tPhoto: Kochia, southwest Montana. © 2020 Delena Norris-Tull
Insects & Other Invertebrates as Biocontrol Agents
Summaries of the research and commentary by Dr. Delena Norris-Tull, Professor Emerita of Science Education, University of Montana Western, July 2020, updated December 2024.
“After some assessment of cost-benefit ratio, the process involves collecting exotic natural enemies to control a target invasive weed, usually followed by importing, rearing, testing, and release from quarantine for establishment. Host specificity tests are conducted in artificial and field conditions, and increasingly combined with ecological and molecular evaluations” (Suckling & Sforza, January, 2014).
Louda, et al., 1997, reported that, in the US, 41% of biocontrol projects, using insects from the plant’s native range, have had some success in reducing populations of the target weed, and 20% of projects have shown significant control of the target weed species. However, that means that 39% had been unsuccessful. A significant number of advances have been made in the use of biocontrols since 1997, as described below.
Parker & Gilbert (2007) conducted field studies to compare the impacts of natural biocontrol invertebrate herbivores and fungi on ten native species of clover and 18 non-native species, including both invasive and non-invasive introduced clovers. They wanted to answer two questions: Do native species suffer greater attack by natural enemies compared to attacks on non-native species.? Are some introduced species excluded from native plant communities because they are susceptible to local natural enemies?
Parker & GIlbert, in a four year study involving both field observations and garden experiments, found that native and introduced clovers were equally impacted by native pathogens. And they found that, although in the first two years invertebrate herbivores caused more damage on native clovers, the impact over four years showed no difference in mortality between introduced and native clovers. Thus, in contrast to research by Rand & Louda (2004), and by Parker, Burkepile, & Hay (2006), Parker & Gilbert did not find that natural enemies caused more harm to native plants than to introduced plants. Rather, they found that "the most invasive species showed greater infection, greater prevalence and severity of disease, greater prevalence of herbivory, and greater effects of fungicide on biomass." Also, invasive plants were "indistinguishable from noninvasive introduced species in all other respects."
In their four year garden experiments, they recorded invertebrate herbivores (thrips, snail, aphids, lepidopterans and coleopterans) and the damage they caused. And they located the most common clover pathogen and conducted "a controlled inoculation experiment to measure both the ability to infect and the tendency to cause disease on native and introduced clover hosts." They compared "prevalence of disease symptoms and herbivory. And they recorded severity of damage symptoms." Because they "found no relationship between the percentage of leaf damage per plant for disease and for herbivory," they concluded that "herbivores do not drive disease dynamics." And there was "no evidence in this system of a group of host-specific natural enemies that preferentially attack native species."
They found "no significant differences between native and nonnative species for the degree to which leaves were infected by fungi." However, "The trend was for the two most invasive species to show the highest levels of infection." But "The most invasive introduced species...never showed less damage from pathogens than did the noninvasive introduced species."
Damage from herbivores "tended to be more prevalent on the most invasive introduced species... Severity of damage was not significantly different between invasive and noninvasive species."
In years three and four, they compared the effectiveness of two antifungal agents. The agents had no phytotoxic effects. "Fungicides were successful in reducing, but not eliminating, fungal infection... in year 4, infection was reduced by 21% and symptoms by about half... The reduction in mortality due to the fungicide treatment was not significantly different between native and introduced species." In ear 4, "Mortality was not lower in fungicide plants than in control plants." "We found no significant difference between the response of native and introduced plants to fungicide."
Herbivore impact on flowers
Jogesh, et al. (2014), state that, "Selection pressure exerted by herbivory is thought to have shaped a wide range of plant defence traits over macroevolutionary time... Herbivores may also cause rapid adaptive evolution in plants over contemporary timescales. However, documenting microevolutionary change is challenging given that response to selection can be influenced by other processes within the ecological community. Only a handful of studies have quantified real-time ecological and evolutionary change in plants in response to selection by herbivores." Among other defenses that have evolved in plants, to repel animals from eating them, are chemical defenses. The work by Jogesh, et al., is so comprehensive and unique in its design, that I have here included quite a bit of information from their research.
Jogesh, et al. (2014), conducted a unique study examining the chemical content of wild parsnip, Pastinaca sativa, in its native habitat in Europe, and in two locations where it has been introduced and become invasive, New Zealand and the USA. Wild parsnip has been in the USA for about 400 years, and in New Zealand since about 1867. Wild parsnip contains a variety of toxic chemicals in all of its aboveground tissues, which discourage herbivores.
