The suitability of non-target native mangroves for the survival and development of the lantana bug Aconophora compressa, an introduced weed biological control agent

Aconophora compressa (Hemiptera: Membracidae), a biological control agent introduced against the weed Lantana camara (Verbenaceae) in Australia, has since been observed on several non-target plant species, including native mangrove Avicennia marina (Acanthaceae). In this study we evaluated the suitability of two native mangroves, A. marina and Aegiceras corniculatum (Myrsinaceae), for the survival and development of A. compressa through no-choice field cage studies. The longevity of females was significantly higher on L. camara (57.7 ± 3.8 days) than on A. marina (43.3 ± 3.3 days) and A. corniculatum (45.7 ± 3.8 days). The proportion of females laying eggs was highest on L. camara (72%) followed by A. marina (36%) and A. corniculatum (17%). More egg batches per female were laid on L. camara than on A. marina and A. corniculatum. Though more nymphs per shoot emerged on L. camara (29.9 ± 2.8) than on A. marina (13 ± 4.8) and A. corniculatum (10 ± 5.3), the number of nymphs that developed through to adults was not significantly different. The duration of nymphal development was longer on A. marina (67 ± 5.8 days) than on L. camara (48 ± 4 days) and A. corniculatum (43 ± 4.6 days). The results, which are in contrast to those from previous glasshouse and quarantine trials, provide evidence that A. compressa adults can survive, lay eggs and complete nymphal development on the two non-target native mangroves in the field under no-choice condition.

Agonosoma trilineatum (Heteroptera: Scutelleridae) a biological control agent of the weed bellyache bush, Jatropha gossypiifolia (Euphorbiaceae)

Bellyache bush, Jatropha gossypiifolia L., is a serious weed of northern Australia. Agonosoma trilineatum (F.) is an insect from tropical America released in Australia in 2003 as a biological control agent against bellyache bush. It feeds on seeds and has the potential to reduce seed production, thereby potentially reducing the rate of spread and recruitment. To test the host specificity of A. trilineatum, four biological responses to host plant species were determined: development of nymphs, oviposition preferences, adult feeding and frequency of mating. Development of nymphs to adults and adult feeding only occurred on three Jatropha spp. These species also supported mating and oogenesis but only J. gossypiifolia was accepted for oviposition. Mating did not occur in the presence of other plant species. The evidence indicates that there is little risk associated with the release of this insect species in Australia and probably other countries where this weed is a problem. The probability of this insect expanding its host range is low because multiple aspects of the biology would need to change simultaneously. A. trilineatum was released in Australia between 2003 and 2007. A Climex model indicated that coastal areas of Queensland and the Northern Territory would be climatically most suitable for this insect.

Evaluation of Ustilago sporoboli-indici as a classical biological control agent for invasive Sporobolus grasses in Australia

Sporobolus pyramidalis, S. africanus, S. natalensis, S. fertilis and S. jacquemontii, known collectively as the weedy Sporobolus grasses, are exotic weeds causing serious economic losses in grazing areas along Australia's entire eastern coast. In one of the first attempts to provide biological control for a grass, the potential of a smut, Ustilago sporoboli-indici, as a biological control agent for all five weedy Sporobolus spp. found in Australia was evaluated in glasshouse studies. Application of basidiospores to 21-day-old Sporobolus seedlings and subsequent incubation in a moist chamber (26 °C, 90% RH, 48 h) resulted in infection of S. pyramidalis, S. africanus, S. natalensis and S. fertilis but not S. jacquemontii. Host-range trials with 13 native Australian Sporobolus spp. resulted in infection of four native species. Evaluation of damage caused by the smut on two Australian native and two weedy Sporobolus spp. showed that the total numbers of flowers infected for the four grasses were in the following order: S. creber > S. fertilis > S. elongatus > S. natalensis with percentage flower infections of 21%, 14%, 12% and 3%, respectively. Significant differences (P = 0.001) were found when the numbers of infected flowers caused by each treatment were compared. The infection of the four native Sporobolus spp. by the smut indicated that it was not sufficiently host specific for release in Australia and the organism was rejected as a potential biological control agent. The implications of these results are discussed.

