Cookies on CABI

Like most websites we use cookies. This is to ensure that we give you the best experience possible.


Continuing to use  means you agree to our use of cookies. If you would like to, you can learn more about the cookies we use.

Search this site
Sign up for the CABI e-zine Newsletter
Improving lives by solving problems in agriculture and the environment

Biological control of garlic mustard

Crushed garlic mustard leaves and seeds smell like cultivated garlic and have been used as flavouring in cooking for centuries. Garlic mustard is a brassica from Eurasia that was accidentally taken to North America and became invasive in many of its forests. Together with partners, CABI is exploring the possibility of using specially selected and tested insects from the native range in order to safely control the plant’s spread and impact in the introduced range.

Project Overview

So, what's the problem

Garlic mustard (Alliaria petiolata) is not related to garlic but is a biennial cruciferous (brassica) plant. Native to Eurasia, it was accidentally introduced to North America in the 19th century. It is one of the few non-indigenous herbaceous species that can invade and dominate the understory of North American forests. It is considered one of the most serious invaders in the northeastern and midwestern USA and southeastern Canada. Several methods have been used to control its proliferation in natural areas; hand removal can be effective in small infestations, while fire, cutting and herbicide treatments have been used to reduce densities in large infestations. However, these treatments are costly, need to be repeated over several years and may face regulatory restrictions.

What is this project doing?

This project to investigate the potential for biological control of the weed was initiated in 1998 by Prof. Bernd Blossey (Cornell University, USA). A team from CABI’s centre in Switzerland has been surveying for natural enemies and assessing host specificity of selected insects.

Biological control is based on the concept that a plant may become invasive because of the absence of natural enemies that keep it in check in its area of origin. The project aims to identify and introduce host-specific natural enemies as biological control agents. A guiding principle is that an agent should not impact plants other than the target. Risk of non-target damage is assessed by testing whether a potential agent feeds or develops on other plant species. 

By reviewing the literature, we found records of 69 herbivorous insect species and seven fungi associated with garlic mustard in Europe. 30 species were collected in subsequent field surveys in Switzerland, Germany, Austria and Daghestan (Russia) in 1998–2000. Six insects were prioritized as potential biological control agents: the root-feeding flea beetle Phyllotreta ochripes, the two shoot-mining weevils Ceutorhynchus alliariae and C. roberti, the two seed-feeding weevils C. constrictus and C. theonae and the root-crown weevil C. scrobicollis.


We have so far discarded three species: the flea beetle P. ochripes and the stem-mining weevil C. alliariae which were not sufficiently specific. The seed-feeding weevil C. theonae was difficult to find and rear while we suspended work on the stem-mining weevil C. roberti to concentrate on other species.

The root-crown weevil C. scrobicollis was specific enough and after several attempts to apply for permission to release it, the USDA-APHIS Technical Advisory Group (TAG) recommended field release in February 2017. The weevil now has to pass further US environmental regulations (section 7 consultation with US Fish and Wildlife Service, tribal and public consultations etc.) before being permitted for release.

A petition to also release the weevil in Canada was submitted in December 2017. Data from Esther Gerber’s PhD study and modelling by Adam Davis (USDA-ARS, Illinois) showed the agent’s potentially large impact on garlic mustard demography.

The seed-feeding weevil C. constrictus seems the most specific agent. Of 103 species and subspecies tested, it developed to adult or mature larva on Brassica nigra (black mustard), B. juncea (brown mustard) and the target weed in the laboratory, but attacked only garlic mustard in field trials. It is also effective as our experiments demonstrated that 79% of garlic mustard seeds were destroyed.

We are completing testing in collaboration with the University of Minnesota.

The team

Project Manager

Staff image of Ghislaine Cortat

Ghislaine Cortat  Research scientist in Weed Biological Control

Rue des Grillons 1 CH-2800 Delémont
T +41 (0)32 4214879

Project team

Staff image of Ivo Toševski

Ivo Toševski

Research Scientist, Weed Biological Control

Staff image of Hariet L Hinz

Hariet L Hinz

Country Director and Head Weed Biological Control