Environment Prize Recognizes Biocontrol's Diversity

Amongst the three recipients of the 2003 Tyler Prize for environmental achievement are two leading lights of the biocontrol world, Hans Herren of ICIPE (International Centre of Insect Physiology and Ecology, Kenya) and Yoel Margalith (Ben-Gurion University of the Negev, Israel). (The third recipient of the Tyler Prize is the pioneering cancer scientist, Sir Richard Doll.) Although both Herren and Margalith are well known and respected for subsequent work, the central achievements for which they are honoured illustrate two different biological control approaches that have led to progress in securing the world's food supply and human health over the last three decades:

• Classical biological control of an exotic pest through the introduction of a natural enemy from their common area of origin
• Inundative biological control using an entomopathogen as a microbial agent, which may be formulated and applied as a biopesticide, against an indigenous or exotic disease vector

During the 1980s, Hans Herren, then at the International Institute of Tropical Agriculture (IITA), led the team that coordinated the implementation of one of the largest and most successful classical biological programmes in the world. The Africa-wide Biological Control Program was responsible within Africa for introducing the South American encyrtid parasitoid, Anagyrus lopezi (= Epidinocarsis/Apoanagyrus lopezi), to control the cassava mealybug, Phenacoccus manihoti , which was devastating cassava production across vast areas of Africa at that time. Uncontrolled, it was feared that the pest, which had been accidentally introduced from cassava's own home range in the New World, would cause catastrophic losses to a crop that has vital significance as a security food crop in Africa. Widespread famine affecting up to 200 million people across tropical and subtropical Africa was feared. At first, headline-catching touches, such as distributing the natural enemies by a low-flying light aircraft, received more publicity than the spread of the biocontrol agent and its impact on the pest and cassava yields. The scientific evidence was clear, however, and public acceptance followed when, within 12 years, the formerly devastating pest was controlled across all of continental Africa. In a rare `silver bullet' success story, control has been sustained ever since, with this one natural enemy keeping the pest in check. However, the programme was also responsible for introducing the concept of biological control to many countries. The outstanding success of the programme was instrumental in facilitating acceptance of this approach in Africa, and contributed substantially to building the capacity that enabled countries to use biological control against other pests.

In 1976, during a World Health Organization sponsored project in Israel, Yoel Margalith discovered a strain of Bacillus thuringiensis in the Negev Desert which was significantly more toxic to mosquitoes than other known bacterial strains at that time. Later identified as B. thuringiensis ssp. israelensis (Bti) , it proved to have that useful combination of being both lethal and relatively specific to dipteran immature stages. In particular, it was highly pathogenic to culicids and simuliids. Its potential for controlling the vectors of diseases such as malaria, dengue, yellow fever and river blindness (onchocerciasis) was quickly recognized. Its commercial potential has since been exploited and many Bti -based products are now registered. The use of Bti transformed mosquito and blackfly control in many countries, with extensive control programmes based on Bti implemented in West Africa, the USA and Europe. For example, Bti was used against river blindness along the Volta River in eleven African countries. The sight of millions was saved and repopulation of deserted river valleys initiated. What has made Bti even more useful is that little resistance has been reported, perhaps owing to synergistic action of the complex of toxic proteins. Use of Bti led to malarial infections from mosquitoes resistant to other pesticides dropping by 90% along the Yangtze River, China, which has a population of over 20 million people. Thus Bti has had an enormous effect on human health and environmental quality and, with increasing global travel and the emergence and spread of diseases such as West Nile virus to new regions, the existence of this environmentally benign yet effective tool assumes new significance.

❑

When the Mist Clears, What Will We See?

One of the perceived disadvantages of weed biological control is that because agents are generally chosen that will target a single weed at a time, there is a danger that the plant targeted will simply be replaced by other unwanted exotics. Therefore it is important not only to demonstrate that a weed has been reduced and maintained below a desired threshold, but also to show the end result that is really wanted: that the weed has been replaced by more desirable vegetation. This will be the true measure of whether or not a weed biological control project has been successful.

Mist flower (Ageratina riparia, Asteraceae) is an aggressive and fast growing weed originating in Central America. It's a perennial herb or sub-shrub, 0.3-2 m tall which produces masses of small white flowers that develop into highly mobile seeds. The plant is particularly problematic in wet areas (such as riverbanks) in tropical and warm temperate regions including northern Australia, Hawaii and South Africa. In New Zealand it has invaded pastures and native forests in the North Island. Landcare Research has been conducting an active biological control programme against this weed. Following the example of a successful biological control programme against mist flower conducted in Hawaii, two natural enemies of the weed: the white smut fungus Entyloma ageratinae , and the gall fly Procecidochares alani , were introduced into New Zealand in 1998 and 2001 respectively.

In 1999 Landcare Research established a small, multi-year project, in an area of native forest near Auckland (the Waitakere Ranges), to look at what plants replaced mist flower as its cover (hopefully) declined. Small (4 m²) permanent monitoring plots were established in the summer of 1999/2000, some with mist flower and others without the weed. In the plots with mist flower, the health of the weed declined between 1999/2000 and 2001/2002, with the percentage of leaves infected by the white smut fungus increasing from 18% to 62%, and the total percentage of dead leaves increasing from 8% to 23%. During these two years the total percentage cover of mist flower decreased from 74% o 16%, a decline of more than 50%. We attribute this decline in cover to defoliation caused by the white smut fungus: the gall fly had not yet reached the plots and there appeared to be no other environmental or management changes that could have caused such a dramatic loss of foliage.

When the plots were first examined, in 1999/2000, there were found to be significantly fewer native plant species in the plots with mist flower than in the plots without it. In contrast, there was no significant difference between the number of exotic species (excluding mist flower) in plots with mist flower compared with those without the weed. This appears to confirm the view of land managers that mist flower was having a negative impact on the regeneration of native species while not inhibiting other exotics. At least the presence of mist flower was not correlated with a higher number of other exotic species.

Encouragingly, we didn't find more exotic species arriving in plots with mist flower as the cover of the weed declined. Even better, the change in the number of native species was significantly different between the plots with or without mist flower: The number of native plant species present increased in plots with mist flower (as the weed declined in percentage cover over the 2 years). In contrast, there was a small unexplained decrease in the number of native plant species in the plots without mist flower. Overall, it appears that the biological control of mist flower is benefiting native species rather than other weedy exotics.

Further information: Fröhlich, J.; Fowler, S.; Gianotti, A.; Hill, R.; Killgore, E.; Morin, L; Sugiyama, L.; Winks, C. (2000): Biological control of mist flower: transferring a successful programme from Hawai'i to New Zealand. In: Spencer, N.R. ( ed ) Proceedings of the X International Symposium on Biological Control of Weeds. Bozeman, Montana, USA, 4-9 July 1999, pp. 51-57.