Pre-release evaluation of the efficacy of the leaf-sucking bug Carvalhotingis visenda (Heteroptera: Tingidae) as a biological control agent for cat's claw creeper Macfadyena unguis-cati (Bignoniaceae)

In classical weed biological control, assessing weed response to simulated herbivory is one option to assist in the prioritization of available agents and prediction of their potential efficacy. Previously reported simulated herbivory studies suggested that a specialist herbivore in the leaf-feeding guild is desirable as an effective biological control agent for cat's claw creeper Macfadyena unguis-cati (Bignoniaceae), an environmental weed that is currently a target for biological control. In this study, we tested (i) whether the results from glasshouse-based simulated herbivory can be used to prioritise potential biological control agents by evaluating the impact of a leaf-sucking tingid bug Carvalhotingis visenda (Drake & Hambleton) (Hemiptera: Tingidae) in quarantine; and (ii) the likely effectiveness of low- and high-densities of the leaf-sucking tingid after its release in the field. The results suggest that a single generation of C. visenda has the potential to reduce leaf chlorophyll content significantly, resulting in reduced plant height and leaf biomass. However, the impact of one generation of tingid herbivory on below-ground plant components, including the roots and tuber size and biomass, were not significant. These findings are consistent with results obtained from a simulated herbivory trial, highlighting the potential role of simulated herbivory studies in agent prioritisation.

The leaf-tying moth Hypocosmia pyrochroma (Lep., Pyralidae), a host-specific biological control agent for cat's claw creeper Macfadyena unguis-cati (Bignoniaceae) in Australia

Cat's claw creeper, Macfadyena unguis-cati, a major environmental weed in coastal and sub-coastal areas of Queensland and New South Wales, Australia is a target for classical biological control. Host specificity of Hypocosmia pyrochroma Jones (Lep., Pyralidae), as a potential biological control agent was evaluated on the basis of no-choice and choice larval feeding and survival, and adult oviposition preference tests, involving 38 plant species in 10 families. In no-choice tests, larval feeding and development occurred only on cat's claw creeper. In choice tests, oviposition and larval development was evident only on cat's claw creeper. The results support the host-specificity tests conducted in South Africa, and suggest that H. pyrochroma is a highly specific biological control agent that does not pose any risk to non-target plants tested in Australia. This agent has been approved for field release by relevant regulatory authorities in Australia.

Improving the agent selection, release and evaluation process: the role of bioclimatic modelling

The value of CLIMEX models to inform biocontrol programs was assessed, including predicting the potential distribution of biocontrol agents and their subsequent population dynamics, using bioclimatic models for the weed Parkinsonia aculeata, two Lantana camara biocontrol agents, and five Mimosa pigra biocontrol agents. The results showed the contribution of data types to CLIMEX models and the capacity of these models to inform and improve the selection, release and post release evaluation of biocontrol agents. Foremost among these was the quality of spatial and temporal information as well as the extent to which overseas range data samples the species’ climatic envelope. Post hoc evaluation and refinement of these models requires improved long-term monitoring of introduced agents and their dynamics at well selected study sites. The authors described the findings of these case studies, highlighted their implications, and considered how to incorporate models effectively into biocontrol programs.

