A genetic analysis of the species and intraspecific lineages of Dactylopius Costa (Hemiptera: Dactylopiidae)
- Authors: Van Steenderen, Clarke Julian Mignon
- Date: 2020
- Subjects: Dactylopius
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/151491 , vital:39135
- Description: The Cactaceae family comprises 15 genera and nearly 2000 species. With one exception, these are all native to the Americas. Numerous cactaceous species are invasive in other parts of the world, resulting in considerable damage to ecosystem functioning and agricultural practices. The most successful biological control agents used to combat invasive Cactaceae belong to the Dactylopius genus (Hemiptera: Dactylopiidae), comprising eleven species. The Dactylopiidae are exclusively cactophagous and are usually host-specific. Some intraspecific lineages of dactylopiids, often referred to as `biotypes', also display host-specificity, and are used to control particular species of invasive Cactaceae. To date, two lineages within Dactylopius opuntiae (`ficus' and `stricta'), and two within D. tomentosus (`cholla' and `imbricata') have been released in South Africa to control Opuntia ficus-indica and O. stricta, and Cylindropuntia fulgida and C. imbricata, respectively. The `californica var. parkeri' lineage is currently under consideration for release in South Africa for the control of C. pallida. Australia has already released these five lineages, and approved the release of an additional three in 2017; namely D. tomentosus `bigelovii', `cylindropuntia sp.', and `acanthocarpa x echinocarpa'. Many of the Dactylopius species are so morphologically similar, and in the case of lineages, identical, that numerous misidentifications have been made in the past. These errors have had serious implications, such as failed attempts at the biological control of cactus weeds. This thesis aimed to generate a multi-locus genetic database to enable the identification of the species and lineages in the Dactylopiidae family, and to test its accuracy. Seven species were included in the analysis, including two lineages within D. opuntiae and six within D. tomentosus. Genetic characterisation was achieved through the DNA sequencing of three gene regions; namely mitochondrial 12S rRNA and cytochrome c oxidase I (COI), nuclear 18S rRNA, and fragment analysis using two inter-simple sequence repeats (ISSRs). Nucleotide sequences were very effective for species-level identification, where the 12S, 18S, and COI regions showed 100%, 94.59%, and 100% identification accuracy rates, respectively. Additionally, the 12S and COI markers distinguished between half of the D. tomentosus lineages (`californica', `cholla', and `imbricata'), with identification accuracies of 100%. The `echinocarpa x acanthocarpa', `bigelovii', and `cylindropuntia sp.' lineages formed one clade. None of the DNA genetic markers showed a separation between the `ficus' and `stricta' lineages within D. opuntiae. Fragment analysis through the use of ISSRs provided higher-resolution results, and addressed this gap by showing a well-supported separation between the two lineages, and between wild populations collected in the Eastern Cape Province in South Africa. The identification accuracy of the `ficus' and `stricta' lineages was 81.82%. This is the first time that a method has been developed that can distinguish between these lineages. An additional component of this thesis was the creation of three user-friendly R-based programs to assist with: 1. ISSR data processing. 2. The identification of query Dactylopius nucleotide sequences relative to the gene databases created here. 3. A graphical user interface (GUI) version of the R package `SPIDER', which is useful for the assessment of the accuracy of genetic barcode data. A successful biological control programme relies on the correct identification of the agent in question, and so it is imperative that cactus biological control practitioners are able to distinguish between Dactylopius species and lineages in order to release the most effective ones onto target Cactaceae. The laboratory protocols reported, and data processing tools created here, have largely addressed this need and offer valuable practical applications. These include: 1. The flagging of potential new species, cryptic species, and lineages of dactylopiid species released as new biocontrol agents. 2. Validating the identifications made by taxonomists based on morphology. 3. Confirming to which species, and, where applicable, to which lineage, a field-collected sample belongs. 4. Identifying hybrids resulting from lineage crosses. Ensuring that the correct Dactylopius species are utilised for biological control will improve the control of invasive Cactaceae and protect biodiversity and agricultural productivity.
