Relative host plant species use by the lantana biological control agent Aconophora compressa (Membracidae) across its native and introduced ranges

Aconophora compressa Walker (Hemiptera: Membracidae) was released in 1995 against the weed lantana in Australia, and is now found on multiple host plant species. The intensity and regularity at which A. compressa uses different host species was quantified in its introduced Australian range and also its native Mexican range. In Australia, host plants fell into three statistically defined categories, as indicated by the relative rates and intensities at which they were used in the field. Fiddlewood (Citharexylum spinosum L.: Verbenaceae) was used much more regularly and at higher densities than any other host sampled, and alone made up the first group. The second group, lantana (Lantana camara L.: Verbenaceae; pink variety) and geisha girl (Duranta erecta L.: Verbenaceae), were used less regularly and at much lower densities than fiddlewood. The third group, Sheena’s gold (another variety of D. erecta), jacaranda (Jacaranda mimosifolia D. Don: Bignoniaceae) and myoporum (Myoporum acuminatum R. Br.: Myoporaceae), were used infrequently and at even lower densities. In Mexico, the insect was found at relatively low densities on all hosts relative to those in Australia. Densities were highest on L. urticifolia, D. erecta and Tecoma stans (L.) Juss. ex Kunth (Bignoniaceae), which were used at similar rates to one another. It was found also on a few other verbenaceous and non-verbenaceous host species but at even lower densities. The relative rate at which Citharexylum spp. and L. urticifolia were used could not be assessed in Mexico because A. compressa was found on only one plant of each species in areas where these host species co-occurred. The low rate at which A. compressa occurred on fiddlewood in Mexico is likely to be an artefact of the short-term nature of the surveys or differences in the suites of Citharexylum and Lantana species available there. These results provide further incentive to insist on structured and quantified surveys of non-target host use in the native range of potential biological control agents prior to host testing studies in quarantine.     

Indirect nontarget effects of host-specific biological control agents: Implications for biological control

Classical biological control of weeds currently operates under the assumption that biological control agents are safe (i.e., low risk) if they do not directly attack nontarget species. However, recent studies indicate that even highly host-specific biological control agents can impact nontarget species through indirect effects. This finding has profound implications for biological control. To better understand the causes of these interactions and their implications, we evaluate recent case studies of indirect nontarget effects of biological control agents in the context of theoretical work in community ecology. We find that although particular indirect nontarget effects are extremely difficult to predict, all indirect nontarget effects of host specific biological control agents derive from the nature and strength of the interaction between the biological control agent and the pest. Additionally, recent theoretical work suggests that the degree of impact of a biological control agent on nontarget species is proportional to the agent’s abundance, which will be highest for moderately successful control agents. Therefore, the key to safeguarding against indirect nontarget effects of host-specific biological control agents is to ensure the biological control agents are not only host specific, but also efficacious. Biological control agents that greatly reduce their target species while remaining host-specific will reduce their own populations through density-dependent feedbacks that minimize risks to nontarget species.     

Integrated biological control of bacterial speck and spot of tomato under field conditions using foliar biological control agents and plant growth-promoting rhizobacteria