Biology and host range of Ophiomyia camarae Spencer (Diptera: Agromyzidae), a potential biocontrol agent for Lantana spp. (Verbenaceae) in Australia

The life history and host range of the herringbone leaf-mining fly Ophiomyia camarae, a potential biological control agent for Lantana spp., were investigated. Eggs were deposited singly on the underside of leaves. Although several eggs can be laid on a single leaf and a maximum of three individual mines were seen on a single leaf, only one pupa per leaf ever developed. The generation time (egg to adult) was about 38 days. Females (mean 14 days) lived longer than males (mean 9 days) and produced about 61 mines. Oviposition and larval development occurred on all five lantana phenotypes tested. Eleven plant species representing six families were tested to determine the host range. Oviposition and larval development occurred on only lantana and another nonnative plant Lippia alba (Verbenaceae), with both species supporting populations over several generations. A CLIMEX model showed that most of the coastal areas of eastern Australia south to 30°16' S (Coffs Harbour) would be suitable for O. camarae. O. camarae was approved for release in Australia in October 2007 and mines have been observed on plants at numerous field sites along the coast following releases.

Specificity of Carvalhotingis visenda (Hemiptera: Tingidae) as a biological control agent for cat's claw creeper Macfadyena unguis-cati (Bignoniaceae) in Australia

Biological control is considered the most suitable management option for cat's claw creeper, Macfadyena unguis-cati, a major environmental weed in coastal and sub-coastal areas of Queensland and New South Wales, Australia. The potential host range of the leaf-sucking bug, Carvalhotingis visenda (Hemiptera: Tingidae) was evaluated on the basis of nymphal survival and development, adult feeding and survival, and oviposition preference using choice and no-choice tests involving 38 plant species in 10 families. In no-choice tests, although adults survived on a few of the non-target plants, no eggs were laid on any of the non-target plants. In no-choice condition, the tingid oviposits and completes nymphal development only on M. unguis-cati. There was also no visible feeding damage on any of the non-target plants. In choice tests, adults showed distinct preference for M. unguis-cati, and the preference level increased over time as the tingids moved away from the non-target plants. At the end of the trial no adults were evident on any of the non-target plants. Host specificity tests confirm that the tingid is a highly host specific biocontrol agent, and does not pose risk to any non-target plants in Australia. This agent has been approved for field release by the relevant regulatory authorities in Australia.

The host range and biology of Cometaster pyrula; a biocontrol agent for Acacia nilotica subsp. indica in Australia

Prickly acacia, Acacia nilotica subsp. indica (Benth.) Brenan, a major weed of the Mitchell Grass Downs of northern Queensland, Australia, has been the target of biological control projects since the 1980s. The leaf-feeding caterpillar Cometaster pyrula (Hopffer) was collected from Acacia nilotica subsp. kraussiana (Benth.) Brenan during surveys in South Africa to find suitable biological control agents, recognised as a potential agent, and shipped into a quarantine facility in Australia. Cometaster pyrula has a life cycle of approximately 2 months during which time the larvae feed voraciously and reach 6 cm in length. Female moths oviposit a mean of 339 eggs. When presented with cut foliage of 77 plant species, unfed neonates survived for 7 days on only Acacia nilotica subsp. indica and Acacia nilotica subsp. kraussiana. When unfed neonates were placed on potted plants of 14 plant species, all larvae except those on Acacia nilotica subsp. indica and Acacia nilotica subsp. kraussiana died within 10 days of placement. Cometaster pyrula was considered to be highly host specific and safe to release in Australia. Permission to release C. pyrula in Australia was obtained and the insect was first released in north Queensland in October 2004. The ecoclimatic model CLIMEX indicated that coastal Queensland was climatically suitable for this insect but that inland areas were only marginally suitable.

A systematic approach to biological control agent exploration and prioritisation for prickly acacia (Acacia nilotica ssp. indica)

Agent selection for prickly acacia has been largely dictated by logistics and host specificity. Given that detailed ecological information is available on this species in Australia, we propose that it is possible to select agents based on agent efficacy and desired impact on prickly acacia demography. We propose to use the 'plant genotype' and 'climatic' similarities as filters to identify areas for future agent exploration; and plant response to herbivory and field host range as 'predictive' filters for agent prioritisation. Adopting such a systematic method that incorporates knowledge from plant population ecology and plant-herbivore interactions makes agent selection decisions explicit and allow more rigorous evaluations of agent performance and better understanding of success and failure of agents in weed biological control.