By: Jane Barton (née Fröhlich), 353 Pungarehu Rd, RD 5, Te Kuiti, New Zealand
Email: [email protected]

Jane Barton works under subcontract for Landcare Research. The project was supported by Auckland Regional Council, with contributions from Northland Regional Council, the Department of Conservation, Environment Waikato and the New Zealand Government. Auckland University students (Jonathan Boow, Krystian Ragiel and Kate Edenborough) carried out most of the fieldwork.

❑

Biocontrol of Invasive Insect Threatening Galápagos Islands

In January 2002, biological control was used for the first time in the Galápagos Islands with the release of the Vedalia beetle Rodolia cardinalis to control the invasive cottony cushion scale, Icerya purchasi. In the presence of town officials and with the collaboration of college students, beetles were released simultaneously on the inhabited islands of Santa Cruz, San Cristòbal, Isabela, and Floreana. The release programme was a joint effort between the Galápagos National Park Service (GNPS) and the Charles Darwin Foundation (CDF) and followed a 6-month intensive educational campaign to inform the public of the threats of invasive species and of the rigorous studies carried out to evaluate the safety of introducing this biological control agent. Since then, beetles have been released in priority areas on the islands of Marchena, Fernandina, Pinta, Pinzòn and Rabìda where endangered species of plant are seriously affected by I. purchasi. Since releases began, just over 1500 adult beetles have been released.

This is the culmination of a 6-year research programme by the Department of Terrestrial Invertebrates of the Charles Darwin Research Station (CDRS) to evaluate the risks associated with using biological control to mitigate the impacts of I. purchasi. In 1996, serious outbreaks of I. purchasi on several islands alerted the GNPS and CDF to the threats of this pest to plant species of conservation value. Chemical control was quickly ruled out as a possibility for reducing pest numbers and biological control was considered the only viable solution. Legislation for the release of exotic natural enemies in the Galápagos is only now being developed, and at this time responsibility for importation decisions lay with the GNPS. Consequently, a technical advisory committee was formed with ten members of the CDF and GNPS to evaluate the possibility of employing biological control for the first time on the Galápagos Islands. The committee concluded that given the immediate threat of the cottony cushion scale to rare or endangered species an evaluation of the risks associated with the introduction of R. cardinalis should be carried out at the same time as studies to confirm that the impacts of I. purchasi merited the introduction of this biological control agent.

A risk assessment methodology was drawn up based on the guidelines of the FAO's Code of Conduct for the Import and Release of Exotic Biological Control Agents. This was expanded on and procedures were developed to address three key questions:

• Does I. purchasi weaken plant growth and contribute to the mortality of native flora and fauna in the Galápagos?
• Are there any natural enemies of I. purchasi already present in the Galápagos that could be used to suppress this insect?
• Will R. cardinalis have any detrimental effects on nontarget organisms or the environment in general?

Field surveys and experimental studies demonstrated that high infestations of the cottony cushion scale were influencing the survival of native plant communities including threatened species or their habitats. However, the paucity of baseline data on these endangered plant species, and the effects of other stress factors such as lack of water and nutrients, made it difficult to isolate the effects of I. purchasi in the field. Since it was introduced in 1982, the cottony cushion scale has colonized 15 islands in the archipelago where it attacks 31 endemic species and 31 native plant species, 16 of which are threatened according to the IUCN (World Conservation Union) criteria. Mortality has been recorded on nine endemic and 10 native species so far, including populations of threatened species such as the Critically Endangered daisy tree Scalesia atractyloides and the white mangrove, habitat of the Critically Endangered mangrove finch. One species has been reclassified as threatened explicitly as a result of damage caused by I. purchasi. Furthermore, local extinctions of specialist endemic arthropod fauna dependent on threatened plant species have been observed.

No monophagous natural enemies of I. purchasi were found in the Galápagos, eliminating the possibility of using augmentative biological control. The cottony cushion scale has been successfully controlled in 60 countries by R. cardinalis, but virtually nothing was known about the feeding range of this biological control agent and whether it has had any impact on native fauna. Consequently, tests were deemed necessary to evaluate whether R. cardinalis would use Galápagos invertebrates as alternate prey. Beetles were donated by CSIRO Entomology (Brisbane, Australia) and no-choice tests carried out in a newly constructed insect containment facility at the Charles Darwin Research Station [See also BNI 20(3), 71N (September 1999), Host specificity testing of Rodolia cardinalis... one hundred years late? ]. Immature stages of R. cardinalis were unable to complete development or feed on a wide range of prey species. Similarly, adult R. cardinalis were unable to use a small range of Homoptera as temporary sources of food. Rodolia cardinalis was only able to feed on Margarodes similis, the closest relative to the cottony cushion scale, but this species is subterranean and would not be at risk from predation.

At the request of the advisory committee additional research was carried out to confirm experimentally that the beetle would not impact insectivorous vertebrates such as the finches and other small birds. This was because R. cardinalis adults reflex bleed heavily from their joints producing a haemolymph that might contain a toxic alkaloid. Experimental trials were carried out on two species of Galápagos finch, but neither species showed adverse reactions after being fed R. cardinalis.

A risk analysis was presented to a technical advisory committee in 2001 and it was concluded that sufficient evidence existed to demonstrate the costs and benefits of liberating R. cardinalis into the Archipelago. A post-release monitoring programme involving the participation of the local community has been initiated to determine whether the beetles have established and are reducing cottony cushion scale numbers on threatened plant species. Additionally, information is being gathered on the feeding behaviour of the beetle. This information is particularly important as it will determine whether the beetle is having any negative impact on the Galápagos biota, which in turn will allow us to evaluate whether the risk assessment was accurate in its predictions.

By: Charlotte Causton, Research Associate, Department of Terrestrial Invertebrates, Charles Darwin Research Station, A.P. 17-01-3891, Quito, Ecuador
Email:

❑

Tropical Soda Apple Biocontrol Bears First Fruit

Tropical soda apple (TSA), Solanum viarum, also known as `the plant from hell', is a perennial prickly bush (family: Solanaceae) native to Brazil, northeast Argentina and Paraguay. It has been spreading rapidly in the USA since it was discovered in Glades County, Florida in 1988. TSA has become a serious weed of pasture and natural habitats, displacing other vegetation and forming impenetrable thickets. In 1992 approximately 150,000 acres (60,700 ha) of pasture-land was estimated to be infested and economic losses to Florida cattle ranchers at this time were put at US$11 million (or 1% of total state beef sales). Ten years on, the infested area has increased to more than one million acres (some 404,700 ha) of improved pastures, citrus groves, sugarcane fields, ditches, vegetable crops, sod (turf grass) farms, forest lands (including oak- and cypress-dominated tree islands in pasture, which provide shade for cattle), and natural areas. Carrying-capacity falls in TSA-infested pasture both because the plant's foliage is unpalatable to cattle and because dense TSA stands prevent cattle access to shade. TSA also causes as-yet unquantified losses to vegetable growers (a sector worth some $1.7 billion annually in Florida) because six plant viruses are transmitted by insect vectors from TSA to important solanaceous crops.