Integration of foliar bacterial biological control agents and plant growth promoting rhizobacteria (PGPR) was investigated to determine whether biological control of bacterial speck of tomato, caused by Pseudomonas syringae pv. tomato, and bacterial spot of tomato, caused by Xanthomonas campestris pv. vesicatoria and Xanthomonas vesicatoria, could be improved. Three foliar biological control agents and two selected PGPR strains were employed in pairwise combinations. The foliar biological control agents had previously demonstrated moderate control of bacterial speck or bacterial spot when applied as foliar sprays. The PGPR strains were selected in this study based on their capacity to induce resistance against bacterial speck when applied as seed and soil treatments in the greenhouse. Field trials were conducted in Alabama, Florida, and California for evaluation of the efficacy in control of bacterial speck and in Alabama and Florida for control of bacterial spot. The foliar biological control agent P. syringae strain Cit7 was the most effective of the three foliar biological control agents, providing significant suppression of bacterial speck in all field trials and bacterial spot in two out of three field trials. When applied as a seed treatment and soil drench, PGPR strain Pseudomonas fluorescens 89B-61 significantly reduced foliar severity of bacterial speck in the field trial in California and in three of six disease ratings in the field trials in Alabama. PGPR strains 89B-61 and Bacillus pumilus SE34 both provided significant suppression of bacterial spot in the two field trials conducted in Alabama. Combined use of foliar biological control agent Cit7 and PGPR strain 89B-61 provided significant control of bacterial speck and spot of tomato in each trial. In one field trial, control was enhanced significantly with combined biological control agents compared to single agent inoculations. These results suggest that some PGPR strains may induce plant resistance under field conditions, providing effective suppression of bacterial speck and spot of tomato, and that there may be some benefit to the integration of rhizosphere-applied PGPR and foliar-applied biological control agents.     

Role of plant phenology in mediating interactions between two biological control agents for spotted knapweed

The role of spotted knapweed phenology on the attack rate of two seed-head insects Urophora affinis and Larinus minutus was assessed in a series of field studies at four study sites in south-eastern British Columbia, Canada. Slow or later developing knapweed plants had more seed heads that contained only single or multiple U. affinis whereas early or faster developing plants had more seed heads containing L. minutus alone or in combination with U. affinis. L. minutus did not distinguish between seed heads with or without U. affinis larvae when laying eggs. However, seed heads with multiple U. affinis present, produced fewer L. minutus adults than expected. The probability of single or multiple U. affinis galls being present increased with seed-head diameter but was not affected by seed-head height. Attack by L. minutus increased with seed-head diameters >5 mm and was lower at plant heights above 50 cm. These results demonstrate two mechanisms that enable U. affinis to successfully coexist with L. minutus: differences between the species in their response to the developmental phenology of knapweed heads, and increased survivorship of U. affinis in heads with multiple U. affinis galls through niche interference competition. These mechanisms provide a possible explanation for the persistence of U. affinis populations on spotted knapweed, in spite of high levels of within seed-head mortality that have been observed with increasing L. minutus populations.     

The development and multiple uses of a standardised bioassay method to select hypocrealean fungi for biological control of aphids

A technically standardised bioassay method was designed, evaluated and used to assess virulence and host range of hypocrealean fungi against aphids. A track mounted sprayer was used to apply conidia because hand held versions of the same sprayer can be used for field applications, thereby allowing the outcome from laboratory experiments to predict activity in the field accurately. Eighteen fungal isolates were assessed in single concentration bioassays against the black bean aphid Aphis fabae Scopoli. Isolates comprised commercially available mycoinsecticides (based on Beauveria bassiana and Lecanicillium longisporum) and isolates of B. bassiana, Lecanicillium spp., Paecilomyces fumosoroseus and Metarhizium anisopliae. Aphid mortality was in excess of 80% for 15 isolates, and HRI 1.72 (L. longipsorum), Z11 (P. fumosoroseus), Mycotech strain GHA (B. bassiana) and ARSEF 2879 (B. bassiana) were studied further. Multiple concentration bioassays identified HRI 1.72 as the most virulent isolate against A. fabae with significantly smaller LC₅₀ and LT₅₀ values compared to other isolates. A precise LC₅₀ value (2.95 × 10² conidia ml⁻¹) was calculated for HRI 1.72 using a second multiple concentration assay with smaller concentrations of conidia. The four isolates were applied at a single concentration (1 × 10⁸ conidia ml⁻¹) against Myzus persicae, A. fabae, Acyrthosiphon pisum, Metopolophium dirhodum, Sitobion avenae and Rhopalosiphum padi. A ranking of aphid susceptibility was obtained, such that S. avenae > M. persicae, A. pisum, A. fabae > R. padi. Results indicate the importance of standardising bioassay methods to reduce bioassay variability without compromising the ability to use the bioassay to investigate fungus–host interactions under varying abiotic and biotic conditions.     

