Biological control of garlic mustard

Overview

Results

The team

Related information

So what's the problem?

Garlic mustard, Alliaria petiolata, is a biennial cruciferous plant of European origin. It is one of the few non-indigenous herbaceous species able to invade and dominate the understory of North American forests, and is considered one of the most serious invaders in the north-eastern and midwestern USA and south-eastern Canada.

This project to investigate the potential for biological control of the weed was started in 1998 in collaboration with Prof. Bernd Blossey of Cornell University in the US.

What is this project doing?

CABI conducted a literature review which revealed 69 phytophagous (plant feeding) insect species and seven fungi associated with garlic mustard in Europe. In subsequent field surveys, 30 species were collected and six of these were prioritized as potential biological control agents. Host-specificity tests (to ensure that the biological control agents don’t attack and breed on any other plants) showed that one of the species, a root-mining flea beetle, is not specific enough to be considered further, and another species was difficult to obtain and rear. Work therefore continued with the remaining four species. Between 2000 and 2006, a CABI staff member (Esther Gerber) conducted her PhD in the framework of the garlic mustard project on the impact and interactions of three of these. At present, CABI is investigating four weevil species: Ceutorhynchus alliariae, C. constrictus, C. roberti and C. scrobicollis.

Results so far

The two stem-mining weevils C. alliariae and C. roberti show high spatial and temporal niche overlap (i.e. they use the same plant parts at the same time), appear to be equally competitive, and have similar impact on garlic mustard. Host-specificity tests conducted early on in the project showed that both weevils can develop to adult on three European species closely related to garlic mustard: Peltaria alliacea, Nasturtium officinale and Thlaspi arvense. An open-field test conducted in 2008 with these three species revealed some differences between the two otherwise very similar stem miners. Overall, attack on non-target species was lower for C. alliariae than for C. roberti. Ceutorhynchus alliariae also develops in Rorippa sinuata, a Brassicaceae species native to North America. Subsequent single-choice tests (where the weevil can choose between two plants) revealed that C. alliariae attacked R. sinuata in the presence of garlic mustard as well. An open field test conducted in 2010 revealed however, that attack on R. sinuata is strongly reduced compared to its main host, garlic mustard. No-choice tests in 2010 and 2011 further revealed a limited potential for C. alliariae to develop on Lobularia maritima and Caulanthus inflatus, and for C. roberti on Streptanthus farnsworthianus (L. maritima is an ornamental plant, the latter two are Brassicaceae species native to North America).

Females of the root-mining weevil, C. scrobicollis, lay eggs from mid-September through to April. Impact experiments have shown that C. scrobicollis is the most promising of the prioritized agents. Host-specificity tests revealed that it can develop to adult on the same species as C. alliariae and C. roberti (see above). Tests with C. scobicollis have however revealed that R. sinuata is not attacked under single-choice conditions; we therefore believe that the risk of attack by C. scrobicollis under field conditions is very low. A petition for field release for C. scrobicollis in North America has been prepared in cooperation with our US collaborators and was submitted in July 2008. In a reply delivered by the Technical Advisory Group (TAG) in January 2009, seven reviewers recommended the agent to be released; three indicated approval for release with reservation, and two asked for additional information. To address the concerns expressed by reviewers, extra tests with plant species from additional tribes including endangered species in North America and representative species from tribes containing species sold in the nursery trade. These were conducted at the University of Minnesota, USA and at CABI. Tests confirmed C. scrobicollis as a strong biological control candidate. The species was not able to complete development under no-choice conditions on any of the ten additional plant species tested. Ceutorhynchus scrobicollis further, was not able to complete development in Nasturtium officinale (watercress) in water-saturated soils, i.e. under the conditions N. officinale is grown as a crop. Test results were summarized and a response to TAG was submitted at the beginning of September 2011.

Larvae of the fourth species, C. constrictus, develop in the seeds of garlic mustard, and each destroys 2–3 seeds seeds during development. So far, only Brassica nigra, apart from garlic mustard, has supported complete development to adult. Open-field tests showed however that B. nigra is not a normal field host of C. constrictus. Mature larvae emerged from Brassica juncea as well, however, no complete development to adult was found. In subsequent single-choice arrangements, i.e. when C. constrictus was simultaneously offered test and garlic mustard plants, results varied according to test conditions: no eggs were laid on B. juncea when plants were offered in a field cage, while a few eggs were found on the test plant species when weevils were released into a gauze bag covering one inflorescence (cluster of flowers) each of B. juncea and garlic mustard. Further, Ceutorhynchus obstrictus (cabbage seedpod weevil) was observed on B. juncea plants in the greenhouse. Eggs and larvae found in B. juncea might therefore originate from this weevil known to feed on several other Brassicaceae. To clarify this, eggs and larvae were sent for identification using molecular analyses. At this point, C. constrictus appears to have the most restricted host range of the agents under investigation. An impact experiment testing two different infestation levels revealed that up to 79% of seeds were destroyed on plants onto which three pairs of C. constrictus had been released.

Work in 2012 will mainly concentrate on the seed feeder C. constrictus. Depending on results of molecular analyses, additional single-choice cage tests and/or open-field tests with Brassica juncea will be conducted. Additionally, an initial shipment of C. constrictus to the quarantine facility at the University of Minnesota will be made and additional host-specificity tests (to ensure it only attacks and is effective on the target) on North American Brassicaceae species unavailable at CABI, will be carried out there. We also plan to further explore the acceptance of R. sinuata by C. alliariae and its impact on it.

Selected publications
Rauth, S.J., Hinz, H.L., Gerber, E. and R.A. Hufbauer (2011). The benefits of pre-release population genetics: A case study using Ceutorhynchus scrobicollis, a candidate agent of garlic mustard, Alliaria petiolata. Biological Control, 56, 67-75

Gerber, E., Cortat, C., Hinz, H.L., Blossey, B., Katovich, E., Skinner, L., 2009. Biology and host specificity of Ceutorhynchus scrobicollis (Curculionidae; Coleoptera), a root-crown mining weevil proposed as biological control agent against Alliaria petiolata in North America. Biocontrol Science and Technology, 19 (2), 117-138.

Gerber, E., Hinz, H.L., Blossey, B., 2008. Pre-release impact assessment of two stem-mining weevils proposed as biological control agents for Alliaria petiolata. Biological Control 45, 360–367.

Gerber, E., Hinz, H.L., Blossey, B., 2007. Interaction of specialist root and shoot herbivores of Alliaria petiolata and their impact on plant performance and reproduction. Ecological Entomology 32, 357-365.

Gerber, E., Hinz, H.L., Blossey, B., 2007. Impact of the belowground herbivore and potential biological control agent, Ceutorhynchus scrobicollis, on Alliaria petiolata performance. Biological Control 42, 355–364.

Davis, A.S., Landis, D.A., Nuzzo, V., Blossey, B., Gerber, E., Hinz, H.L., 2006. Demographic models inform selection of biocontrol agents for garlic mustard (Alliaria petiolata). Ecological Application 16, 2399-2410.

Blossey, B., Nuzzo, V., Hinz, H., Gerber, E., 2001. Developing biological control of Alliaria petiolata (M. Bieb.) Cavara and Grande (garlic mustard). Natural Areas Journal 21, 357-367.