So, what's the problem
Dyer's woad (Isatis tinctoria) is a plant of Eurasian origin in the mustard family. It was introduced to North America by early colonists as a textile dye crop and then accidentally spread as a contaminant of crop seed. Today, it is officially recognized as a serious weed in ten western US states.
While many invasive plants need some kind of disturbance to aid establishment and spread, dyer's woad does not. This makes it a particular problem in well-vegetated sites, such as rangelands, because it can readily invade and dominate them.
Control is difficult. Chemical control is limited in rangeland and forests by often inaccessible terrain, the risk of undesirable environmental impacts and questionable economic returns. Cultivation can be effective against the weed in crop areas, but it must be repeated 2–3 times a year for several years to be successful, which is onerous and costly.
What is this project doing?
In 2004, Mark Schwarzländer (University of Idaho, USA) and Jim Hull (Weed Superintendent, Idaho) invited CABI to participate in an initiative to investigate the potential for biological control of dyer's woad.
The ideal biological control agent is very specific, attacking only the target weed. From literature and field surveys in Europe, we identified several potential biological control agents, which we subsequently studied to varying degrees. The aim was to identify species that are both specific to the target plant and cause sufficient damage to indicate they could limit the weed’s impact and spread if released in the USA.
To assess how specific a natural enemy is, we follow recognized protocols. First, we test whether adult females will lay eggs on test plants (no-choice oviposition tests). If they do, we test whether larvae can develop, especially to adult, on plants of this species (larval development tests). If larval development occurs, we conduct tests under more natural conditions either in field cages or in the open field, and give insects the choice of the target weed and the test plants of concern. For promising candidates we also conduct impact experiments to assess what kind and how much damage they cause.
Two potential agents, one weevil and one flea beetle, were determined to be not specific enough to be further considered.
Current work concentrates on two weevils: the seed feeder Ceutorhynchus peyerimhoffi and the root-crown mining C. rusticus. We tested both species on over one hundred plant species, two-thirds of which are native to North America and include species of conservation concern.
C. peyerimhoffi laid eggs and developed in only two non-target plant species in no-choice tests. Unfortunately one of them, Boechera hoffmannii, is a federally listed endangered species. However, when we exposed B. hoffmannii together with dyer’s woad in a cage test, no feeding or eggs were recorded on the endangered species. We plan to conduct additional tests to confirm that C. peyerimhoffi is no threat to it.
C. rusticus developed to adult in seven species, six native to North America, and it developed to some extent in several more species. In field tests, dyer’s woad was highly preferred for egg-laying and we therefore believe that C. rusticus will not have a significant impact on nontarget species.
In impact experiments, C. peyerimhoffi adult and larval feeding destroyed up to nearly 100% of dyer’s woad seeds. Attack by C. rusticus reduced dyer’s woad biomass by some 46%, seed production by 72%, and heavily attacked plant rosettes were killed. A combination of both agents would probably be ideal to control dyer’s woad’s vigour and spread in North America.
We are planning – together with Massimo Cristofaro (Biotechnology and Biological Control Agency, Rome, Italy) and Radmila Petanovic (University of Belgrade, Serbia) – to start working on an eriophyid mite in the genus Metaculus, which is probably a species new to science.
Technical Research Assistant