Natural control

Does it work?

There have been many notable successes in the natural control of weeds. The general consensus is that any contribution to a reduction in current control costs and efforts that outweighs the cost of the programme should be considered a success. A recent review of natural control programmes revealed an average cost: benefit ratio in excess of 1:200 and though there was some variation, all were found to be positive. A good example is the use of the weevil Cyrtobagous salviniae to control Salvinia weed in Sri Lanka. This plant was introduced into the country during WWII to prevent enemy aircraft from identifying waterways. It did the job so well that almost all water-bodies in the country were affected. The weevil was released in 1986 and within four years it destroyed around 80% of weed infestations. Since its discovery this weevil has been a successful control agent against this weed in more than 10 tropical countries around the world and is still working today.

What if it goes wrong?
Of the 1,400 worldwide releases against weeds, less than 1% produced non-target effects and all bar two were predicted by scientists prior to release. However, the predicted damage to, in some cases rare species, was deemed to be an acceptable risk at the time, given the scale of the problems. However, given the same data and situation, it is highly unlikely that today’s decision makers would authorize such a release.

The care taken prior to releasing a natural control agent is in stark contrast to the wholesale importation of exotic species, either as pets or garden plants and their pests, or as stowaways within many commodity crops.

Why can’t we just leave things as they are?
Himalayan balsam and other non-native species have one thing in common; they were introduced into the UK without the suite of natural enemies which keep the plant in check in the native range. This gives Himalayan balsam an unfair advantage over their native counterparts and enables them to become highly invasive species out-competing our native flora and degrading natural habitats. Just leaving things as they are will just enable Himalayan balsam to spread further and degrade additional areas. Integrating natural control methods with more traditional control methods should be the goal to best reduce the economic and ecological impacts of invasive weed species. In the early days of the release of the rust against Himalayan balsam, the rust will be establishing and spreading through the population. Therefore, there will still be a need to apply traditional control methods especially if Himalayan balsam is encroaching on sensitive habitats. Natural control should act to reduce the vigour and spread of Himalayan balsam over time, making it easier to control Himalayan balsam with conventional means.

Is there a danger of releasing another alien species that will become a problematic invasive?
The proposed control agent will only be released after this vigorous testing, peer review and public consultation. It has been tested on 56 plant species to ensure that it does do not attack other plants.

Whilst the natural control agent is not native to the UK, it is ‘native’ to balsam as a host. Only co-evolved natural enemies are considered as control agents and some of these will have sacrificed their ability to feed on other species in order to specialize on the target weed. These are the agents that biocontrol scientists select. By applying internationally-accepted safety testing procedures to a selected agent, it is possible to demonstrate that the risk to UK native biodiversity or crops is negligible and the organism will not be harmful.

Is this anything like GM?
No. Genetic modification involves human intervention to provide an organism with certain genes that code for desirable traits. In the case of natural control ‘Mother Nature’ has done the modifying for us through the process of 1,000’s of years of evolution. Natural control aims to allow a natural balance to be restored by the re-association of an invasive plant and its natural control agent.

How can it be that these natural enemies only attack one plant?
Most fungal and insect species are restricted in what they are able to infect or feed on, and some are monophagous, meaning they will only attack one species of plant. This is not as surprising as it seems. Many endangered insects are under threat because their only host plant or its habitat has become rare and they are unable to feed on anything else. Of course some insect and fungal species will feed or infect a wide range of plant species but these are rejected as potential natural control agents early in the safety testing regime.

Will releasing a natural control agent eradicate Himalayan balsam in Great Britain?
No. Eradication is not the normal outcome of natural control since it is not in the agent’s best interest to eradicate its only food source. Long-term control, below an economic or environmental threshold, should be anticipated. The aim of the biological control programme is to reduce the occurrence Himalayan balsam on our river systems and areas with high conservation status. By reducing the occurrence of Himalayan balsam on rivers, this will reduce the impact threshold of the plant on native biodiversity.

Is this like the azolla weevil?
In some way yes, although it was not introduced as part of a natural control programme, as it was in South Africa. The azolla weevil (Stenopelmus rufinasus) was an accidental but fortuitous introduction from North America, that has become ‘ordinarily resident’ in the UK. The weevil is now currently being redistributed to control the problematic aquatic weed Azolla filiculoides on ponds and slow-moving rivers. (see

Will anyone make money out of a release?
No. Natural control projects are activities for the ‘public good’ and there is no money to be made. CABI is a not-for-profit organization and has only received the funding required for the research phase of the programme from the consortium of sponsors.

Are you looking at using natural control for any other invasive species?
Natural control could be considered for any non-native invasive species providing that its impact is deemed sufficient for the research to be justified. In the plant world, CABI is currently working on Japanese knotweed (Fallopia japonica), floating pennywort (Hydrocotyle ranunculoides) and Australian swamp stonecrop (Crassula helmsii). CABI’s team in Switzerland are also considering using natural control against insects such as the Lilly leaf-beetle (Lilioceris lilii) and the horse chestnut leafminer (Cameraria ohridella) as well as Ambrosia weed (Ambrosia artemisiifolia).

Has biological control been considered for weeds in the UK before?
Yes. In 2010 the Japanese knotweed psyllid (Aphalara itadori) was released against Japanese knotweed (Fallopia japonica) as part of a staged release plan. It was the first intentional release of a non-native organism to control an invasive non-native plant in Western Europe. However, there is a long history of using natural enemies to control insect pests in Europe. An example is the release of a specialist predator to control the invasive spruce bark beetle in many EU countries which to this day is still providing a long-term solution to this economic pest. In addition, biocontrol agents are regularly used on a smaller scale within the horticultural industry and even in householders’ glasshouses for example to control aphids, whitefly, mites and other crop pests.

Are there any other examples of where non-native species have been used to control other non-native species in the UK?
Yes. There is an example regarding an animal species. The non-native predatory beetle, Rhizophagus grandis, was released under licence in the mid 1980’s to tackle the invasive non-native spruce bark beetle (Dendroctonus micans). The predator is now well established and follows the spruce bark beetle as it spreads, keeping the latter under control. Occasional releases are made where distribution of the spruce bark beetle “jumps” e.g. is accidentally transported some distance. This is an example of natural, or biological control as a non-native species was deliberately introduced in order to tackle another invasive non-native species.

What about the rest of the world?
There have been over 1,400 releases of natural control agents against weeds around the world. In all, more than 400 different agents have been released against more than 150 different target weeds. CABI is a world expert in natural control and played a vital role in the development of an International Code of Conduct on the use of it. CABI also has a very good international track record of researching and releasing natural control agents. In countries like the US, Canada, Australia and New Zealand natural control is often viewed as the first line of defence when a new and problematic invasive species is identified.

What about the cane toad?
The cane toad (Bufo marinus), was introduced to Australia by the sugar cane industry in 1935 in an ill-judged attempt to control pest beetles. This was done against the recommendations of scientists at the time. Thousands of toads were released without any host specificity testing and not only failed to control the beetle but turned their carnivorous attention to anything that moved and was small enough to be swallowed. They went on to become a significant problem themselves. Although carried out in the name of biological control, today’s practitioners consider this unfortunate case to have been a highly irresponsible act in an age when there was no real regulation and safety testing. Today’s pest risk analyses should prevent such a thing happening again.