Estimating susceptibility of biological control agents to pesticides: influence of life history strategies and population structure

In this study we examined the influences that differing life history strategies and population structures at the time of pesticide exposure have on population susceptibility to pesticides. We used life table data and a matrix projection model to incorporate combinations of mortality (lethal effect) and reductions in fecundity (sublethal effect) into estimates of intrinsic population growth rates (r) for a predator, the seven-spot lady beetle, Coccinella septempunctata L., and its prey, the pea aphid, Acyrthosiphon pisum Harris, and an aphid parasitoid, Diaeretiella rapae (M’Intosh). All three species exhibited differences in key life history variables. The aphid had the highest r and shortest generation time, the ladybeetle had the lowest r and longest generation, while the parasitoid exhibited intermediate life history characteristics. When the model was run with populations started as neonates (aphids) or eggs (lady beetle, parasitoid) for each species, ladybeetle populations were much more susceptible than either aphid or parasitoid populations 30 days after simulated exposure to a pesticide. For example, 50% mortality and a 50% reduction in fecundity resulted in a population headed toward extinction (negative r) for the ladybeetle while the parasitoid population grew exponentially (positive r) even after sustaining 70% mortality and a 70% reduction in fecundity. The aphid species maintained exponential growth after sustaining 80% mortality and an 80% reduction in fecundity. Thus, differences in life history variables accounted for the greater susceptibility of the ladybeetle to a pesticide than its aphid prey or the parasitoid over a set time interval. These differences in susceptibility were greatly reduced when the model was run starting with a mixed age/stage population (the stable age distribution) for each species indicating that population structure at the time of pesticide exposure plays a critical role in population susceptibility. These results suggest that life history attributes as well as population structure at the time of pesticide exposure both play a major role in population susceptibility to pesticides, highlighting the need to explicitly consider differences in life history variables among species when calculating compatibility of pesticides and biological control agents as well as the population structure of beneficial species at the time of pesticide application.     

Assessing the damage potential of the stem-boring weevil Lixus aemulus for the biological control of Chromolaena odorata

In a damage trial in pots, Lixus aemulus, a stem-boring weevil imported from Brazil, caused 66.6% mortality of the stems and reduced the dry mass of infested Chromolaena odorata stems by 46.6%. There was a 94.4% reduction in the number of achenes produced on the infested branches. These results indicate that L. aemulus has potential to have impact as a biological control agent of C. odorata.     

Short-distance dispersal behavior and establishment of the parasitoid Gonatocerus ashmeadi (Hymenoptera: Mymaridae) in Tahiti: Implications for its use as a biological control agent against Homalodisca vitripennis (Hemiptera: Cicadellidae)

The egg parasitoid Gonatocerus ashmeadi Girault (Hymenoptera: Mymaridae), was introduced into French Polynesia as a biological control agent to control the invasive plant feeding pest Homalodisca vitripennis (Germar) (Hemiptera: Cicadellidae). The short-distance dispersal of G. ashmeadi was monitored as part of the biological control program. G. ashmeadi showed exponential dispersal capacity with 47 m/day being a minimum estimate of its natural rate of spread at high host densities (>150 nymphs per minute of sweep net sampling) in urbanized areas at sea level, which were characterized by a high diversity of exotic ornamental plants. This rate of spread contrasted starkly with almost nonexistent establishment and dispersal where host densities were very low (<2 nymphs per minute of sweep net sampling) at high elevation (800 m) with relatively undisturbed native vegetation. Survey results across different altitudes revealed an effect of vegetative diversity and host density on the measurable mobility and establishment of G. ashmeadi. In contrast, no significant influence of wind direction was found on G. ashmeadi dispersal rate or direction. Survey results for G. ashmeadi from French Polynesia suggest that the best release establishment strategies for classical biological control of H. vitripennis are: (1) many small releases where host density is high, or (2) larger and fewer releases where host densities are low.     