One of the types of herbivores that attacks this species in its native habitat is a florivore, an herbivore that eats the products of flowers, including buds and petals, grain or seeds, fruit, pollen, and/or nectar. In Europe, wild parsnip has coevolved with a specialist florivore, the parsnip webworm, Depressaria pastinacella. For the first several hundred years that wild parsnip grew in the USA, the parsnip webworm was not here. The webworm was introduced in the USA in 1869, and New Zealand in 2004.
Jogesh, et al. (2014), wanted to find out if evolutionary processes occur rapidly enough to alter the chemistry of wild parsnips in New Zealand. So they compared plant chemistry of plants from New Zealand, the USA, and Europe.
Seeds were collected from nine New Zealand populations from 2006 to 2009, and, due to lack of access to a greenhouse, the seeds were planted in a common garden in Sawyers Bay, Otago. Seeds were collected from five populations in central Illinois in 2008. These seeds were sprouted in greenhouses and the seedlings were then transplanted to common gardens at the University of Illinois. From each population, at least 40 seeds from each of the 14 populations were collected and planted in the spring.
The following spring, all plants that survived were scored for herbivore damage. In the two weeks between oviposition and egg hatch, one member of each half-sibling pair was sprayed with insecticide, to create an herbivore-free environment. Flowers were collected for chemical analysis. Seeds were weighed, as an indicator of plant fitness. After all seeds were collected, the plants were dried and weighed to measure aboveground biomass.
Statistical Analyses and Results:
Jogesh, et al. (2014), state that, "Of the 1238 parsnips planted in the US garden, 685 plants bolted, flowered and set seed. In NZ, 1603 parsnip seeds were sown and 626 plants bolted, flowered and set seed. Of the plants that bolted, some were lost to wind damage and some individuals could not be measured for certain parameters due to losses during processing."
(1) Impact of webworms on parsnip fitness:
Because wild parsnip is a biennial, Jogesh, et al., were able to analyze chemical content of the first year rosettes and the second year adult plants. Flowers were also analyzed for dry weight and chemistry. The data analysis was designed to examine the "impact of webworms on fitness and to compare susceptibility of US and NZ plants (genotypic effects) grown in different gardens (environmental effects). Realized fitness was the dependent variable and country of origin (USA, NZ), garden (2010, 2011) and spray treatment were included as fixed effects, population was a random effect, and plant size (stalk weight) was added as a covariate in a linear mixed-effects model. Realized fitness was estimated as log-transformed (to ensure equal variance) total seed weight. (They) used this model to determine the effects of (1) spray treatment (do webworms have an impact on fitness?), (2) country X spray (are NZ parsnips more susceptible to webworms than US parsnips?), (3) garden X spray (does the NZ environment favour webworm susceptibility?), (4) country X garden X spray (is there a genotype-by-environment interaction in webworm susceptibility?) (5) garden X country (is there a genotype-by-environment interaction in realized fitness?) and (6) stalk weight (does realized fitness change with plant size?)."
"To quantify impacts of webworm herbivory on the cost of seed production, (they) calculated plant reproductive effort as the ratio of seed weight to stalk weight. Reproductive effort measures the proportion of available resources (biomass) allocated to seed production and as such accounts for variation in realized fitness that may be due to resource availability and thus the relative cost of investing in seeds. Reproductive effort was fitted with a linear mixed model with country of origin, spray treatment and garden location as fixed effects, and population as a random effect."
Results: "Irrespective of locality, both realized fitness and reproductive effort (the allocation of biomass to seed production) were higher in plants without webworms. (They) did not find a significant country X spray treatment interaction or a garden X spray treatment interaction, suggesting that the effect of webworm florivory on realized fitness does not differ with environment or genetic background ... However, insecticide treatment increased reproductive effort in the US garden but not in the NZ garden (...significant spray X garden effect), suggesting that environmental factors influence the effect of webworms on the cost of seed production (allocation of biomass to seeds) but not the effect of webworms on overall realized fitness."
Conclusion: Webworms reduce parsnip fitness.
(2) Impact of florivores on pollination success:
Jogesh, et al. (2014), also "compared pollination success between (1) US and NZ parsnips; (2) sprayed vs. unsprayed plants; (3) the two different garden plots; and (4) plant size; to determine whether pollination success was influenced by the country of origin (genotype); florivory, the environment, or plant size, or interactions among these factors. Pollination success was estimated from the weight of fertilized seeds and the total number of available seeds."