Temporal patterns in incidence and abundance of Aconophora compressa (Hemiptera: Membracidae), a biological control agent for Lantana camara, on target and nontarget plants

The membracid Aconophora compressa Walker, a biological control agent released in 1995 to control Lantana camara (Verbenaceae) in Australia, has since been collected on several nontarget plant species. Our survey suggests that sustained populations of A. compressa are found only on the introduced nontarget ornamental Citharexylum spinosum (Verbenaceae) and the target weed L. camara. It is found on other nontarget plant species only when populations on C. spinosum and L. camara are high, suggesting that the presence of populations on nontarget species may be a spill-over effect. Some of the incidence and abundance on nontarget plants could have been anticipated from host specificity studies done on this agent before release, whereas others could not. This raises important issues about predicting risks posed by weed biological control agents and the need for long-term postintroduction monitoring on nontarget species.

Studies on the rust Prospodium tuberculatum, a new classical biological control agent released against the invasive alien weed Lantana camara in Australia. 2. Host range

A strain of the rust Prospodium tuberculatum from Brazil was screened as a potential biocontrol agent against 40 Australian Lantana camara forms and 52 closely related, non-target plant species. Results under glasshouse conditions showed that the Brazilian rust strain is pathogenic to only two flower colour forms: pink and pink-edged red. Macro- and microsymptoms were recorded using 11 assessment categories and four susceptibility ratings. No macrosymptoms were observed on any of the non-target plants.

A scientific approach to agent selection

The prioritisation of potential agents on the basis of likely efficacy is an important step in biological control because it can increase the probability of a successful biocontrol program, and reduce risks and costs. In this introductory paper we define success in biological control, review how agent selection has been approached historically, and outline the approach to agent selection that underpins the structure of this special issue on agent selection. Developing criteria by which to judge the success of a biocontrol agent (or program) provides the basis for agent selection decisions. Criteria will depend on the weed, on the ecological and management context in which that weed occurs, and on the negative impacts that biocontrol is seeking to redress. Predicting which potential agents are most likely to be successful poses enormous scientific challenges. 'Rules of thumb', 'scoring systems' and various conceptual and quantitative modelling approaches have been proposed to aid agent selection. However, most attempts have met with limited success due to the diversity and complexity of the systems in question. This special issue presents a series of papers that deconstruct the question of agent choice with the aim of progressively improving the success rate of biological control. Specifically they ask: (i) what potential agents are available and what should we know about them? (ii) what type, timing and degree of damage is required to achieve success? and (iii) which potential agent will reach the necessary density, at the right time, to exert the required damage in the target environment?

How widespread is Parthenium hysterophorus and its biological control agent Zygogramma bicolorata in South Asia?

Parthenium hysterophorus is a weed of global significance causing severe economic, environmental, human and animal health problems in Asia, Africa, Australia and the Pacific. In South Asia, P. hysterophorus occurs in India, Pakistan, Sri Lanka, Bangladesh and Nepal. A host-specific leaf-feeding beetle Zygogramma bicolorata from Mexico was introduced into India in 1984, as a biological control agent for P. hysterophorus. In this study, a GIS-based distribution map of P. hysterophorus and its biological control agent Z. bicolorata in South Asia based on meta-analysis is presented. The map highlights the limited published information on P. hysterophorus incidence in many of the states and territories in India, as well as in neighbouring Bangladesh, Bhutan, Nepal and Pakistan. Incidence of Z. bicolorata was recorded as three distinct clusters, covering many states in India. In Pakistan, Z. bicolorata was recorded in the Punjab region bordering India. A CLIMEX model based on the current distribution of Z. bicolorata in India suggests that the geographic range of this agent in India and Pakistan can extend to other P. hysterophorus-infested areas in the region. The CLIMEX model also suggests that all of Bangladesh and Sri Lanka, and parts of Nepal are climatically suitable for Z. bicolorata.