The system management approach of biological weed control: Some theoretical considerations and aspects of application

Theoretical considerations behind the system management approach of biological weed control are presented. These include, a part describing and explaining the effects of parasitic fungi on crop – weed competition, a part describing and explaining the epidemic spread of parasitic fungi on weeds, and a part relating crop – weed competition at the population level to epidemics. The theoretical framework developed may also provide a basis for the use of other natural enemies, like insects, for biological weed control following the system management approach. Aspects of application are discussed using data of the interaction between the annual weed Senecio vulgaris and the rust fungusPuccinia lagenophorae.     

Characterizing biological aerosol in a chamber: an approach to estimation of viable organisms in a single biological particle

There are practical and valid reasons topresent biological field trial referenceresults as agent containing particles per literof air (ACPLA). However, workers in biologicalaerosol research have a need to know how manyviable individual organisms are contained in asingle particle of a given diameter. Anecdotalevidence may exist suggesting that the task hasbeen accomplished but without a way toreplicate the measurements, it is difficult toaccept unsubstantiated claims. It is verydifficult to declare a finite number thatsatisfies all the experimental requirements, asthe problem is a statistical probability issue. This paper describes a method for estimatingthe number with practical instructions forreplicating the work in other laboratories.The test aerosol was contained in a 90 m³chamber at concentrations as low as 5 ACPLA. Amodified version of slit sampler collectedviable particles. A statistical method was usedto demonstrate sampling predictability at 95%confidence level. By using glass fiber filtersmounted in a dichotomous sampler, samplingefficiencies were estimated for a variety ofcommon aerosol collectors. The accumulated datapermitted the estimation of the number ofviable spores per particle. For a 2.5 to4 µm particle, arguments have beenpresented for considering 4.5 as the mostprobable ACPLA value.     

Use of indigenous fishes to control schistosome snail vectors in Mala i, Africa

The expansion of water resource development projects in the tropics and subtropics resulted in an increased transmission of human schistosomes. This study was conducted to test the feasibility of using two snail-eating fishes, Trematocranus placodon (Trewavas) and Maravichromis anaphyrmis (Trewavas), as biocontrol agents of schistosomes in fish ponds. The results suggest that T. placodon has a potential for controlling the snail vectors of Schistosoma mansoni (Sambon) and Schistosoma haematobium (Bilharz). The use of M. anaphyrmis, however, to control schistosome-carrying snails was not satisfactory.     

Field assessment of host plant specificity and potential effectiveness of a prospective biological control agent, Aceria salsolae, of Russian thistle, Salsola tragus

The eriophyid mite, Aceria salsolae de Lillo and Sobhian, is being evaluated as a prospective classical biological control agent of invasive alien tumbleweeds, including Salsola tragus, S. collina, S. paulsenii and S. australis, in North America. Previous laboratory experiments to determine the host specificity of the mite indicated that it could sometimes persist and multiply on some nontarget plants, including Bassia hyssopifolia and B. scoparia. These are both European plants whose geographic range overlaps that of the mite, but the mite has never been observed on them in the field. A field experiment was conducted in Italy to determine if the mite would infest and damage these plants under natural outdoor conditions. The results indicate that this mite does not attain significant populations on these nontarget plants nor does it significantly damage them. Salsola tragus was heavily infested by A. salsolae, and plant size was negatively correlated to the level of infestation. Although S. kali plants were also infested, their size did not appear to be affected by the mites. The other nontarget plants were not as suitable for the mite in the field as in previous laboratory experiments. We conclude that there would be no significant risk to nontarget plants as a result of using A. salsolae as a biological agent to control Salsola species in North America.     