Results: They "did not find a significant country X spray treatment interaction or a garden X spray treatment interaction, suggesting that the effect of webworm florivory on realized fitness does not differ with environment or genetic background... However, insecticide treatment increased reproductive effort in the US garden but not in the NZ garden (...significant spray X garden effect), suggesting that environmental factors influence the effect of webworms on the cost of seed production (allocation of biomass to seeds) but not the effect of webworms on overall realized fitness."
Conclusion: Effect of webworms on pollination success is mediated by plant size.
(3) Influence of genotype and environment on fitness traits:
They also "compared plant size, total damage, proportional damage and plant chemistry between US and NZ parsnips grown in different gardens to determine whether these traits: (1) differed between parsnips depending on their country of origin (genotype); (2) differed between gardens (environment); and (3) were influenced by genotype-by-environment interaction."
Results: Parsnips grown in NZ were on average 15 times larger than their half-sibs in the US garden. The substantial difference in size is likely a reflection of different abiotic conditions between the two gardens. Plant size was a very strong predictor of realized fitness with larger plants having proportionally higher fitness compared with smaller plants... Total damage and pollination success were also strongly correlated with plant size... Significant genotype-by-environment effects and differences between half-sibs in both gardens suggests that many traits are phenotypically plastic."
Conclusion: Genotype-by-environmental interactions influence parsnip fitness.
(4) Half-sib comparisons to determine environmental effects:
They "compared eight floral compounds, size, realized fitness and damage between half-sibs grown in US and NZ gardens to test for trait differences attributable to environmental factors."
Results: "Floral chemistry was strongly determined by genotype-by-environment interactions (country of origin X garden...), indicating phenotypic plasticity in the expression of many floral compounds. Examination of individual compounds and comparison of half-sibs suggests that some compounds (bergapten, xanthotoxin and octyl acetate) are not environmentally variable, whereas others (imperatorin, isopimpinellin, sphondin, myristicin and octyl butyrate) are strongly environmentally determined... The phenotypic plasticity of octyl butyrate, isopimpinellin and imperatorin was partially driven by plant size; large plants produced higher amounts of octyl butyrate and lower amounts of isopimpinellin and imperatorin."
Conclusion: Genotype-by-environment interactions influence floral chemistry
(5) Evolution of parsnips in New Zealand:
"To determine whether susceptibility to webworms decreased in plants associated with webworms over time, we compared NZ parsnips with different infestation histories (infested for 3–6 years and never infested) over the period 2006 to 2009."
Results: "Reduced realized fitness as a result of webworm florivory was consistent in parsnip populations with no infestation history from 2006 to 2009. However, parsnip populations with 3 and 6 years of webworm infestation showed a fitness loss due to florivory in the years 2006 and 2007 but a smaller effect of webworms in 2008 and 2009, suggesting the evolution of tolerance to herbivory although with small sample sizes and high year-to-year variation, this interaction did not prove to be statistically significant... Parsnips from infested populations were significantly larger after 3–6 years of infestation compared with parsnips from uninfested populations in 2009...Parsnips with a history of infestation did not have lower total damage or higher pollination success in 2009 compared with parsnips with no infestation history."
Conclusion: Large size evolves in infested New Zealand parsnip populations.
Discussion: Jogesh, et al. (2014), "found no evidence of change in overall floral chemistry in all (New Zealand) populations." The results demonstrate that "6 years of reassociation between wild parsnip and its specialized herbivore has resulted in the evolution of increased plant size in New Zealand. Plant fitness was strongly dependent on plant size and size also mediated the effect of florivores on pollination success. Even though we found no evidence for increased chemical defences, large plants have much higher pollination success and realized fitness in spite of higher webworm damage and size is likely under strong selection after 3–6 years of intense florivory."
Longterm use of Biocontrols ineffective in general
Sheley, et al., 2011, conducted an extensive review of the research on the long-term effectiveness of biological control agents on rangeland ecoystems. They concluded that, with the exception of a few highly effective agents, overall, biological control agents have not proven effective for large-scale infestations over the long-term. And “the risks of deleterious off-target effects increase with the number of releases.” The few documented successes include two beetles that have been highly effective with reducing infestations of St. John’s wort, and three insects used on ragwort. They also found some success with the Diorhabda species (leaf beetles) used on salt cedar (see concerns about these species, described under “Indirect effects”). The advantage of biological control agents is that they are relatively inexpensive, when compared with herbicides (which are often unsuccessful). They found that a number of studies demonstrated that native species do successfully re-emerge, after reduction of invasive species via biological control agents, although not all studies had this result.