Although probably first introduced to Florida by unwitting human vectors (likely pathways are on footwear or through escape from cultivation), the rapid spread of TSA in the southeastern USA is attributed to its tremendous reproductive potential and ease of spread. It produces high numbers of viable seeds and can regenerate vegetatively from an extensive root system. Spread is aided by cattle and wildlife which ingest the seeds. The weed can also be spread in TSA-contaminated hay, sod, seed, soil, compost, machinery and running water.

TSA has been reported in nine other states (Alabama, Georgia, Mississippi, Louisiana, Texas, North Carolina, South Carolina, Tennessee and Pennsylvania). The weed has the potential to expand its range even further in the USA, based on temperature and photoperiod threshold experiments conducted in controlled environmental chambers. TSA was placed on the Florida Noxious Weed List in 1994 and the Federal Noxious Weed List in 1995, and it is listed as one of the most invasive species in Florida by the Florida Exotic Pest Plant Council.

The Tropical Soda Apple Task Force was formed to develop appropriate control strategies for this unwanted biological pollutant. The Task Force, composed of research, regulatory and industry expertise, developed recommendations for research, education/awareness and regulatory programmes. It has been active in assisting with the development of Best Management Practices (BMPs), locating funding for research, and ongoing industry/public awareness of the importance of taking appropriate action to control TSA. Regulatory measures have gone in the direction of TSA management using established BMPs, voluntary compliance and contract specifications requiring regulated articles to be free from TSA (e.g. sod has to be certified TSA-free). The eradication of TSA from Florida is not feasible owing to the general infestation of the state and lack of tools and resources. However, the management of TSA is achievable with the application of biological controls as the best long-term strategy to reduce the impact to an acceptable level.

Until now, three types of control have been used to limit the spread of TSA: chemical, mechanical, and regulatory. Herbicides and mowing only provide temporary weed suppression and, in addition to being expensive, they are not always practical in inaccessible areas. Moreover, herbicides can have negative environmental effects, which include leaving undesirable chemical residues in the ecosystem and in commodities, and adversely affecting non-target organisms. Several southern states are trying to prevent the spread of TSA by means of regulatory control, regulating the movement of cattle, hay, sod, manure, seed and soil from infested areas to areas free of infestation. For example, cattle are held in an area free from TSA fruit for a 5- to 7-day period to allow for the seed to pass through the rumen.

A biological control project on this invasive non-native weed was initiated in January 1997 by University of Florida researchers in collaboration with Brazilian and Argentinian researchers. The Florida Department of Agriculture & Consumer Services - Division of Plant Industry and the USDA-APHIS-PPQ (US Department of Agriculture - Animal and Plant Health Inspection Service - Plant Protection and Quarantine) have been providing funding for the project.

From exploratory surveys conducted in South America, several insects were identified as potential biological control agents of TSA. Two chrysomelid leaf beetles, Gratiana boliviana and Metriona elatior were initially selected for screening because of the extensive plant defoliation attributed to these beetles in their native range. Three other promising candidates currently undergoing host range determination in Florida-quarantine are two more chrysomelid leaf beetles, Platyphora sp. and G. graiminea, and a curculionid flower bud weevil Anthonomus tenebrosus.

Gratiana boliviana was approved for field release in the USA by TAG (Technical Advisory Group for Biological Control Agents of Weeds) in April 2002. APHIS review and analysis of the potential environmental impacts associated with releasing G. boliviana into the environment are documented in detail in an Environmental Assessment (EA) (February 2003). A high level of specificity and significant defoliation of TSA were demonstrated in host-specificity tests conducted at the Florida Biological Control Laboratory quarantine in Gainesville, the USDA-ARS (Agricultural Research Service) quarantine in Stoneville, Mississippi and the USDA-ARS South American Biological Control Laboratory in Hurlingham, Argentina, and in extensive field surveys and open-field tests conducted in South America. Field releases of G. boliviana in the USA are planned for May 2003.

A virus indigenous to Florida is also showing promise as a biocontrol agent. The tobacco mild green mosaic tobamovirus (commonly called the tobacco mild green mosaic virus, TMGMV) is a common virus that produces a mild mosaic disease in tobacco (Nicotiana tabacum). In TSA, however, it has a deadly impact. Young plants are killed most rapidly but even mature plants succumb. Typically plants die within 14-21 days of being inoculated with the virus. In field trials 83-97% control (complete kill) was achieved for plants of different sizes and ages.

The ease with which the virus multiplies in tobacco provides an excellent basis for mass-producing it. In addition, application of the virus is simple: leaf extract prepared from infected plants is applied to a few leaves of the target plant either by hand or with a simple mechanical implement. However, further development as a biocontrol agent means tackling host specificity issues. Literature records indicate that the virus infects at least 15 species in four families. It has been recorded causing severe symptoms in only a few species, but these include several Capsicum pepper varieties. This would preclude the use of the virus in these crops, and further screening is assessing whether any other cultivated plants are susceptible. However, it is possible to use the virus in TSA-infested fields without risk of wide dispersal. The fact that TMGMV is not an insect-transmitted virus, and therefore is unlikely to have uncontrolled secondary spread from infected plants, is an important safety feature.

Development of the virus as a bioherbicide involves registration by the US Environmental Protection Agency (EPA), a process which may take about 2 years. However, with the chrysomelid beetle about to be released and this promising bioherbicide in development, the University of Florida biocontrol programme is making significant progress against this noxious weed.

Contacts:
Julio Medal, University of Florida
Email:

Raghavan Charudattan,
University of Florida
Email:

Daniel Gandolfo, USDA-ARS South American Biological Control Laboratory
Email:

Jeff Mullahey, University of Florida - West Florida REC Director, Milton
Email:

Richard Gaskalla, FDACS-Division of Plant Industry Director
Email:

James Cuda, University of Florida
Email:

❑

Bug for Bellyache Bush Biocontrol

A seed-sucking bug from South America is set to become the latest Neotropical biocontrol agent to be released against invasive rangeland shrubs in Australia. Bellyache bush (Jatropha gossypiifolia) is a noxious weed across a large proportion of northern Australia. It is naturalized in tropical Queensland, Northern Territory and Western Australia, and its potential distribution includes enormous tracts of these states. Native to the New World tropics and a member of the Euphorbiaceae, it has been introduced to other regions as an ornamental or medicinal plant and has been declared a weed in a number of countries. In Australia bellyache bush invades rangeland, particularly in riparian areas, forming dense thickets which eliminate other more useful species. In addition, all parts of the plant are highly toxic. Although in low doses it is reputed to have medicinal value, in Australia it has been cited as the cause of death of over 300 grazing animals in one shire in one year alone.