Colonization of apple orchards by predators of <Emphasis Type="Italic">Dysaphis plantaginea</Emphasis>: sequential arrival,response to prey abundance and consequences for biological control

Episyrphus balteatus (DeGeer) (Diptera: Syrphidae), Adalia bipunctata (L.) (Coleoptera: Coccinellidae) and Aphidoletes aphidimyza (Rondani) (Diptera: Cecidomyiidae) are the three most abundant natural enemies of Dysaphis plantaginea Passerini (Homoptera: Aphididae) in Asturian (NW Spain) apple orchards. They attack this aphid in sequence: E. balteatus arrived first, followed by A. bipunctata and then by A. aphidimyza. The cecidomyiids arrived too late to have a regulating effect. The syrphids laid an average of 2.3 ± 1.7 eggs per aphid colony and the coccinellids 18.4 ± 9.9 regardless of the degree of the infestation rates of the apple shoots. This value corresponds to the size of an egg batch laid by one female. Therefore, these aphid predators did not respond numerically to the abundance of the pest. The results of this study indicate that natural populations of syrphids and ladybird beetles are unable to control D. plantaginea, and therefore a more complex strategy than waiting for natural enemies is required.     

Susceptibility of yellow starthistle to Puccinia jaceae var. solstitialis and greenhouse production of inoculum for classical biological control programs

In anticipation of large-scale distribution of a Turkish isolate of Puccinia jaceae var. solstitialis in California for biological control of yellow starthistle (Centaurea solstitialis, YST), susceptibility of YST within the state was determined and a protocol for bulk inoculum production was developed. Inoculation was made of 62 field accessions of YST representative of the range of habitats in California. These were determined to be equally susceptible to infection by the isolate approved for release in the United States in 2003. To support a program to speed establishment by release at many locations statewide, protocols for artificial increase of inoculum were developed. Over 64 g of urediniospores were produced with a mass-production system under greenhouse conditions from 2003 to 2006. Yield of inoculum varied by season, with peak production occurring from early spring through early summer. A large-scale urediniospore harvest also was made from a field plot at Davis, California. Our results show that susceptibility of YST in California is not likely to limit establishment of P. jaceae for biological control, and that production of this or other obligate pathogenic fungi (biological control agent) is possible for support of statewide release and research programs.     

Successful biological control of mist flower (Ageratina riparia) in New Zealand: Agent establishment, impact and benefits to the native flora

The white smut fungus (Entyloma ageratinae) and the gall fly (Procecidochares alani) were released in New Zealand in 1998 and 2001 respectively to suppress mist flower (Ageratina riparia). The fungus established and spread rapidly, crossing 80 km of sea to Great Barrier Island within 2 years. The mean number of P. alani galls increased exponentially to 1.96/stem at release sites, but dispersal was slow. The impact of the biocontrol agents was monitored once annually from 1998/99 to 2003/04, at up to 51 sites in the North Island. The mean percentage of live leaves infected with fungus rapidly reached nearly 60%. Maximum plant height declined significantly. In heavy infestations, mean percentage cover of mist flower declined from 81 to 1.5%. Galls were only recorded towards the end of the impact study, and at low mean numbers. As mist flower declined, the species richness and mean percentage cover of native plants increased. In contrast, the species richness and mean percentage cover of exotic plants (excluding mist flower) did not change significantly. Many plant species colonizing the plots were important native mid- or late-successional shrubs or trees. With few exceptions, the exotic plant species common in the plots were not weeds that appeared to threaten native forest habitats. There was only a weak “replacement weed effect” from the potentially serious invader African club moss (Selaginella kraussiana). These data, together with reports of reduced threats to rare endemic plants from mist flower, suggest this rapid, well-monitored weed biocontrol program was very successful.     