Sheley, et al., 2011, found that, “When operating under current protocols there are relatively few documented direct effects… on desirable vegetation… There is, however, mounting evidence suggesting that poor monitoring efforts, difficulty in predicting biocontrol effects, and the largely unrecognized indirect effects biocontrols can have on ecosystems contributes to an underestimation of the detrimental effects…on desirable vegetation… For example, the bulk of biocontrol monitoring focuses on release sites with little attention paid to offsite biocontrol effects even though there is strong evidence demonstrating landscape-scale variation in biocontrol effects on desirable vegetation… In addition, it is estimated that less than half of the biological control efforts… in the United States demonstrated any evidence of control…. [In addition] recent literature [shows] complex indirect effects of biocontrol on desirable vegetation. For example, following the collapse of the target [invasive] population, intense competition among biocontrol agents can cause a transient increase in host [invasive] plant range, which results in the biocontrol agents attacking desirable species… When biocontrol agents only moderately damage invasion plants they may increase invasive plant…compensatory growth… Current procedures do not adequately prevent biocontrol efforts from having significant effects on desirable vegetation.” Other research indicates that biocontrol insects can have negative direct or indirect impacts on native insects and other animals, such as mice, within the food chain.
References:
Next Sections on Insects as Biocontrol Agents:
Additional reports on Biocontrol Agents:
Insects & Other Invertebrates as Biocontrol Agents
Summaries of the research and commentary by Dr. Delena Norris-Tull, Professor Emerita of Science Education, University of Montana Western, July 2020, updated December 2024.
“After some assessment of cost-benefit ratio, the process involves collecting exotic natural enemies to control a target invasive weed, usually followed by importing, rearing, testing, and release from quarantine for establishment. Host specificity tests are conducted in artificial and field conditions, and increasingly combined with ecological and molecular evaluations” (Suckling & Sforza, January, 2014).
Louda, et al., 1997, reported that, in the US, 41% of biocontrol projects, using insects from the plant’s native range, have had some success in reducing populations of the target weed, and 20% of projects have shown significant control of the target weed species. However, that means that 39% had been unsuccessful. A significant number of advances have been made in the use of biocontrols since 1997, as described below.
Parker & Gilbert (2007) conducted field studies to compare the impacts of natural biocontrol invertebrate herbivores and fungi on ten native species of clover and 18 non-native species, including both invasive and non-invasive introduced clovers. They wanted to answer two questions: Do native species suffer greater attack by natural enemies compared to attacks on non-native species.? Are some introduced species excluded from native plant communities because they are susceptible to local natural enemies?
Parker & GIlbert, in a four year study involving both field observations and garden experiments, found that native and introduced clovers were equally impacted by native pathogens. And they found that, although in the first two years invertebrate herbivores caused more damage on native clovers, the impact over four years showed no difference in mortality between introduced and native clovers. Thus, in contrast to research by Rand & Louda (2004), and by Parker, Burkepile, & Hay (2006), Parker & Gilbert did not find that natural enemies caused more harm to native plants than to introduced plants. Rather, they found that "the most invasive species showed greater infection, greater prevalence and severity of disease, greater prevalence of herbivory, and greater effects of fungicide on biomass." Also, invasive plants were "indistinguishable from noninvasive introduced species in all other respects."
In their four year garden experiments, they recorded invertebrate herbivores (thrips, snail, aphids, lepidopterans and coleopterans) and the damage they caused. And they located the most common clover pathogen and conducted "a controlled inoculation experiment to measure both the ability to infect and the tendency to cause disease on native and introduced clover hosts." They compared "prevalence of disease symptoms and herbivory. And they recorded severity of damage symptoms." Because they "found no relationship between the percentage of leaf damage per plant for disease and for herbivory," they concluded that "herbivores do not drive disease dynamics." And there was "no evidence in this system of a group of host-specific natural enemies that preferentially attack native species."
They found "no significant differences between native and nonnative species for the degree to which leaves were infected by fungi." However, "The trend was for the two most invasive species to show the highest levels of infection." But "The most invasive introduced species...never showed less damage from pathogens than did the noninvasive introduced species."