In 1997, the Northern Territory Government began funding a CSIRO (Australia's Commonwealth Scientific and Industrial Research Organisation) Entomology project aimed at the weed's biological control, and further funding from the Queensland Department of Natural Resources was made available from 1998. Surveys for natural enemies conducted in eight countries in tropical America and the Caribbean (Mexico, Honduras, Guatemala, Venezuela, Dominican Republic, Puerto Rico, Trinidad and Curacao) led to 60 species of phytophagous insects and one rust fungus being identified from the plant, and promising (i.e. damaging) candidates were selected for further study. Host specificity testing at CSIRO's Long Pocket Laboratory's quarantine facility on more than 70 species of Euphorbiaceae ruled out some of the potential agents, but a species of seed-feeding scutellerid bug, Agonosoma trilineatum, proved to be highly specific. These tests, which looked at the potential for development of nymphs, and adult oviposition preferences, mating and feeding on different plant species, indicated that only species of Jatropha could act as a host for A. trilineatum (four other species of this genus occur in Australia, all originally introduced as ornamentals), but only J. gossypiifolia was accepted for oviposition (with one minor exception). In addition, adult A. trilineatum mated and fed only on the host plant. An application for release of A. trilineatum was submitted in June 2002, and permission to release has now been granted. Insects are being mass reared, and first releases are planned for both Queensland and the Northern Territory in May 2003.

Immature and adult A. trilineatum feed only on seeds, and in laboratory studies adults proved capable of completely destroying them. In the field, where seeds are more abundant, the impact is as yet unknown but it is hoped that insect populations will reach high levels, and that under these conditions seeds will be killed or at least damaged and their weight and viability reduced. Bellyache bush is spread by seed, so a reduction in seed production should reduce the rate of spread and the rate of recruitment in established infestations. In addition, as the seeds are the most toxic part of this plant, poisoning of livestock may be reduced. However, a single agent species is not expected to provide complete control, and research on other potential agents is continuing. Successful establishment of a diet-based rearing method for a new Lagocheirus species of cerambycid stem weevil means that testing of that species can begin in the Queensland quarantine facility. In addition, the stem boring weevil Cylindrocopturus jatrophae has also been imported for screening in quarantine. Two further species, a stem-mining tineid moth, Xylesthia sp., and an unidentified brentid beetle have been earmarked as potential agents. Elsewhere, the rust fungus has been identified as Phakopsora jatrophicola by CABI Bioscience (UK Centre). A preliminary study showed that this rust has some potential as a biocontrol agent.

In order to be able to measure the impact of this and future agents, Northern Territory and Queensland government field scientists are collecting baseline ecological parameters for comparison with data after the agents are established.

To improve future agent selection, a postdoctoral researcher funded by the Cooperative Research Centre for Australian Weed Management, S. Raghu, is taking an interesting approach. He is evaluating the ability of the weed to respond to different types of simulated herbivory. Agents utilizing the parts of the plant most susceptible to stress, resulting in a negative impact to the plant, will be the focus of subsequent investigations into their suitability and safety for biocontrol.

Contact: Tim Heard, CSIRO Entomology, Long Pocket Labs, 120 Meiers Rd,
Indooroopilly, Qld 4068, Australia
Email: Fax: +61 7 3214 2885

❑

Australia Saddled with another Invasive Weed

Camels brought to Australia during a 1930's gold rush are blamed for the introduction of a drought-resistant tree, which is thriving in the current drought and spreading through large areas of northern Australia's rangelands.

Calotrope, Calotropis procera, is a small tree rising to 3-5 m in height. It originates in arid regions of Asia and Africa and is believed to have been brought to Australia in camel saddles, in which the kapok-like seed comose (tuft of hair) was used as padding. The camels and saddlery were brought from India to Chillagoe in northeast Queensland during a gold rush in the 1930s. When the saddles broke down, seeds, which had been inadvertently included, were released.

The plant has large oval leaves. They are a dull, greenish grey and are coated with a waxy bloom. There are relatively few leaves per branch and these are restricted to the ends of the branches. The branches are brown when young. The stems become corky with age, taking on an ash grey colour. Calotrope plants grow rapidly, reaching more than a metre in the first season and reaching full height in 3-4 years. At maturity the tree has 3-5 stems, which have few branches. The trunk is 40-130 mm in basal diameter when mature.

Towards the end of the second growing season, bunches of attractive white flowers with purple tips appear at the ends of each branch. Thereafter flower production is almost continuous. Fruiting pods are mango like in appearance. They are thin-skinned and mostly full of air. The seeds are produced along a central structure. They are initially white, maturing to dark brown. Once mature, the pod splits, releasing about 300 seeds. The comose or tuft of hair, at one end of the seed, aids in wind and water dispersal.

Calotrope is a member of the Asclepiadaceae, the milkweed family. They are so-named because, when injured, a milky latex is produced. This protects the plant and heals the wound. The latex of calotrope is very bitter to taste and is toxic, causing deaths of livestock in some countries. Anecdotal reports in Australia suggest that livestock stressed by drought may die from eating the plant.

Within Australia there is some confusion regarding the spread and distribution of the plant owing to the use of several different names. These include rubber bush, rubber tree, kapok tree, king's crown, cabbage tree and giant milkweed. The common name changes from one district to the next. To add to the confusion, some of these names are used for completely different weeds in different regions. Additionally there is a second species, C. gigantea , which has white flowers and is larger than C. procera but is otherwise similar in appearance. Within Australia, C. procera is able to produce abundant viable seeds while C. gigantea produces flowers but no viable seeds develop. Its reproduction in Australia is limited to vegetative propagation and hence relies on human intervention. It is thus not considered a weed.

The Asclepiadaceae have very complex pollination mechanisms and self pollination is often genetically prevented. It is possible that C. gigantea was introduced as a living plant or cutting and that there is insufficient genetic difference between individual plants to allow successful cross-pollination. Alternatively, complex relationships with pollinating insects may provide the explanation. The genus Calotropis is largely pollinated by specific members of the carpenter bee subfamily, Xylocopinae. Although present in Australia, they are not well represented, and it may be that species capable of pollinating C. gigantea are absent.

Calotrope prevails in the hot tropics and is spreading rapidly in northern Australia. It survives the driest seasons, establishing on land from which vegetation has been removed by mining, flooding, drought, overgrazing or construction work. It matures quickly into a perennial small tree and once established is difficult to displace. This is due to its ability to colonize exposed areas, put down a deep tap root and reach reproductive maturity rapidly.

The drought now gripping large areas of Australia greatly favours the spread of calotrope. The deep tap root allows the plant to survive and flourish in dry periods. Once established, the mature plant flowers nearly year round and so there are seeds available whenever germination rain falls. This allows the plant to colonize denuded areas. It out-competes useful native pasture plants, rapidly covering areas which would otherwise support grazing. Where thickets occur, mustering cattle becomes difficult and access to watering points reduced.

Calotrope is on the increase in many parts of northern Queensland, as well as the Northern Territory, South Australia and Western Australia. There are no recent audits, but it seems the weed infests tens of thousands of hectares. If it continues to spread unabated, millions of hectares are potentially threatened. There are anecdotal accounts of infestations increasing in area by 250% over a 5-year period. If not controlled, vast areas of Australia will become covered in this weed.

While biological control frequently supplies the best long-term solution, interim measures are needed urgently. Scientists at the Department of Natural Resources and Mines (NR&M) Tropical Weed Research Center, Charters Towers, are conducting trials near Georgetown in northern Queensland to assess the effectiveness of several herbicides using varying rates and different application methods. In conjunction with the Australian Agricultural Company, they are planning aerial application trials in the Gregory River area later in the year.

Herbicides will play only one role in the containment and ultimate control of this weed. Pasture management practices in the form of effective control of grazing pressure and mechanical control will also be essential tools to reduce the risk of further calotrope spread. Evidence from the Northern Territory suggests that a reduced stocking rate and a well maintained competitive pasture is able to reduce the spread and invasiveness of calotrope and in some situations out-compete established calotrope.

NR&M Weed Scientist Peter Wilkinson has recently found a disease on calotrope and, in conjunction with Department of Primary Industries, identified it as a fungal pathogen, with the proposed name, Phaeoramularia calotropidis, which has not previously been reported in Australia. The effectiveness of this disease is not known at present. If the disease does not harm native or useful plants, it may be useful as a mycoherbicide. However, development of reliable biocontrol methods takes time and more conventional control measures should be employed until then.

In the mean time landowners should avoid overgrazing and should mechanically uproot the plant or apply registered herbicides for the control of their calotrope problem.

By: Peter Wilkinson, Tropical Weeds Research Center, Department of Natural Resources and Mines, PO Box 187,
Charters Towers, 4820, Australia
Email: Fax: +61 7 4787 3969

❑

Revisiting History with Water Hyacinth

Biocontrol of water hyacinth (Eichhornia crassipes) has a long history. It was first attempted in 1971 with the release of the mite Orthogalumna terebrantis in Zambia, and other agents were released there and in Zimbabwe shortly after. In the next three decades, biocontrol introductions of up to six arthropod agents were made in 34 countries around the world. There is therefore a lot of experience to draw on, but as knowledge has increased, earlier work can also be reconsidered with the benefit of hindsight and advancing technology. Two recent initiatives have done just that. The first is a new look at a previously discarded prospective biocontrol agent, and the second a retrospective look at the water hyacinth infestations on Lake Victoria.

Old Agent Reassessed for New Region

The grasshopper Cornops aquaticum was discovered in some of the earliest surveys for water hyacinth natural enemies in South America. In 1974 it was described as one of the most damaging insects associated with the plant in its region of origin. However, it was rejected as a biocontrol agent for the USA, when subsequent investigations indicated that under laboratory starvation trials it could feed and complete development on species other than water hyacinth. More recently, as the suite of agents released in South Africa has still left gaps in control, researchers have been looking at this promising insect again.

Over the last 5 years, researchers at the Plant Protection Research Institute, South Africa have been testing the some 150 South African species in the water hyacinth family Pontederiaceae and related families. These no-choice trials with immature and adult Cornops are considered the most cautious of host specificity tests.

The results of the tests indicated that under laboratory conditions Cornops could complete development on five other species besides water hyacinth. Two of these, Pontederia cordata (pickerel weed) and Canna indica (canna), are introduced and potentially invasive in South Africa and thus of no conservation concern. (A taste for pickerel weed, however, precludes Cornops' use in the USA because the plant is indigenous there.)

Amongst the indigenous African Pontederiaceae, some specimens of Cornops were able to complete development on Heteranthera callifolia and plants were quite heavily attacked, but adult females were unable to oviposit on the thin petioles so the grasshopper is not expected to put this species at risk outside the laboratory. Likewise, Cornops is not expected to threaten two other African Pontederiaceae that experienced some damage in the laboratory. One incidence of oviposition only was recorded on Monochoria africana but complete development was never recorded. Equally, although some individuals completed development on Eichhornia natans, neither feeding nor oviposition was recorded, and, as it is a submerged plant, lack of emergent leaf material and submerged petioles are expected to preclude establishment by Cornops on this species in the wild.

A series of adult choice trials are being finalized. Once these results have been analysed a release report will be submitted to the National Department of Agriculture and the Department of Environmental Affairs and Tourism for comment. However, it could still be a year before this very damaging agent for water hyacinth sees the outside of a quarantine laboratory.

Remote Sensing for Historical Perspective

Just how much water hyacinth there was on Lake Victoria at the height of the infestations in the late 1990s and what happened to it has been hotly debated in various fora (including this journal). This is not merely a matter of historical interest, for questions of whether water hyacinth populations are on the rebound still surface regularly, and stakeholders are therefore justifiably concerned about the sustainability of the control exerted by the introduced Neochetina weevils. Part of the trouble has been a lack of reliable lake-wide year-on-year data on the water hyacinth infestations.

A report* available on the US Geological Survey EROS Data Center (EDC) website fills some of the gaps:

http://edcintl.cr.usgs.gov/lakespecialfeature.html

The report is a result of a study by EDC and Clean Lakes, Inc. funded by a USAID/Uganda, Greater Horn of Africa Initiative (GHAI) cooperative agreement with Clean Lakes, Inc. Scientists from EDC processed and analysed remotely sensed imagery in order to develop a geo-referenced database to quantify the distribution and extent of water hyacinth in Tanzanian, Ugandan and Kenyan Lake Victoria, and in some lakes in the Rwandan-Tanzanian Kagera River basin over a 5-year period from the early stages of the infestation. The study also analyses this information in relation to potentially influential factors such as weather, water level fluctuations and control measures. In particular, it looks at the relationship between dates of weevil introductions (which varied around the lake), El Niño events, and water hyacinth decline. An article based on the report appeared in Water Hyacinth News No. 6, see:

www.impecca.net/

The study found that:

• Water hyacinth infestations increased rapidly from 1997 and reached a maximum in 1998 with infestations particularly severe in the northern part of the lake.
• A sharp decline in water hyacinth occurred in 1999.
• A low level of water hyacinth has been maintained, although there was a small lake-wide increase in 2001.

The report notes that Neochetina weevils were first released in Uganda in late 1995, in Kenya in January 1997, and in Tanzania in August 1997. It points out that the noticeable reductions around the lake that began sometime after early 1999 coincided with rapidly increasing Neochetina weevil populations, but also followed the El Niño rains of late 1997/early 1998. It describes how records show that the lake rose by 1.8 m, and argues that severe weather created high wind and wave action that may have been crucial to the break up of plants already waterlogged and damaged through weevil activity.

The low post-weevil level of water hyacinth is just what would be expected in a classical biological control success, but the report points out some anomalies in dates of releases and observed reductions in water hyacinth infestations. It concludes that changes in water hyacinth infestations in Ugandan and Kenyan waters were consistent with weevil action (although the role of weather in hastening the disintegration of the mats should not be discounted). However, it points out that water hyacinth decline along the southern shores in Tanzania followed too rapidly after weevil introduction to be attributed to them, and other factors, particularly the weather at this time, are likely to have been critical.

The report also deals with Lake Mihindi in the Kagera River system in Rwanda, which was more than half-covered with water hyacinth by early 1997. Weevils were not released here until September 2000. Dramatic reductions in this infestation, coincident with declines observed in Lake Victoria, are suggested to have been the result of flood waters associated with heavy rainfall breaching an outlet; the water hyacinth may simply have been washed away. However, unlike Lake Victoria, by September 2000 the water hyacinth infestations had returned. While it may not be entirely valid to make direct comparisons between water bodies of such differing sizes, Lake Victoria had weevils and continuing reduced water hyacinth populations (including along its southern shore where weather was also concluded to be the likely cause of the initial decline) whereas Lake Mihindi had no weevils (at this time) and a resurgence of the weed occurred.

The report contains clear explanations of remote sensing tools and methods, and invaluable discussions on their use as a monitoring tool. It provides fascinating insights into the development of water hyacinth infestations on Lake Victoria, and in doing so raises new questions that will doubtless fuel renewed debate. Importantly, it includes recent data showing that low levels of water hyacinth suitable for growth remain in most parts of the lake, and concludes that continued active and aggressive management will reduce the likelihood of a major resurgence. Whilst the weevils are still very much in evidence, additional measures may be necessary to contain populations in the long term and a sustainable management plan is needed. This study contributes substantially to the baseline knowledge needed to develop a strategy by providing data on the development of the water hyacinth infestation in different parts of the lake.

*Albright, T.; Moorhouse, T.; McNabb,
T. (2002) The abundance and distribution of water hyacinth in lake Victoria and the Kagera River basin, 1989-2001. Sioux Falls, SD/Martinez, CA, USA; USGS/EROS Data Center/Clean Lakes, Inc., 42 pp.

Contact: [Cornops] Martin Hill,
Department of Zoology and Entomology, Rhodes University, P.O. Box 94,
Grahamstown, 6140, South Africa
Email:

[Remote sensing] Tom Albright,
SAIC - USGS/EROS Data Center,
Sioux Falls, South Dakota, USA
Email:

[Remote sensing] Tom Moorhouse,
Clean Lakes, Inc., Martinez,
California, USA
Email:
Website: www.cleanlake.com

❑

Biocontrol Agent Out-Competes Wine Disease

In recent years, New Zealand's wine has gained a fine reputation and some 1% of the world market. As in many other wine-growing regions, however, disease is a constant threat. 'Bunch rot' or 'grey mould' of grapes, caused by the fungus Botrytis cinerea, is arguably the most important disease problem confronting the New Zealand wine industry and is a significant problem for the wine industry worldwide. It has been estimated to seasonally cost the New Zealand wine and grape growing industry up to NZ$18 million in lost grape sales and in addition up to $12 million annually in disease control costs.

Botrytis bunch rot (the correct term for the disease) can occur at any time during the growing season although it's most destructive when warm temperatures follow rainfall either at flowering or later in the growing season. The fungus can develop on dead and dying flower parts, young shoots and newly emerged leaves and as a moist rot on berries in the bunch. Often following rainfall, the fungus is visible as a grey felt-like mat of spores on the infected tissues. The disease spreads easily from berry to berry in the bunch later in the season and from these infected berries masses of airborne spores are produced which are liberated into the vineyard to infect other berries. While limited/managed infection still allows for high quality 'botrytised' wine production, in most cases the fungus continues to rot grapes on the vine reducing yield and causing winemaking problems. In addition, mechanical harvesting means that even low numbers of infected bunches can not be separated during harvesting operations, thereby contaminating the entire crop. Not only does Botrytis bunch rot affect grape-growing but serious damage is also incurred worldwide on berry fruit, kiwifruit, and tomato crops.

Conventional control of B. cinerea is based on repeated sprays of chemical 'botryicides', but efficacy may be limited by the development of resistant Botrytis strains and regulatory restrictions on the timing and number of applications. In addition, there is growing consumer pressure to reduce the number of chemical pesticides used in the vineyard. These impediments opened the way for researching a biological approach to control the disease. HortResearch, with support from the Winegrowers of New Zealand (now 'New Zealand WineGrowers') and Technology New Zealand looked for a suitable antagonist: an organism that could out-compete and suppress botrytis without having any detrimental effects on the vine itself. The scientists isolated a naturally occurring saprophytic fungus from dead leaf litter which proved to be an aggressive and successful competitor of B. cinerea. HortResearch then applied for a patent relating to the use of this saprophytic fungus as a biological control agent.

In April 2001, Botry-Zen Ltd was set up to commercialize the registered product BOTRY-Zen, marketing it as an alternative Botrytis control product for grape growers who wanted to be able to describe their crops as `sustainably grown'. In addition, Bio-Gro New Zealand (which approves products for use by organic growers) has accredited BOTRY-Zen for use in organic wine production. Botry-Zen Ltd has now gained full product Registration in New Zealand and has recently initiated field trialling in the European Union (EU) and in California (USA). The existing New Zealand factory is being expanded considerably and off-shore manufacturing licensing is being closely evaluated so that the product will be available in commercial quantities once Registration is achieved. It is envisaged that final Registration in the EU and the USA will be concluded early in 2005.

The BOTRY-Zen product contains a live preparation of the spores of the non-invasive, non-pathogenic saprophytic fungus isolated by HortResearch. It acts as a true antagonist, aggressively occupying the same physical space and out-competing Botrytis for the nutrients in dead and senescent material in the vines, especially flower tissues such as stamens, flowers caps and aborted fruitlets. A significant added bonus is that this mechanism of action makes it unlikely that resistance to the product will develop.