- Full Text:
- Date Issued: 2020
Augmentative releases of Dactylopius austrinus De Lotto (Dactylopiidae; Hemiptera) for biological control of Opuntia aurantiaca Lindley (Cactaceae), in South Africa
- Authors: Mulateli, Thifhelimbilu
- Date: 2019
- Subjects: Opuntia aurantiaca -- Biolotical control -- South Africa , Invasive plants -- Biolotical control -- South Africa , Dactylopius austrinus De Lotto -- South Africa , Dactylopius -- South Africa
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/92931 , vital:30765
- Description: Opuntia aurantiaca Lindely (Cactaceae) is an invasive alien cactus which has detrimental effects on agroecosystems and indigenous biodiversity in South Africa. Dense infestations over large areas reduce grazing capacity and indigenous biodiversity. Despite the release of a biological control agent, the cochineal insect Dactylopius austrinus De Lotto (Dactylopiidae), the weed is still considered a major problem in many parts of the country. Biological control has relied heavily on classical biological control, with little augmentative biological control implemented. This study investigated the outcome of mass-rearing and augmentative releases of D. austrinus for the control of O. aurantiaca. Augmentative releases are thought to improve the level of control by increasing agent densities in the field and thus increasing the level of damage inflicted to the plants. All data were collected with the intention to optimize release strategies so that the maximum benefit from the biological control agent could be achieved. An impact study was conducted using potted plants in a greenhouse to quantify the efficacy of multiple releases of the agent on the target weed. All three of the release treatments showed consistently higher proportion of cochineal than the controls, as well as the insect exclusion treatments, and these differences were statistically significant. The number of cladodes per plant increased significantly for the insect exclusion and control treatment over the period of the study, whilst all three release treatments decreased steadily over the same period. This study indicated that the agent is damaging to O. aurantiaca and that a single release event was beneficial but that multiple releases did not result in greater levels of control. A post-release evaluation was carried out to quantify the impact of releases of D. austrinus on O. aurantiaca in the field. Plots where the agent was excluded were compared with those where the agent was left at natural field densities and three treatments where agent populations were augmented to varying degrees through releases. The percentage of cochineal infested cladodes for all treatments decreased over time from the initiation of the experiment in October 2017 until the end of the experiment in October 2018. Opuntia aurantiaca densities also decreased over time for all treatments. The insect exclusion treatment had the greatest number of plants for the duration of the study, but this was not significantly different from other treatments. Dactylopius austrinus was damaging to O. aurantiaca, but climatic conditions in the field limited the efficacy of releases. Although O. aurantiaca density decreased during the experiment, it was evident that the reduced number of plants was not due to augmentation of the cochineal populations from the releases that were conducted. The experiment was conducted over a very dry period, when cochineal was particularly effective, so although augmentative releases did not improve the level of control, the natural population of cochineal was high and very damaging to O. aurantiaca over the course of the experiment. Releasing during wet periods, when the agent is less effective, could augment agent populations at a time when natural populations would be low, and hence improve levels of control further. Although this study was limited to a short period of two years, the results of this study suggest that the number of releases is less important than the timing of releases. Releasing immediately after periods of high rainfall is likely to be beneficial, while releasing during dry periods, or during winter when temperatures are low, is less effective. Dactylopius austrinus populations should be constantly monitored so that releases can be conducted when cochineal populations are low and the climatic conditions are correct. If the timing of release events is appropriate, then the over level of control of O. aurantiaca using D. austrinus could be improved.
- Full Text:
- Date Issued: 2019
Reproductive isolation mechanisms of two cryptic species of Eccritotarsus (Hemiptera: Miridae), biological control agents of water hyacinth, Eichhornia crassipes (Martius) Solms-Laubach (Pontederiaceae)
- Authors: Mnguni, Sandiso
- Date: 2019
- Subjects: Eccritotarsus , Meridae , Noxious weeds -- Biological control -- South Africa , Aquatic weeds -- Biological control -- South Africa , Water hyacinth -- Biological control -- South Africa , Biological pest control agents
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/68133 , vital:29202
- Description: Water hyacinth, Eichhornia crassipes (Martius) Solms-Laubach (Pontederiaceae), is one of the world’s worst alien invasive plants. It is indigenous to the Amazon basin in South America but has become a problematic alien invasive in other parts of the world. As such, several host-specific biological control agents have been sourced from the native distributions in South America and have been released to control this plant where it has become problematic. Two of these agents include the geographically and reproductively isolated cryptic species of Eccritotarsus (Hemiptera: Miridae). One of these species was collected in the upper reaches of the Amazon River in Peru, while the other was collected over 3500km away from that site, in Florianopolis, southern Brazil. These cryptic species were thought to be a single species until recently, when DNA barcoding indicated that they were likely to be two species, and the species status has now been confirmed by interbreeding experiments and detailed morphological studies. The Brazilian population remains Eccritotarsus catarinensis (Carvalho), while the Peruvian population is now known as Eccritotarsus eichhorniae (Henry). The aim of this project was to investigate the mating behaviour and other behavioural traits of the two species that have resulted in reproductive isolation, and which could have led to speciation. In addition, investigations involving analysis of chemical compound compositions of the two species aimed to determine the extent to which the compounds played a role in the development and maintenance of reproductive isolation. To achieve the aims, behavioural-observation experiments were conducted in the form of no-choice, bi-choice and multi-choice tests in 1:1, 2:1 and 3:1 sex ratio assessments, both within and between species. Chemical compound compositions of E. catarinensis and E. eichhorniae were also assessed using Nuclear Magnetic Resonance (NMR), Solid-phase micro-extraction (SPME) and Gas-Chromatography Mass-Spectrometry (GC-MS) techniques. In no-choice experiments, the highest number of single and multiple copula incidences, and average total copula duration was found within species while copulation between species was much rarer. In bi-choice experiments, E. eichhorniae females and E. catarinensis males only chose to mate with their respective conspecifics, and within species copulations continued to have higher average total copula duration. In multi-choice experiments, the highest number of single and multiple copula incidences and average total copula duration was also found within species. GC-MS analysis suggested that E. catarinensis females and E. eichhorniae males have unique chemical compounds missing in their conspecifics and same sex of the other species. Further analysis suggested that E. catarinensis females and E. eichhorniae males have similar chemical compound compositions, whereas as E. eichhorniae females and E. catarinensis males have similar chemical compound compositions. These results suggest that there are behavioural differences that led to the development and maintenance of prezygotic reproductive isolation mechanisms, and that this is probably driven by pheromones in chemical compound compositions. These two species were geographically isolated in the native range and the populations have diverged to the point that they are now reproductively incompatible and therefore, distinct species. The main driver of the speciation is most likely mate recognition and attraction, as only reproductively important traits such as pheromones, genitalia, the scent glands and antennae have changed, while other traits, including host range and morphology, have remained remarkably stable. This provides evidence that differences in sexual selection in isolated populations may be important drivers of speciation and reproductive isolation in cryptic species.
- Full Text:
- Date Issued: 2019
The invasion ecology of Nymphaea mexicana Zucc. (Mexican Water lily) in South Africa
- Authors: Naidu, Prinavin
- Date: 2019
- Subjects: Nymphaea Mexicana zuccarini , Nymphaea , Nymphaea -- Biological control -- South Africa , Water lilies , Nymphaea -- Ecology -- South Africa , Water lilies -- Biological control -- South Africa , Invasive plants -- Biological control -- South Africa , Aquatic weeds -- Biological control -- South Africa , Water lilies -- Ecology -- South Africa
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/92920 , vital:30763
- Description: The Mexican water lily, Nymphaea mexicana Zuccarini, is an aquatic perennial, native to southern USA and Mexico, and has been introduced to South Africa via the ornamental plant trade. This species has rapid growth rates and becomes weedy in dams, ponds and rivers. It is currently listed as a NEM:BA category 1b invasive plant in South Africa. One possible management measure for this weed is biological control, but it is a novel target because no biological control programme has been initiated against it anywhere in the world. This study is intended as a baseline for the biological control programme against this plant in South Africa. Assessing the population structure and mode of reproduction of invasive alien plants is an imperative aid to determining if biological control is a suitable management option. Using amplified fragment length polymorphism (AFLP) molecular markers, I compared the amount of genetic variability and differentiation of N. mexicana in its native range (USA), and invasive range (South Africa). Results indicated a large genetic distance between populations in the USA and South Africa, compared to populations within each country. The genetic variability of the invasive populations was higher than that found in the native distribution. This could be due to hybridization in the introduced range, and/or multiple introductions from different source populations. Differences in the morphology of N. mexicana plants in the invasive range and South Africa were also observed which confirm the results of the genetic analyses. I also assessed the reproductive mode of N. mexicana cultivars/hybrids by conducting breeding system experiments and field pollinator studies. Results indicated that the cultivars are sterile, suggesting that the primary mode of reproduction is asexual via fragmentation of tubers. The main pollinators that were found to be associated with the cultivars in South Africa were honeybees, sweat bees, flies and beetles. These insect groups were the same as those that were observed in another study which was conducted on the pollinators associated with the pure N. mexicana in the native range in southern USA. Mechanical and chemical control of N. mexicana and its multiple genotypes have been applied but have not been efficient due to the fast regeneration of shoots, especially in summer. Therefore, these two management options are not long–term solutions and will also be costly due to the widespread occurrence of the hybrids in South Africa. Thus the only cost–effective, environmentally friendly, self–sustainable and long–term management option is biological control. The significant divergence between native and invasive populations of N. mexicana, as well as the possibility of numerous invasive cultivars, may limit future prospects of biological control of this species. However the differences in the root structures between native South African waterlilies, such as N. lotus and N. nouchali, and the introduced waterlilies, such as N. mexicana and its associated hybrids, may play a pivotal role in the success of biological control of the N. mexicana hybrid complex in South Africa. Natural enemies which feed on the hard tuberous roots of N. mexicana and its hybrids, as opposed to the soft bulbs of the native N. nouchali and N. lotus, should be prioritised.
- Full Text:
- Date Issued: 2019
The performance and preference of a specialist herbivore, Catorhintha schaffneri (Coreidae), on its polytypic host plant, Pereskia aculeata (Cactaceae)
- Authors: Egbon, Ikponmwosa Nathaniel
- Date: 2019
- Subjects: Insects and biological pest control agents -- South Africa , Pereskia -- Biological control -- South Africa , Cactus -- Biological control -- South Africa , Coreida-- South Africa , Invasive plants -- Biological control -- South Africa , Catorhintha schaffneri
- Language: English
- Type: text , Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10962/68250 , vital:29223
- Description: Plant species moved beyond their natural ranges may be liberated into enemy-free spaces, where they increase resource allocation to fitness, rather than defence against natural enemies, and become invasive as suggested by the Evolution of Increased Competitive Ability (EICA) Hypothesis. Several cacti are notable invaders and are targeted for biological control. The leafy cactus, Pereskia aculeata Miller, introduced into South Africa from South America, has become a target for biological control after becoming invasive. The absence of natural enemies of P. aculeata in the introduced range may be the reason for its invasiveness. This thesis seeks to investigate the role of the evolution of increased competitive ability (enemy release) as the probable driver of P. aculeata’s success, and ascertain how the plant’s intraspecific variation influences the impact, fitness of, and preference by its biological control agent, Catorhintha schaffneri Brailovsky and Garcia (Coreidae), in South Africa. Enemy release and evolution of traits in P. aculeata were examined by quantifying plant growth parameters of fifteen genotypes of P. aculeata from both the native and invaded distribution of the plant. Ten genotypes of P. aculeata were used in testing the effect of agent herbivory (impact and damage) under similar conditions. These studies indicated that most invaded-range genotypes were more vigorous than the native genotypes. Rapid growth may account for the quick access of invasive genotypes of P. aculeata to tree canopies. Catorhintha schaffneri damage varied between genotypes but differences in the damage and impact from the agent could not be explained by whether the plant originated in the introduced or native distribution. In sum, while the growth of the invasive genotypes largely conforms to the EICA hypothesis, the impact of C. schaffneri did not support the hypothesis. The influence of host variation in P. aculeata on the fitness of C. schaffneri within the context of local adaptation to plant genotypes from different localities was tested using agent survival, stage-specific and total developmental time, and the extent of damage to ten host genotypes. Maw’s Host Suitability Index (HIS) and Dobie’s Susceptibility Index (DSI) showed the preference by and performance of C. schaffneri on the different genotypes of the plant. Catorhintha schaffneri survived to the adult stage on 70% of genotypes tested. Evidence consistent with the assumption that C. schaffneri would be fitter on the native genotypes than the invasive genotypes due to local adaptation was not found. In addition, there was no evidence in support of fitter agents on the invasive genotypes than on the native genotypes as proposed by EICA hypothesis. Catorhintha schaffneri developed equally well on the invasive genotypes of P. aculeata as on the native genotypes. To establish whether host variation would affect diet selection by C. schaffneri, both nymphs and adults were examined in paired-choice and multiple-choice trials. The nymphs and adults chose their hosts regardless of host genotype differences. The agent may be good at selecting good succulent shoots from bad shoots, but is incapable of distinguishing a good host genotype from a poorer one. This thesis shows, therefore, that P. aculeata and its array of genotypes in South Africa could be effectively controlled by C. schaffneri, as it has the potential to suitably utilise and impact the different genotypes of the weed in South Africa with neither any demonstrable preference nor local adaptation for the native genotypes. Consequently, the use of C. schaffneri, as a biological control agent in the weed biological control programme of P. aculeata remains promising, as the agent is insensitive to the intraspecific variation of the invasive host plants.