Damage from herbivores "tended to be more prevalent on the most invasive introduced species... Severity of damage was not significantly different between invasive and noninvasive species."
In years three and four, they compared the effectiveness of two antifungal agents. The agents had no phytotoxic effects. "Fungicides were successful in reducing, but not eliminating, fungal infection... in year 4, infection was reduced by 21% and symptoms by about half... The reduction in mortality due to the fungicide treatment was not significantly different between native and introduced species." In ear 4, "Mortality was not lower in fungicide plants than in control plants." "We found no significant difference between the response of native and introduced plants to fungicide."
Herbivore impact on flowers
Jogesh, et al. (2014), state that, "Selection pressure exerted by herbivory is thought to have shaped a wide range of plant defence traits over macroevolutionary time... Herbivores may also cause rapid adaptive evolution in plants over contemporary timescales. However, documenting microevolutionary change is challenging given that response to selection can be influenced by other processes within the ecological community. Only a handful of studies have quantified real-time ecological and evolutionary change in plants in response to selection by herbivores." Among other defenses that have evolved in plants, to repel animals from eating them, are chemical defenses. The work by Jogesh, et al., is so comprehensive and unique in its design, that I have here included quite a bit of information from their research.
Jogesh, et al. (2014), conducted a unique study examining the chemical content of wild parsnip, Pastinaca sativa, in its native habitat in Europe, and in two locations where it has been introduced and become invasive, New Zealand and the USA. Wild parsnip has been in the USA for about 400 years, and in New Zealand since about 1867. Wild parsnip contains a variety of toxic chemicals in all of its aboveground tissues, which discourage herbivores.
One of the types of herbivores that attacks this species in its native habitat is a florivore, an herbivore that eats the products of flowers, including buds and petals, grain or seeds, fruit, pollen, and/or nectar. In Europe, wild parsnip has coevolved with a specialist florivore, the parsnip webworm, Depressaria pastinacella. For the first several hundred years that wild parsnip grew in the USA, the parsnip webworm was not here. The webworm was introduced in the USA in 1869, and New Zealand in 2004.
Jogesh, et al. (2014), wanted to find out if evolutionary processes occur rapidly enough to alter the chemistry of wild parsnips in New Zealand. So they compared plant chemistry of plants from New Zealand, the USA, and Europe.
Seeds were collected from nine New Zealand populations from 2006 to 2009, and, due to lack of access to a greenhouse, the seeds were planted in a common garden in Sawyers Bay, Otago. Seeds were collected from five populations in central Illinois in 2008. These seeds were sprouted in greenhouses and the seedlings were then transplanted to common gardens at the University of Illinois. From each population, at least 40 seeds from each of the 14 populations were collected and planted in the spring.
The following spring, all plants that survived were scored for herbivore damage. In the two weeks between oviposition and egg hatch, one member of each half-sibling pair was sprayed with insecticide, to create an herbivore-free environment. Flowers were collected for chemical analysis. Seeds were weighed, as an indicator of plant fitness. After all seeds were collected, the plants were dried and weighed to measure aboveground biomass.
Statistical Analyses and Results:
Jogesh, et al. (2014), state that, "Of the 1238 parsnips planted in the US garden, 685 plants bolted, flowered and set seed. In NZ, 1603 parsnip seeds were sown and 626 plants bolted, flowered and set seed. Of the plants that bolted, some were lost to wind damage and some individuals could not be measured for certain parameters due to losses during processing."
(1) Impact of webworms on parsnip fitness:
Because wild parsnip is a biennial, Jogesh, et al., were able to analyze chemical content of the first year rosettes and the second year adult plants. Flowers were also analyzed for dry weight and chemistry. The data analysis was designed to examine the "impact of webworms on fitness and to compare susceptibility of US and NZ plants (genotypic effects) grown in different gardens (environmental effects). Realized fitness was the dependent variable and country of origin (USA, NZ), garden (2010, 2011) and spray treatment were included as fixed effects, population was a random effect, and plant size (stalk weight) was added as a covariate in a linear mixed-effects model. Realized fitness was estimated as log-transformed (to ensure equal variance) total seed weight. (They) used this model to determine the effects of (1) spray treatment (do webworms have an impact on fitness?), (2) country X spray (are NZ parsnips more susceptible to webworms than US parsnips?), (3) garden X spray (does the NZ environment favour webworm susceptibility?), (4) country X garden X spray (is there a genotype-by-environment interaction in webworm susceptibility?) (5) garden X country (is there a genotype-by-environment interaction in realized fitness?) and (6) stalk weight (does realized fitness change with plant size?)."