Trials in New Zealand vineyards since 1997 have demonstrated that BOTRY-Zen, applied twice over the flowering period and twice over bunch closure, provides protection against Botrytis infection at comparable levels to, or better than, standard fungicide programmes.

BOTRY-Zen does not control other grape diseases such as powdery or downy mildew but research commissioned by Botry-Zen Ltd has shown that some commonly used fungicides are compatible with BOTRY-Zen, with some being able to be tank mixed. This is an exciting new development with both botrytis and powdery mildew almost completely controlled this last season with tank mixes. Research is continuing at HortResearch to assess compatibility with other fungicides used for these diseases.

The discovery and underpinning research was undertaken and funded by HortResearch. Subsequent developmental research was jointly funded by HortResearch and by the Wine Institute of New Zealand Incorporated and the New Zealand Grape Growers Council whose commercial arms, Winegrowers of New Zealand Ltd and New Zealand Grape Growers Ltd formed a joint venture called Winegrape Tech. Winegrape Tech has the right to commercially exploit the HortResearch patent rights and other intellectual property rights associated with the saprophytic fungus organisms themselves and has granted to Botry-Zen Ltd an exclusive licence of those rights in all countries that are parties to the Patent Co-operation Treaty. The research was also funded by the Foundation for Research Science and Technology and through a Technology Business Growth grant.

Source: Botry-Zen Ltd website:

www.botryzen.co.nz/INDEX.HTML

Contact: John Scandrett, CEO,
BotryZen Ltd, PO Box 1777,
Dunedin, New Zealand
Email: Fax: +64 3 4772953

Philip Elmer (principle research scientist), HortResearch Ruakura Research Centre, Hamilton, New Zealand
Email:

Peter Wood (viticulture scientist), HortResearch Hawkes Bay Research Centre,
New Zealand
Email:

❑

Brazil Quarantine Laboratory Celebrates 12 Years

The Brazilian National Quarantine Laboratory `Costa Lima' Embrapa Environment has for the last 12 years fostered international exchange and undertaken quarantine of beneficial organisms for biological control. It is the only institution in Brazil authorized, since 1991, by the Ministry of Agriculture and Supply (MA) to introduce natural enemies for pest control and other beneficial organisms for scientific research. It is also one of the mandates of the laboratory to interact with foreign institutions for export of biocontrol agents.

The main services provided by the Quarantine Laboratory are:

• To prepare documents needed by interested professionals to request permits for importation (which are issued by the Ministry of Agriculture).
• To assist interested professionals in locating suitable sources of biocontrol agents for introduction.
• To collect and evaluate information concerning import and export of natural enemies.
• To provide information to the Ministry of Agriculture concerning safety of proposed introductions, in consonance with a supporting Technical Committee of Brazilian researchers expert in biocontrol.
• To provide quarantine services.
• To monitor releases and evaluate efficacy.
• To collaborate with international research institutions in the exploration and exchange of native natural enemies under a previous cooperative agreement.

During the 12 years of activities, the laboratory has processed 181 introductions of biocontrol agents and other microorganisms involving 17 species of insect parasitoids, two of predators, nine of mites, seven of nematodes and 146 of different microorganisms. Also several international biological control projects have received cooperation from the quarantine laboratory, including USDA-ARS (US Department of Agriculture - Agricultural Research Service), University of Florida (USA), IITA (International Institute of Tropical Agriculture), IRD (Institut de Recherche pour le Développement, France), CIAT (Centro Internacional de Agricultura Tropical) and Amsterdam University (the Netherlands) amongst others.

Other activities carried out by the laboratory are:

• To manage the International Information System on Biological Control via the Internet:
  www.bdt.org.br/bdt/biocontrol
  together with a discussion list:
  
• To maintain a database about biological control agents in Brazil.
• To alert the public about bringing organisms into Brazil that are not officially approved, providing pamphlets, information brochures, meetings, videos and lectures.

The Brazilian Quarantine facility intends to be an overseas laboratory for foreign research institutions around the world in order to promote biological control programmes, and biocontrol workers are welcome to contact us to discuss possible initiatives.

By: J. Tambasco; F. Lucchini; L. A. N. de Sá; E. A. B. De Nardo and M. A. B. Morandi

Lab. de Quarentena `Costa Lima', Embrapa Meio Ambiente, Cx. P. 69, 13820-000, Jaguariúna, SP, Brazil
Email:

❑

Indian Bio-Control Research Laboratories Progress

Bio-Control Research Laboratories (BCRL), established at Bangalore in 1981, is the R&D division of Pest Control (India) Private Ltd (PCI). BCRL is engaged in research, production and field trials with biological control agents and pheromones. Since February 2001, when it moved into a newly constructed laboratory complex, staff have been busy with, amongst other activities, scaling up and increasing the range of biocontrol products. The new facilities include a laboratory complex for production and quality testing of biological control agents, walk-in incubators and cold storage rooms, along with glasshouses and experimental fields. The construction of the state-of-the-art laboratory complex was a major landmark in the history of biological control of crop pests in the country. BCRL, which is a Department of Scientific and Industrial Research (Government of India) recognised laboratory, was also the recipient of its National Award for Excellence in R&D Efforts in Industry in the Agro Industry Category for the year 1993.

BCRL has made several innovations and standardized mass production techniques for a variety of natural enemies and their hosts. These include:

• Development of a breeding-cum-shipment container for the predatory mite Amblyseius tetranychivorus
• Design and development of a unique, easy-to-use insect water trap for mass trapping of sugarcane borers and pests of other field crops, in association with sex pheromone lures
• Establishment of a mass production facility for continuous year-round laboratory production of Helicoverpa armigera and Spodoptera litura
• Doubling productivity of eggs of the alternate laboratory host Corcyra cephalonica
• Increasing temperature tolerance and searching capacity of Trichogramma spp.
• Devising blister packing for shipment of Cryptolaemus montrouzieri grubs to avoid cannibalism during transportation
• Development of technology for mass production and formulation of entomopathogenic fungi

In addition, under an innovative industry-institution partnership project with Tamil Nadu Agricultural University, Coimbatore it has successfully developed methods for improving commercial scale production, formulation and field efficacy of nuclear polyhedrosis viral (NPV) pesticides under the National Agricultural Technology Project, Indian Council for Agricultural Research (ICAR).

The efforts made by BCRL have played a crucial role in popularizing and transferring biocontrol technology from research laboratories to farmers' fields in India. An exclusive Field Extension and Education Unit was established in 2002 to create awareness about biological control technology at the grass root level. Qualified supervisory staff and trained field assistants are being recruited and posted in different parts of the country for carrying out field demonstrations to validate the performance of biocontrol products in farmers' fields. The service network of 85 establishments of the parent company is increasingly being used to provide biocontrol inputs directly to the farmers all over India.

BCRL has been an active participant in the All India Coordinated Research Project on Biological Control since its inception. In 2003 it signed a Memorandum of Understanding with the University of Agricultural Sciences, Bangalore for collaborative teaching, research and extension. In addition, it has been operating collaborative projects with international research organizations such as Natural Resources Institute, UK.

Contact: K. P. Jayanth, General Manager, Bio-Control Research Laboratories,
A division of Pest Control (India) Pvt. Ltd., PO Box 6426, Yelahanka,
Bangalore 560 064, India
Email: / Fax: +91 80 8468838
Website: www.pcilindia.com

❑

Guidance for Inundative Biocontrol

The European Union-funded ERBIC (Evaluating Environmental Risks of Biological Control Introductions into Europe) project has published a paper proposing methodology for assessing the risk of biocontrol introductions*. However, BioControl' s editor-in-chief and project team member Heikki Hokkanen notes in the Editorial to the issue, the topic is a controversial one, so the paper should be viewed as opening scientific discussion on how to demonstrate the safety or otherwise of candidate biocontrol agents.

In the last issue of BNI, a news article on proposals for the revision of the Code of Conduct for the Import and Release of Exotic Biological Control Agents (ISPM No. 3) reminded us that ISPM No. 3 does not set out in detail how to assess whether a proposed introduction is safe. No risk analysis process is explained, although the requirement for collection of information in dossiers implied that risk analysis would be the basis of decisions. The article highlighted the need for detailed guidance, and thus the ERBIC-authored paper is very timely.

The paper points out that although biocontrol has a good track record in terms of safety, the current popularity of commercial inundative biocontrol may result in problems as an increasing number of activities are undertaken without appropriate training or understanding of risk. It is this area of biocontrol that the ERBIC project has focused on (although some of the principles and approaches will apply to classical biocontrol). The authors recognize that the challenge is to achieve a balance in developing protocols and guidelines to prevent serious mistakes occurring in importing and releasing harmful exotic species while still allowing safe biocontrol to proceed. They identify the most critical ecological issues as being able to estimate (a) the probabilities of attack on non-target organisms and (b) the dispersal and establishment capacities of the biocontrol agent. The methodology described in this paper integrates information on the potential for an agent to establish in a non-target habitat, its potential dispersal distance, its host range (number of species), and its potential direct and indirect effects on non-target organisms.

One of the most difficult parameters to assess is the potential for adverse effects on non-target organisms (and the ecosystems in which they function), especially for generalist natural enemies that impact on many species. Host range forms a central element of the described risk evaluation process, because a lack of host specificity could lead to unacceptable risk if such an agent were to establish and disperse widely whereas, in contrast, a strictly monophagous agent would not be expected to pose serious risk however well established or widely dispersed it became. Although few natural enemies are truly monophagous, many have a restricted host/prey range. For classical biocontrol, the narrower the host range, in general, the better. However, the biocontrol industry sometimes favours the more polyphagous natural enemies precisely because they will control a range of taxonomically unrelated pests. This tendency is particularly likely to give rise to non-target effects.

In an illustrative case history exercise, the proposed methodology is applied to a number of natural enemy species currently mass reared and released for biocontrol of pests in glasshouse and open-field conditions in Europe, drawing on published information and expert opinion. Agents are assigned a numerical value (1-5) for the likelihood and magnitude of each criterion considered, and these values are used to calculate a risk index for each agent**. The results obtained in this exercise confirmed a number of expectations in relation to previous experience. The authors also identified both under- and over-estimates of risk by this method (and under-estimates particularly where there is limited knowledge of the release area ecosystem). In addition, attention is drawn to variations between risk index values for the same species in different release areas. Based on their collective experience the authors give index ranges for three risk categories: low (usually meaning no objection to release), intermediate (where further specified information would be requested before a decision was made) and high (indicating that release would normally be advised against).

The potential for the use of such an index is clear. For example, agents used against a single pest could be ranked in terms of risk for a given release area. In addition, while the current study does not allow conclusions about risks for specific groups of natural enemies to be drawn, analysis of more cases might allow such generalizations to be made. The authors caution, however, against narrow interpretation of the index, noting that different criteria may be more or less important in different circumstances, and that high risk values are not by definition negative. They conclude that while meaningful ranking of natural enemies in risk categories now seems possible, such an index should be used by biological control experts as an element in a wider assessment of the safety of a potential biocontrol agent.

The authors recognize the biocontrol industry's fear that the common trend towards more stringent regulatory requirements by countries will lead to lengthy procedures and thus high costs, which in turn could threatens the sector's financial viability. However, they argue that the risk assessment procedure they propose is a first step towards light and harmonized registration that will not be prohibitive for the industry.

*Van Lenteren, J.C.; Babendreier, D.; Bigler, F.; Burgio, G.; Hokkanen, H.M.T.; Kuske, S.; Loomans, A.J.M.; Menzler-Hokkanene, I.; Van Rijn, P.C.J.; Thomas, M.B.; Tommasini, M.G.; Zeng, Q.Q. (2003) Environmental risk assessment of exotic natural enemies used in inundative biological control. BioControl 48, 3-38.

**See the ERBIC website for further information:

http://honeybee.helsinki.fi/MMSBL/MAEL/Hankkeet/ERBIC/index.htm

❑

UK Non-Natives Review

The UK Government has recently published its non-native species policy review. A series of working groups were set up to consider current arrangements for dealing with the introduction, establishment and spread of non-native species, and assess the main pathways through which non-native species are introduced and spread. The report makes recommendations to improve measures to limit the ecological and economic impact of invasive non-native species in Great Britain. In general the review follows the recommendations of the Convention on Biological Diversity (CBD) guiding principles on invasive alien species and the three-stage hierarchical approach.

As readers will be aware classical biological control is considered novel in the UK as well as in Europe so it is an important fact that the review recommends that: "strategic funding should be made available to support the development of novel control techniques for invasive non-native species". Furthermore, in the draft of their first Science and Innovation Strategy, the Department for Environment, Food and Rural Affairs (DEFRA) have followed up the review team's recommendation and committed to launching a new research programme on alien and invasive species including further work on biological control of Japanese knotweed (Fallopia japonica) which was one of the case study species in the original review.

This represents an important endorsement of biological control by the UK government and should lead to more use in the UK of what has become the mainstay of sustainable control for arthropod and plant pests in many countries around the world.

Review of Non-native Species Policy:

www.defra.gov.uk/wildlife-countryside/resprog/findings/non-native/index.htm

Science & Innovation Strategy: Delivering the Evidence:

www.defra.gov.uk/Science/S_IS/

❑