The co-occurrence of an introduced biological control agent (Coleoptera: <Emphasis Type="Italic">Coccinella septempunctata</Emphasis>) and an endangered butterfly (Lepidoptera: <Emphasis Type="Italic">Lycaeides melissa samuelis</Emphasis>)

Whether a biological control agent presents a non-target risk to a native species depends if they co-occur spatially and temporally, and if the agent will harm the native species. We sampled two study sites during 1993 in Minnesota and Wisconsin to survey predators and parasitoids of the extant populations of the United States federally endangered Karner blue butterfly, Lycaeides melissa samuelis. We found the introduced coccinellid Coccinella septempuntata co-occurring spatially and temporally with eggs, larvae and adults of L. m. samuelis. The two species were also observed together on the latters sole host plant, Lupinus perennis, and in Wisconsin, an adult C. septempunctata was observed consuming second instar larvae of L. m. samuelis. Using a simple model to hypothesize the risk that C. septempunctata presents to L. m. samuelis, we showed that increases in predator density could greatly increase mortality to L. m. samuelis. At these sites, C. septempunctata were reproducing and had access to summer aphids and suitable overwintering habitat. Nearby agricultural crops could provide spring aphids for oogenesis, and assist with C. septempunctata population build-up. Maintaining a minimum isolation distance between agricultural crops known to harbor aphids and extant L. m. samuelis populations may need to be considered as part of the butterfly management program.     

Field assessment of the risk posed by Diorhabda elongata, a biocontrol agent for control of saltcedar (Tamarix spp.), to a nontarget plant, Frankenia salina

The biological control program for saltcedar (Tamarix spp.) has led to open releases of a specialist beetle (Chrysomelidae: Diorhabda elongata) in several research locations, but the controversy over potential impacts to native, nontarget plants of the genus Frankenia remains unresolved. To assess the potential for nontarget impacts under field conditions, we installed cultivated Frankenia spp. (primarily two forms of Frankenia salina but also including Frankenia jamesii) at locations in Nevada and Wyoming where D. elongata densities and saltcedar defoliation were expected to be very high, so insects would be near starvation with high probability of attacking nontargets if these were suitable hosts. Subsequent insect abundance was high, and only minor impact (<4% foliar damage) was observed on both forms of F. salina under these ‘worst case’ conditions; there was no impact to F. jamesii. No oviposition nor larval development were observed on any plants, there was no dieback of damaged F. salina stems, and plants continued growing once insect populations subsided. These results under ‘natural’ field conditions contrast with caged host-range tests in which feeding, development and minor oviposition occurred on the nontarget plant. Other ecological factors, such as distance from target plants to natural Frankenia spp. populations, inhospitable conditions for agent survival in such sites, and intrinsic insect behavior that makes colonization and/or genetic adaptation highly unlikely, lead us to conclude that nontarget impacts following program implementation will be insignificant or absent. Host range testing of new agents, while necessary to ensure safety, must put greater attention on assessing the ecological context where agents will be establishing, and on balancing speculated risks against potential benefits of biological control.     

Recombinant and classically selected factors of potato plant resistance to the Colorado potato beetle, Leptinotarsa decemlineata, variously affect the potato aphid parasitoid Aphidius nigripes

Different forms of crop resistance developed against majorpotato pests such as the Colorado potato beetle (CPB), Leptinotarsadecemlineata, may be variously compatible with biological controlof secondary pests such as aphids. We compared the performance of theparasitoid Aphidius nigripes developing in the aphid hostMacrosiphum euphorbiae, on five potato lines, including atransgenic `Superior-BT' line expressing the CryIIIA toxin ofBacillus thuringiensis, which is specific to Coleoptera; andthe `NYL 235-4' line derived from Solanum berthaultii,characterized by relatively unspecific resistance to herbivores based onmoderately-high density of glandular trichomes. The other lines testedwere a `Kennebec-OCI' transgenic line expressing the protease inhibitorrice cystatin I (OCI), a potential resistance factor against coleopteranpests; and the commercial cultivars `Superior' and `Kennebec' used ascontrols. Parasitoid immature survival and adult size were reducedcompared to controls when the wasps developed on aphids fed the`Superior-BT' potato. In contrast, parasitoid size and fecundityincreased above control when the wasps developed on aphids fed the`Kennebec-OCI' potato. Parasitoids reached the adult stage faster andwere more fecund on `NYL 235-4' than control lines. The results indicatethat the different forms of potato resistance currently developed mainlyagainst the CPB had various unexpected effects on aphid parasitoidfitness. These effects on the parasitoid were complex but were generallyinterpretable in terms of host aphid quality variation among potatolines used as food by the aphids during parasitoiddevelopment.     