"To quantify impacts of webworm herbivory on the cost of seed production, (they) calculated plant reproductive effort as the ratio of seed weight to stalk weight. Reproductive effort measures the proportion of available resources (biomass) allocated to seed production and as such accounts for variation in realized fitness that may be due to resource availability and thus the relative cost of investing in seeds. Reproductive effort was fitted with a linear mixed model with country of origin, spray treatment and garden location as fixed effects, and population as a random effect."
Results: "Irrespective of locality, both realized fitness and reproductive effort (the allocation of biomass to seed production) were higher in plants without webworms. (They) did not find a significant country X spray treatment interaction or a garden X spray treatment interaction, suggesting that the effect of webworm florivory on realized fitness does not differ with environment or genetic background ... However, insecticide treatment increased reproductive effort in the US garden but not in the NZ garden (...significant spray X garden effect), suggesting that environmental factors influence the effect of webworms on the cost of seed production (allocation of biomass to seeds) but not the effect of webworms on overall realized fitness."
Conclusion: Webworms reduce parsnip fitness.
(2) Impact of florivores on pollination success:
Jogesh, et al. (2014), also "compared pollination success between (1) US and NZ parsnips; (2) sprayed vs. unsprayed plants; (3) the two different garden plots; and (4) plant size; to determine whether pollination success was influenced by the country of origin (genotype); florivory, the environment, or plant size, or interactions among these factors. Pollination success was estimated from the weight of fertilized seeds and the total number of available seeds."
Results: They "did not find a significant country X spray treatment interaction or a garden X spray treatment interaction, suggesting that the effect of webworm florivory on realized fitness does not differ with environment or genetic background... However, insecticide treatment increased reproductive effort in the US garden but not in the NZ garden (...significant spray X garden effect), suggesting that environmental factors influence the effect of webworms on the cost of seed production (allocation of biomass to seeds) but not the effect of webworms on overall realized fitness."
Conclusion: Effect of webworms on pollination success is mediated by plant size.
(3) Influence of genotype and environment on fitness traits:
They also "compared plant size, total damage, proportional damage and plant chemistry between US and NZ parsnips grown in different gardens to determine whether these traits: (1) differed between parsnips depending on their country of origin (genotype); (2) differed between gardens (environment); and (3) were influenced by genotype-by-environment interaction."
Results: Parsnips grown in NZ were on average 15 times larger than their half-sibs in the US garden. The substantial difference in size is likely a reflection of different abiotic conditions between the two gardens. Plant size was a very strong predictor of realized fitness with larger plants having proportionally higher fitness compared with smaller plants... Total damage and pollination success were also strongly correlated with plant size... Significant genotype-by-environment effects and differences between half-sibs in both gardens suggests that many traits are phenotypically plastic."
Conclusion: Genotype-by-environmental interactions influence parsnip fitness.
(4) Half-sib comparisons to determine environmental effects:
They "compared eight floral compounds, size, realized fitness and damage between half-sibs grown in US and NZ gardens to test for trait differences attributable to environmental factors."
Results: "Floral chemistry was strongly determined by genotype-by-environment interactions (country of origin X garden...), indicating phenotypic plasticity in the expression of many floral compounds. Examination of individual compounds and comparison of half-sibs suggests that some compounds (bergapten, xanthotoxin and octyl acetate) are not environmentally variable, whereas others (imperatorin, isopimpinellin, sphondin, myristicin and octyl butyrate) are strongly environmentally determined... The phenotypic plasticity of octyl butyrate, isopimpinellin and imperatorin was partially driven by plant size; large plants produced higher amounts of octyl butyrate and lower amounts of isopimpinellin and imperatorin."
Conclusion: Genotype-by-environment interactions influence floral chemistry
(5) Evolution of parsnips in New Zealand:
"To determine whether susceptibility to webworms decreased in plants associated with webworms over time, we compared NZ parsnips with different infestation histories (infested for 3–6 years and never infested) over the period 2006 to 2009."