- Full Text:
- Date Issued: 2019
Investigating thermal physiology as a tool to improve the release efficacy of insect biological control agents
- Authors: Griffith, Tamzin Camilla
- Date: 2018
- Subjects: Aquatic weeds -- Biological control , Water hyacinth -- Biological control , Insects -- Physiology , Miridae -- Effect of low temperatures on , Cold adaptation , Insects as biological pest control agents , Eccritotarsus catarinensis
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/63301 , vital:28391
- Description: Biological control is commonly used for the control of invasive aquatic weeds, which often involves the release of multiple host-specific agents. Releasing multiple agents has inherent safety concerns as the introduction of each new agent is associated with risks, but is often required to improve control where establishment is limited. Climatic incompatibility between the agent’s thermal physiology and its introduced range often causes agents to fail to establish. However, it has been suggested that the thermal physiology of insects is plastic. Therefore, the potential to manipulate their thermal physiologies before releasing them into the field needs to be explored; reducing the need to release additional agents, thereby ensuring the safety of biological control. This thesis therefore aimed to firstly, determine whether season and locality influenced the thermal physiology of two field populations of a water hyacinth (Eichhornia crassipes) control agent, the mirid Eccritotarsus catarinensis; one collected from the hottest establishment site, and one collected from the coldest establishment site in South Africa. Their thermal physiology was significantly influenced by season and not by the sites’ climate, suggesting their thermal physiology is plastic under field conditions. Secondly, the classical method of determining the lower critical thermal limit (CTmin), and a new respirometry method of determining this limit, compared the thermal physiology of two Eccritotarsus species reared in quarantine. Eccritotarsus catarinensis was significantly more cold tolerant than the more recently released Eccritotarsus eichhorniae, despite similar maintenance conditions, and as such, was used to establish whether cold hardening under laboratory conditions was possible. Successfully cold hardened E. catarinensis had a significantly lower CTmin compared to the field cold acclimated population, suggesting that cold hardening of agents could be conducted before release to improve their cold tolerance and increase their chances of establishment, allowing for further adaptation to colder climates in the field to occur. Increasing establishment of the most effective agents will decrease the number of agents needed in a biological control programme, thus encouraging a more parsimonious approach to biological control.
- Full Text:
- Date Issued: 2018
Potential impact and host range of Pereskiophaga brasiliensis Anderson (Curculionidae): a new candidate biological control agent for the control of Pereskia aculeata Miller (Cactaceae) in South Africa
- Authors: Mdodana, Lumka Anita
- Date: 2018
- Subjects: Curculionidae -- South Africa , Cactus -- South Africa , Biological pest control agents , Alien plants-- South Africa , Pereskiophaga brasiliensis Anderson (Curculionidae) , Pereskia aculeata Miller (Cactaceae)
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/62350 , vital:28157
- Description: Pereskia aculeata Miller (Cactaceae) is a damaging invasive alien plant in South Africa that has negative impacts to indigenous biodiversity and ecosystem functioning. Mechanical and chemical control are not effective against P. aculeata so biological control is considered the only viable option. Two biological control agents, the leaf-feeding beetle Phenrica guerini Bechyne (Chrysomelidae) and the stem-wilting bug Catorhintha schaffneri (Coreidae), have been released in South Africa thus far. Post-release evaluations have indicated that P. guerini will not reduce P. aculeata densities to acceptable levels alone, while C. schaffneri was released very recently, so it is too soon to determine how effective that agent will be. Even if C. schaffneri is extremely damaging, it is likely that further agents will be required to reduce the densities of P. aculeata to acceptable levels within a reasonable time-scale. Additional agents should target the woody stems of P. aculeata which are not impacted by the damage of either of the released agents. Pereskiophaga brasiliensis Anderson (Curculionidae) is a promising potential candidate agent that feeds on the thick woody stems of the plant in the larval stage. Climatic matching, genetic matching and field based host specificity observations all indicated that P. brasiliensis was a promising candidate. In this study, the impact of P. brasiliensis to the target weed, P. aculeata, was quantified under quarantine conditions to determine whether it was sufficiently damaging to warrant release. This was followed by host specificity testing to determine whether P. brasiliensis was suitably host specific for release in South Africa. Impact studies indicated that P. brasiliensis was damaging to P. aculeata at insect densities that would be expected in the field. Pereskiophaga brasiliensis reduced the number of leaves of P. aculeata to a greater extent than it reduced shoot lengths, but both plant parameters were significantly reduced due to the feeding damage from the insect. This suggests that the damage from P. brasiliensis may be compatible with that of C. schaffneri which reduces shoot length to a greater degree than the number of leaves. Pereskiophaga brasiliensis is therefore sufficiently damaging to warrant release, and although interaction studies with the other agents would be required, it is expected that it should complement other existing agents. Although P. brasiliensis is sufficiently damaging, at present the host specificity data indicates that it is not suitably specific for release in South Africa because oviposition and larval development to the adult stage was recorded on both indigenous and alien plant species within the families Cactaceae and Basellaceae. This non-target feeding was recorded during no-choice tests, which are very conservative, but significant non-target damage and development to the adult stage was recorded on an indigenous plant from a different family to the target weed. Further host specificity testing, including paired and multiple choice tests, are required to confirm the broad host range of P. brasiliensis. Other biological control agents that damage the woody stems of P. aculeata should be considered. The stem-borer, Acanthodoxus machacalis (Cerambycidae) is considered the most promising of the other candidate agents as it can be sourced from a climatically matched region where genetically suitable P. aculeata plants are found, it is sufficiently damaging to the woody stems of P. aculeata and there is no evidence that the species has a broad host range. Acanthodoxus machacalis should be sourced from Rio de Janeiro, Brazil, and imported into quarantine in South Africa for host specificity testing.
- Full Text:
- Date Issued: 2018
Interactions between two biological control agents released on Pereskia aculeata Miller (Cactaceae), in South Africa
- Authors: Mnqeta, Zezethu
- Date: 2017
- Subjects: Pereskia aculeata -- Biological control , Cactus -- Biological control -- South Africa , Alien plants -- Biological control -- South Africa , Flea beetles -- South Africa , Coreidae -- South Africa , Insects as biological pest control agents -- South Africa
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/7949 , vital:21327
- Description: Pereskia aculeata Miller (Cactaceae) is an alien invasive plant introduced into South Africa from Brazil, which has negative impacts on native biodiversity in South Africa. Phenrica guerini Bechyne (Chrysomelidae) and Catorhintha schaffneri Brailovsky & Garcia (Coreidae) are two biological control agents released against P. aculeata in South Africa. Phenrica guerini was first released against P. aculeata, in 1991, followed by C. schaffneri in 2014. The overall aim of this study was to improve the biological control programme against P. aculeata in order to decrease its density to a level where it does not threaten the native biodiversity of South Africa. The first part of this study evaluated the efficacy of P. guerini on P. aculeata at Port Alfred (Eastern Cape) to better understand the role of P. guerini in the biological control of P. aculeata. An insecticide exclusion experiment was conducted over 100 days. Plots with P. guerini had a mean of 187 (SE ± 62) fewer leaves/m2 than plots without P. guerini. The agent reduced percentage cover in plots with P. guerini, with a mean of 19.42% (SE ± 3.15) lower cover than plots without P. guerini. Although P. guerini had an impact on P. aculeata at Port Alfred previous studies have indicated that a reduction to below 50% cover is required for native biodiversity to recover and the agent only reduced cover to 62% at Port Alfred. Phenrica guerini has therefore not reduced percentage cover sufficiently to completely control the weed. The data collected from Port Alfred was compared to the performance of the agent nationwide. Although P. guerini was found at far more sites than previously recorded, there were very few sites with comparable levels of damage to Port Alfred. This evidence suggests that P. guerini is not sufficiently damaging to reduce P. aculeata to acceptable levels and other biological control agents should be considered.Interactions between two biological control agents can have complex and unexpected impacts for a biological control programme. The second part of this study was to investigate interactions between C. schaffneri and P. guerini under laboratory conditions to test whether the two agents, individually or jointly, enhanced or reduced their impact on P. aculeata. Potted P. aculeata plants were exposed to one of four treatments: control (no agents), P. guerini only, C. schaffneri only and both species in combination. Four stocking densities, ranging from 2 to 12 insects per plant were used. Catorhintha schaffneri alone at high densities was more damaging than all other treatments with a significantly greater reduction in the mean number of leaves, 11.7 (SE ± 1.29), and shoot lengths, 2.17cm (SE ± O. 75). Even at lower density treatments, the combination of the two agents was not significantly more damaging than C. schaffneri alone and C. schaffneri was always more damaging than P. guerini alone. Mortality of P. guerini was significantly higher than C. schaffneri at the highest stocking density when in combination. Phenrica guerini contributes towards the biological control of P. aculeata at some sites in South Africa but not enough to completely control the weed. The antagonistic interaction between P. guerini and C. schaffneri suggests that these agents should not be released together because this would impact negatively on the overall biocontrol programme against P. aculeata. Catorhintha schaffneri should be released at sites were P. guerini is not present and evaluations of the success of this agent in the field should be conducted. Extrapolation of laboratory-based studies into the field is often challenging so mass-rearing of P. guerini should continue until there is convincing proof that C. schaffneri alone is more effective than P. guerini in the field.
- Full Text:
- Date Issued: 2017
Investigating herbivory and plant origin on tall-statured grasses in South Africa
- Authors: Canavan, Kim N
- Date: 2017
- Subjects: Insects as biological pest control agents -- South Africa , Arundo donax , Giant reed -- South Africa , Giant reed -- Biological control -- South Africa , Phragmites australis , Phragmites mauritianus , Phragmites , Tetramesa romana , Biological invasions -- South Africa , Wasps -- Host plants , Wasps -- South Africa
- Language: English
- Type: text , Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10962/6147 , vital:21051
- Description: South African riparian zones have been heavily degraded through anthropogenic activities such as dam construction and extraction of water for irrigation, which has resulted in a loss of ecosystem services and functioning. A consequence of such disturbances to riparian areas is in their susceptibility to invasive alien species (IAS). One such IAS is the giant reed, Arundo donax L. (Poaceae), introduced to South Africa in the 1700s largely for erosion control. Arundo donax has since greatly expanded in the country and is now one of the most abundant IAS. Arundo donax has been found to displace native vegetation and in South Africa this will most likely lead to the displacement of the native tall-statured grasses, Phragmites australis (Cav.) Trin. ex Steud. and Phragmites mauritianus Kunth. This study aimed to enhance our understanding of the tall-statured grasses A. donax, P. australis and P. mauritianus to better manage them in riparian areas. For A. donax, biological control is seen as the most viable option to control stands in the long-term. However, before such a programme is put in place, it is important to first collect baseline data that can be used to guide the direction of the biological control project in South Africa. For the Phragmites spp., despite being a dominant vegetative type in riparian areas, very little is known about their status in South Africa. Furthermore, there have been increasing reports of both Phragmites species having an expansion of their range and abundance. In North America, there has been a similar trend of reed expansion and through molecular work it was determined that a cryptic invasion has occurred with the introduction of an invasive non-native haplotype from Europe. It is therefore unknown if Phragmites spp. populations are expanding due to anthropogenic activities or due to a cryptic invasion. To address these shortfalls in knowledge the study investigated the tall-statured grasses in two parts; firstly, molecular techniques are used to explore the plant origin and genetic diversity of A. donax, P. australis and P. mauritianus and secondly using the Enemy Release Hypothesis as a framework, herbivore assemblages for each reed was determined across their distribution in South Africa. Molecular-techniques determined that both P. australis and P. mauritianus had only one haplotype - known as haplotype K and haplotype V respectively, across their distribution. For P. australis, haplotype K shares a close connection with populations from a Mediterranean lineage and this was further confirmed with a shared grass-waxy band. The direction and timing of genetic exchange between the two regions could not be ascertained and thus still remains unknown. Microsatellite analysis determined that both Phragmites spp. had a high genetic diversity compared to worldwide lineages. With no evidence of any cryptic invasions of haplotypes from other regions, both Phragmites spp. populations are likely to be native to South Africa. For A. donax all populations across South Africa were determined to be haplotype M1; a cosmopolitan haplotype that has an ancient native range in Afghanistan and Pakistan (Indus Valley). Populations were found to have no genetic diversity and thus can be considered one clone. A pre-introductory survey determined a list of herbivores associated with each tall- statured grass. For A. donax, a total of seven herbivores were found. Of these, one herbivore, a galling wasp, Tetramesa romana Walker (Hymenoptera: Eurytomidae) was found to be highly abundant and widely distributed in South Africa. Tetramesa romana is already a biological control agent in North America and thus is likely exerting some pressure on A. donax populations in South Africa. For both Phragmites spp. a total of ten herbivores were found, although having higher species richness compared to A. donax, when compared to other regions, these native species have a relatively low species richness. Providing baseline data on plant origin, genetic diversity and herbivory on A. donax, P. australis and P. mauritianus has provided important information on managing these species in riparian ecosystems in South Africa. For the Phragmites spp. with no evidence of any cryptic invasions, it is recommended that reed stands continue to be managed as native species. Phragmites spp. are important dominant vegetative species and thus should be protected; however, if reed stands become expansive, control methods can be put in place to focus on managing spread and abundance. For A. donax, this study was able to provide pivotal information in guiding the biological control programme. By determining the ancient lineage of South African populations, research can be focused in this area to find potential biological control agents. Lastly, the pre-introductory survey determined that a biological control agent, T. romana was already established with an unknown introduction and also highlighted potential plant parts that should be targeted. In particular, no rhizome feeding herbivores were found in South Africa and therefore this highlights an important niche that should be explored in biological control agents.
- Full Text:
- Date Issued: 2017
Post-release evaluation of the biological control programme against Cereus jamacaru De Candolle (Cactaceae), in South Africa
- Authors: Sutton, Guy Frederick
- Date: 2017
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/7982 , vital:21330
- Description: Cereus jamacaru De Candolle (Cactaceae) is an environment-transforming weed of both agricultural and wildlife habitats in South Africa. Weed infestations reduce carrying capacity of the land, and pose a risk to livestock, wildlife and human safety due to the plant’s spiny thorns. The weed is considered to be under complete biological control in South Africa, due to its introduced control agent, Hypogeococcus festerianus Lizer y Trelles (Hemiptera: Pseudococcidae), although observational reports suggest that the level of success achieved is variable. In this thesis, a formal post-release evaluation of this biological control programme was conducted, specifically to determine the efficacy of H. festerianus as a biological control agent, and to identify factors which may limit or constrain the level of success achieved by the control agent. These data were collected with the intention of improving the control of C. jamacaru in South Africa. A field-based study of C. jamacaru population demographics investigated the efficacy of H. festerianus as a biological control agent of the weed, by integrating weed growth, fecundity and survival metrics with C. jamacaru population dynamics and demographic patterns from 8 sites where H. festerianus was present and 14 sites where the control agent was absent. The findings indicated that H. festerianus significantly reduced weed fecundity, which resulted in fewer seedling recruits, and that levels of plant mortality were greater at sites where H. festerianus was present. The reduction in weed fecundity and survival translated into negative population-level consequences for H. festerianus. Weed-population age frequency distributions in the absence of H. festerianus demonstrated a “reverse J-shaped” distribution, indicative of high recruitment rates and population stability, while C. jamacaru populations infected with H. festerianus were described by bell-shaped distributions, and were typified by limited recruitment, or a complete lack thereof. By constraining recruitment and inhibiting selfregeneration, H. festerianus appears to regulate populations of C. jamacaru. Predation and parasitism of H. festerianus was believed to be a limiting factor for the biological control programme against C. jamacaru in South Africa, although no formal evaluation of this claim had been undertaken. Accordingly, the assemblage of natural enemies acquired by H. festerianus in South Africa was identified by field-collections of infected H. festerianus gall-material. Further, timed point-count surveys of natural enemies associated with H. festerianus were performed and integrated with the data on the impact of H. festerianus on weed population dynamics to assess the impact of two prominent predaceous taxa on H. festerianus efficacy as a biological control agent. Although H. festerianus had acquired a diverse suite of novel natural enemies in South Africa, this has not prevented the biocontrol agent from having an impact on C. jamacaru populations, although other subtler effects cannot be ruled out. This study showed that biological control efforts employing H. festerianus for the management of C. jamacaru have been successful. Furthermore, these data demonstrated the utility of retrospective analyses in developing and improving the science of biological control, specifically how to improve candidate agent prioritisation, determining how many agents are required for successful biological control, and how to evaluate the success of biological control efforts. Improvements in our theoretical understanding of biological control will undoubtedly reduce costs of biological control programmes, improve success rates, and increase the predictability of biological control.
- Full Text:
- Date Issued: 2017