The contribution of conservation biological control to integrated control of Bemisia tabaci in cotton

Integrated control systems are based on the complimentary contribution of chemical and biological control fostered by conservation of natural enemies. Yet, in the 50 years since the integrated control concept [ICC] [Stern, V.M., Smith, R.F., van den Bosch, R., Hagen, K.S., 1959. The integrated control concept. Hilgardia 29, 81–101] was introduced there are few operational programs and even fewer attempts to analyze the mechanisms that allow chemical and biological control to act in concert. The dearth of demonstrable evidence for the ICC has eroded the credibility of biological control and its usage in operational IPM plans. We used in situ life tables within an experimental design to measure and compare the contribution and interaction of biological control and insecticides as tactical components within three pest management systems for Bemisia tabaci (Gennadius) in cotton. Insecticides were the key factor immediately following applications of broad-spectrum materials or one of two selective insect growth regulators (IGRs), and this mortality replaced that provided by natural enemies. Two to six weeks later, however, mortality from natural enemies, primarily predation, in the IGR regimes rebounded to the high levels observed in untreated controls and became the key factor. Mortality from natural enemies remained depressed in the broad-spectrum insecticide regime. Single IGR applications were sufficient to suppress B. tabaci populations throughout the season, while up to five broad-spectrum applications were needed to achieve comparable control. The chemical residual of IGRs was limited to several weeks, demonstrating a key role for mortality from conserved natural enemies that extended the control interval. This “bioresidual” allows for long-term, commercially-acceptable pest suppression following the use of selective insecticides. We provide a rare experimental illustration of integrated control, where chemical and biological controls “augment one another”. Our approach and methodology could be applied to demonstrate and validate integrated control in many other systems, addressing a critical need for implementation of biological control in practicing IPM systems.     

The effect of two biological control agents, the weevil Neochetina eichhorniae and the mirid Eccritotarsus catarinensis on water hyacinth, Eichhornia crassipes , grown in culture with water lettuce, Pistia stratiotes

We assessed the effect of two biological control agents, the mirid Eccritotarsus catarinensis (Carvalho) and the weevil Neochetina eichhorniae (Warner), singly or in combination, on the competitive ability of their host plant, water hyacinth, Eichhornia crassipes (Mart.) Solms-Laub., grown in a screen house, in competition with another aquatic plant (Pistia stratiotes L.). Water hyacinth plant growth characteristics measured included fresh weight, leaf and petiole lengths, number of inflorescences produced, and new shoots. Without herbivory, water hyacinth was 18 times more competitive than water lettuce (across all experimental combinations of initial plant densities), as estimated from fresh weights. Both insect species, singly or in combination, reduced water hyacinth plant growth characteristics. E. catarinensis alone was less damaging than the weevil and under normal conditions, i.e., floating water hyacinth, is not expected to increase control of water hyacinth beyond that of the weevil. When combined with the weevil, half the inoculum of weevils and half the inoculum of mirids produced the same growth reduction as the full inoculum of the weevil. Under conditions where the weevils are not effective because water hyacinths are seasonally rooted in mud, the mirid, which lives entirely on leaves, should become a useful additional biological control agent. Handling Editor: John Scott.