Results: "Reduced realized fitness as a result of webworm florivory was consistent in parsnip populations with no infestation history from 2006 to 2009. However, parsnip populations with 3 and 6 years of webworm infestation showed a fitness loss due to florivory in the years 2006 and 2007 but a smaller effect of webworms in 2008 and 2009, suggesting the evolution of tolerance to herbivory although with small sample sizes and high year-to-year variation, this interaction did not prove to be statistically significant... Parsnips from infested populations were significantly larger after 3–6 years of infestation compared with parsnips from uninfested populations in 2009...Parsnips with a history of infestation did not have lower total damage or higher pollination success in 2009 compared with parsnips with no infestation history."
Conclusion: Large size evolves in infested New Zealand parsnip populations.
Discussion: Jogesh, et al. (2014), "found no evidence of change in overall floral chemistry in all (New Zealand) populations." The results demonstrate that "6 years of reassociation between wild parsnip and its specialized herbivore has resulted in the evolution of increased plant size in New Zealand. Plant fitness was strongly dependent on plant size and size also mediated the effect of florivores on pollination success. Even though we found no evidence for increased chemical defences, large plants have much higher pollination success and realized fitness in spite of higher webworm damage and size is likely under strong selection after 3–6 years of intense florivory."
Longterm use of Biocontrols ineffective in general
Sheley, et al., 2011, conducted an extensive review of the research on the long-term effectiveness of biological control agents on rangeland ecoystems. They concluded that, with the exception of a few highly effective agents, overall, biological control agents have not proven effective for large-scale infestations over the long-term. And “the risks of deleterious off-target effects increase with the number of releases.” The few documented successes include two beetles that have been highly effective with reducing infestations of St. John’s wort, and three insects used on ragwort. They also found some success with the Diorhabda species (leaf beetles) used on salt cedar (see concerns about these species, described under “Indirect effects”). The advantage of biological control agents is that they are relatively inexpensive, when compared with herbicides (which are often unsuccessful). They found that a number of studies demonstrated that native species do successfully re-emerge, after reduction of invasive species via biological control agents, although not all studies had this result.
Sheley, et al., 2011, found that, “When operating under current protocols there are relatively few documented direct effects… on desirable vegetation… There is, however, mounting evidence suggesting that poor monitoring efforts, difficulty in predicting biocontrol effects, and the largely unrecognized indirect effects biocontrols can have on ecosystems contributes to an underestimation of the detrimental effects…on desirable vegetation… For example, the bulk of biocontrol monitoring focuses on release sites with little attention paid to offsite biocontrol effects even though there is strong evidence demonstrating landscape-scale variation in biocontrol effects on desirable vegetation… In addition, it is estimated that less than half of the biological control efforts… in the United States demonstrated any evidence of control…. [In addition] recent literature [shows] complex indirect effects of biocontrol on desirable vegetation. For example, following the collapse of the target [invasive] population, intense competition among biocontrol agents can cause a transient increase in host [invasive] plant range, which results in the biocontrol agents attacking desirable species… When biocontrol agents only moderately damage invasion plants they may increase invasive plant…compensatory growth… Current procedures do not adequately prevent biocontrol efforts from having significant effects on desirable vegetation.” Other research indicates that biocontrol insects can have negative direct or indirect impacts on native insects and other animals, such as mice, within the food chain.
References:
- Louda, S.M., Kendall, D., Connor, J., & Simberloff, D. (22 August, 1997). Ecological effects of an insect introduced for biological control of weeds. Science 277 (5329), 1088-1090.
- Jogesh, T., Stanley, M.C., & Berenbaum, M.R. (2014). Evolution of tolerance in an invasive weed after reassociation with its specialist herbivore. Journal of Evolutionary Biology, 27, 2334-2346.
- Parker, I.M, & Gilbert, G.S. (2007). When there is no escape: The effects of natural enemies on native, invasive, and noninvasive plants. Ecology, 88(5), 1210-1224.
- Sheley, R.L., James, J.J., Rinella, M. J., Blumenthal, D., & DiTomaso, J.M. (2011). Invasive plant management on anticipated conservation benefits: A scientific assessment. In D.D. Briske (Ed.) Conservation benefits of rangeland practices: Assessment, recommendation, and knowledge gaps. (pp. 293-336). USDA Natural Resources Conservation Service.
- Suckling, D.M., & Sforza, R.F.H. (January, 2014). What magnitude are observed non-target impacts from weed biocontrol? PLoS ONE 9(1).
Next Sections on Insects as Biocontrol Agents:
Additional reports on Biocontrol Agents: