the branches. When flower buds get larger, the weevil chews a small hole in the side, where it lays an egg and covers it with black frass (Fig. 11). These oviposition sites can be readily seen on yellow starthistle flower heads when the hairy weevil is present. One larva develops inside the flower head and consumes most of the developing seed. It pupates inside the flower head (Fig. 12), and adults emerge by the end of the summer to hibernate off the plant.

The flower weevil, Larinus curtus (Coleoptera: Curculionidae), is often seen on an open flower, with its head buried deep among the florets (Fig. 13). It lays an egg in the open flower head, and the larva consumes developing seeds (Fig. 14). It pupates inside the flower head, and adults emerge by the end of the summer to hibernate off the plant.

The gall fly, Urophora sirunaseva (Diptera: Tephritidae; Fig. 15), creates a hard nut-like gall from the flower ovary, inside of which its larva develops (Fig. 16). Adults have clear wings with a black 'UV' pattern on the wings. The flies can have more than one generation during the summer, and they spend the winter as mature larvae inside the flower head.

The yellow starthistle rust, Puccinia jacea var. solstitialis(Uredinales: Pucciniaceae), has a complicated life cycle with five life stages (Fig. 17). It is known to infect only yellow starthistle, bachelor's button (Centaurea cyanus) and to a very limited extent safflower (Bruckart 2006). The rust produces resting spores (teliospores) that resist cold temperatures during winter. Resting spores germinate in the spring and produce a sequence of different kinds of spores that infect plants. The final type of spores (urediniospores) are able to reinfect plants during the growing season. Optimal conditions for infection are cool (65-70°F) and humid (at least 8 hours of dew). Infected plants produce pustules in 2 or more weeks, depending on temperature, which release spores that disperse in the wind. In the summer resting spores are produced, which are crucial for initiating infections in the following spring. The rust was released at 176 sites in 40 counties during 2004-2006, but it persisted at only one site close to San Francisco Bay in Sonoma County.

Biological Control Agents


Common name




Bangasternus orientalis

bud weevil



First released in 1985, more abundant in Coastal Mtns in northwest CA

Ceratapion basicorne

rosette weevil



First release in April 2020 in Solano Co.

Chaetorellia australis

peacock fly



First released in 1988, more common on bachelor's button than YST

Chaetorellia succinea

false peacock fly



Accidental introduction in 1991. Not a permitted agent. The most abundant & widespread species.

Eustenopus villosus

hairy weevil



First released in 1990, more abundant in northern CA, Sierra Foothills & Coastal Mtns

Larinus curtus

flower weevil



First released in 1992

Puccinia jacea var. solstitialis

yellow starthistle rust



First released in 2003; almost no establishment.

Urophora sirunaseva

gall fly



First released in 1984; more abundant in the coastal mountains

How the Technique Is Employed

Look for presence of insects and their damage to determine what species are present at your site. It is not worth releasing insects if they are already present. Before flowers bloom, the hairy weevil feeds on small flower buds, causing them to 'flag' (Fig. 5). This damage also changes the plant's architecture, as secondary buds develop into flowers. The presence of seed head insects can be detected by opening mature flower heads and looking for larvae and/or signs of damage (Figs. 1, 3, 4, 6 and 7).

Collect adult insects in the field by sweep net, although this is difficult once the spines appear on the flower heads.

Detailed information and photographs can be found in DiTomaso et al. (2006) and Randall et al. (2017).

Special Tips

It is not known where the weevils overwinter, but it is likely in sheltered places, such as in tree bark, under rocks or in leaf litter. The fly species overwinter inside flower heads as mature larvae and may be susceptible to predation by rodents or grazing ungulates eating flower heads.

Some of these agents, particularly the hairy weevil and the false peacock fly, also attack tocalote or Maltese starthistle (C. melitensis) and Sicilian starthistle (C. sulphurea).


Knowing what insects are present may help you to integrate other management strategies. The three weevil species emerge from the flower heads in the summer and overwinter as adults away from the plant. The fly species overwinter inside flower heads as larvae. Thus, for example, fall grazing of flower heads by goats would kill most of the flies, but not affect the weevils.

Mowing yellow starthistle in the spring will delay flowering, which may reduce the effectiveness of the insects, who will grow old before they can lay eggs.

Herbicides that kill yellow starthistle before it produces mature flower heads will deprive seed head insects of the ability to reproduce. However, the adults will search for the remaining plants that have not been killed. Thus, biological control can be complementary to herbicide or other control treatments, especially if there are areas that are not treated (e.g., too difficult to access or too environmentally sensitive to treat).

The false peacock fly is not a permitted agent and should not be redistributed. It is known to attack two native species of basket flower, neither of which occur in California.

There are no reports of non-target use by the permitted biocontrol agents.

Where Can I Get These?

Some insects may be available from your county Agricultural Commissioner. Historically, some insects have been commercially available by some vendors such as Biological Control of Weeds, Inc. (www.bio-control.com).


Balciunas, J.K. and B. Villegas. 2001. Unintentionally released Chaetorellia succinea (Diptera: Tephritidae): Is this natural enemy of yellow starthistle a threat to safflower growers? Environ. Entomol. 30(5): 953-963.

Balciunas, J.K. and B. Villegas. 2007. Laboratory and realized host ranges of Chaetorellia succinea(Diptera: Tephritidae), an unintentionally introduced natural enemy of yellow starthistle. Environ. Entomol. 36(4): 849-857.

Bruckart, W.L., III. 2006. Supplemental risk evaluations of Puccinia jaceae var. solstitialis for biological control of yellow starthistle. Biological Control 37: 359-366.

DiTomaso, J., G.B. Kyser and M.J. Pitcairn. 2006. Yellow Starthistle Management Guide. Cal-IPC Publication 2006-03. California Invasive Plant Council, Berkeley, California. 74 p.

Pitcairn, M.J., D.M. Woods, D.B. Joley, DG. Fogle and V. Popescu. 1999. Impact of seedling pathogens on yellow starthistle in California. In:D.M. Woods (ed.), Biological Control Program Annual Summary, 1998. California Department of Food and Agriculture, Plant Health and Pest Prevention Services, Sacramento, CA. pp. 74-76.

Pitcairn, M.J., D.M. Woods and V. Popescu. 2005. Update on the long-term monitoring of the combined impact of biological control insects on yellow starthistle. In D.M. Woods (ed.), Biological Control Program Annual Summary, 2004. California Department of Food and Agriculture, Plant Health and Pest Prevention Services, Sacramento, California, pp. 27-30.

Randall, C.B., R.L. Winston, C. Jette, M.J. Pitcairn and J.M. DiTomaso. 2017. Biology and Biological Control of Yellow Starthistle. Fourth Edition. USDA Forest Service, Forest Health Technology Enterprise Team, Morgantown, West Virginia. FHTET-2016-08.

Woods, D.M., A.J. Fisher and B. Villegas. 2010. Establishment of the Yellow Starthistle Rust in California: Release, Recovery, and Spread. Plant Disease 94: 174-178.
DOI: 10.1094/PDIS-94-2-0174

Contributing Authors

Dr. Michael J. Pitcairn, Program Manager, California Department of Food and Agriculture

Dr. Lincoln Smith, Research Entomologist, USDA