So, what's the problem
Field bindweed (Convolvulus arvensis) is a perennial vine of Eurasian origin. The plant was introduced to other continents including North America and is now considered one of the most noxious weeds of agricultural fields throughout temperate regions. Forming dense tangled mats of vegetation, it outcompetes native forbs and grasses, and can severely reduce crop yields. It can harbour plant diseases and is toxic to horses. Its extensive root system and long-lived seeds make it difficult to control by conventional means. Biological control offers an alternative approach: one reason for the plant’s impact may be the absence of natural enemies that attack it in its area of origin.
The weediness of field bindweed is largely attributable to its extensive root system and North American insects attacking the leaves are having little impact. So the United States Department of Agriculture (USDA) initiated a programme to manage field bindweed using biological control in the 1970s. Two biological control agents were introduced from Europe: the gall mite Aceria malherbae and the bindweed moth Tyta luctuosa. The impact of the gall mite varies, while local establishment of the bindweed moth has only recently been reported from the western USA and its impact is not yet clear. The weed continues to be a problem and additional biological control agents are being sought.
What is this project doing?
The project is being revisited through an initiative set-up by the late Dr Richard Hansen (USDA-APHIS-CPHST). CABI’s centre in Switzerland is investigating additional potential agents. So far, we have studied five insect species that showed potential for biological control, and are planning work on two more species.
A guiding principle for biological control is that any released agent should not impact plants other than the target weed. Risk of potential non-target damage is assessed by testing whether a candidate agent feeds or develops on other plant species that it might encounter if introduced. Assessing the impact an agent might have on the target weed is also important so the most-damaging ones can be prioritised.
We have already rejected two root-feeding flea beetles, Longitarsus pellucidus and L. rubiginosus, which proved insufficiently specific. We also rejected two leaf-feeding species, the moth Emmelia trabealis and the tortoise beetle Hypocassida subferruginea. Laboratory tests left questions about the host-specificity of the stem-mining agromyzid fly, Melanagromyza albocilia. In no-choice tests (offering one plant species at a time), itlaid eggs on six plant species, including four native to North America, and larvae were found on three of these, all in the genus Calystegia. Moreover, M. albocilia completed its development to adult on the North American native C. macrostegia. To assess its specificity under more natural conditions, we exposed three native Calystegia species that were attacked under no-choice conditions, as well as Ipomea batatas (sweet potato) and the target C. arvensis in an open-field test in southern Germany. Signs of both feeding and larval development were only found on the target plant. We are planning to set up further open-field tests with additional plants that were attacked in the lab.
We started investigating the root-mining clear-wing moth Tinthia brosiformis, which is only recorded from C. arvensis. Larval feeding can cause the plant to dieback. A rearing colony was established at the Institute for Plant Protection and Environment in Serbia, and methods will be developed to conduct host-specificity tests with this moth.
Research Scientist, Weed Biological Control