Mayr, Linsley and Usinger (1953) have defined three levels of taxonomic study; the first or alpha level dealing with the recognition and description of new species, the second or beta level dealing with the arrangements of these described species into the hierarchical system and the third or gamma level dealing with the study of infraspecific variation and the evolution of taxa. It is probably the general rule that, in any one group, progress is made by passing through these various levels in turn as they suggest in the following quotation.

Gryllus bimaculatus de Geer is one of the most widely distributed species of field crickets and ranges from South Africa, the Mediterranean basin, Southern Asia and the East Indies to Japan. (Cousin, 1956, 1961). Chopard (1938) found that this species does not range north of the northern shore of the Mediterranean. Cousin (1941) was the first to recognize the significance of a regular phenomenon in the variation of pigmentation of G. bimaculatus specimens reared in the laboratory from collections made in the field in southern France.

The rice water weevil (RWW), Lissorhoptrus oryzophilus (Kuschel), is the most important insect pest of rice in the United States. Integrated pest management strategies for RWW in Louisiana consist of cultural controls, resistant cultivars and chemical insecticides. The fourth component of IPM, biological control, is largely absent from the literature for RWW, making exploration of biological control a logical next step in developing a full set of IPM strategies. The three main types of biological control are augmentation, classical and conservation. With little known about RWW predators, conservation biological control makes the most sense as local natural enemy abundance is increased. Detrital subsidies have been shown to cause trophic cascades in agricultural system that can ultimately reduce herbivore populations via increased predator abundance. During the summer of 2013 and 2014, field experiments were carried out to determine if compost-manure additions to rice fields would cause an increase in invertebrate diversity and translates to a reduction in RWW numbers. Surveying of treatment (compost-manure additions) and control plots (no additions) for differences in invertebrates used four different sampling methods: root/soil corer, Gee crayfish trap, aquatic netting and floating pitfall traps. Based on sampling from this experiment, Notonecta sp., immature Pantanla sp. and T. lateralis were chosen to be used in aquaria experiments to test for predator effects on RWW. Detrital subsidies in both years failed to increase diversity of invertebrates or reduce RWW numbers. Plotting of feeding guilds over the course of both years showed predator populations paralleling prey populations. RWW oviposition and larval emergence was unaltered whether Notonecta sp. or Pantala sp. were present or not. Tropisternus lateralis, a herbivore/scavenger, also failed to alter RWW fecundity, suggesting that RWW may either not alter their behavior when other organisms are present or they may be able to differentiate between dangerous and non-dangerous arthropods. These experiments failed to cause a trophic cascade or identify predators of RWW. They did however demonstrate that a large prey population is present and that is utilized by a diverse predator assemblage still with the potential to be increased by detrital subsidies.

Plants are attacked by a variety of herbivore feeding guilds and respond with specific responses to specific attacks, which may be localized or systemic. How a plant defends against one feeding guild may alter the plants response against a different feeding guild. A better understanding of these interactions will allow for the development of refined pest management programs. One situation in which this may occur is in interactions between chewing and piercing/sucking herbivores, such as aphids. Aphids are important crop pests primarily due to their ability to transmit viruses, the efficacy of which can be affected by plant defenses. To determine if systemic induction has an effect on aphid feeding behaviors, three soybean varieties, (Progeny 4906RR, Davis, and Lyon) were induced with either soybean looper (SBL), Chrysodeixis includes (Walker), larvae, foliar jasmonic acid, or foliar salicylic acid. Three days post induction, green peach aphid (GPA), Myzus persicae (Sulzer), apterae feeding behavior was recorded on induced and control plants using the Electrical Penetration Graph (EPG) technique. SBL feeding bioassays were used to assess the effect of previous SBL herbivory, JA or SA on SBL larval weight, and to confirm systemic induction. Both previous herbivory and JA reduced SBL larval weights when fed Progeny 4906RR tissue, while herbivory had no effect on SBL larval weights when fed Davis tissue, suggesting SBL does not induce a defensive response in Davis. SA did not induce a response in either variety. Neither herbivory nor JA had an effect on SBL larval weight in Lyon, suggesting it does not induce a defensive response. However, SA increased SBL larval weights. SBL herbivory decreased several behaviors associated with nonpersistant virus transmission in both Progeny and Davis, but had no effect on aphid feeding behavior in Lyon. Jasmonic acid induction increased several behaviors associated with nonpersistant virus transmission in both Progeny and Davis. Exogenous SA application also increased behaviors associated with nonpersistant virus transmission in all three varieties. These results suggest that inducing host plant resistance using exogenous JA application may have positive effects by reducing herbivore performance. However, both JA and SA may potentially have negative impacts by increasing nonpersistant virus transmission.

Three field experiments were conducted in a mango orchard in Croix-des-bouquets (Haïti) to develop an effective artisanal McPhail trap, less expensive than the commercial traps, for mass trapping Anastrepha obliqua (Macquart) and Anastrepha suspensa (Loew). A field trial conducted to evaluate the effectiveness of two newly made artisanal trap models (AT1 and AT2) with the commercial McPhail trap (MP) demonstrated that the artisanal traps yielded similar results in the average number of fruit flies caught (8.9±2.6, 13±2.9, and 16±4.1 respectively). Moreover, the cost-efficacy ratio was a lot higher in the artisanal trap models (AT1: 0.42 $ per flies caught, AT2: 0.28 $ per flies, and MP: 0.69 $ per flies), even if the total number of fruit flies was higher in the commercial trap (319 flies) compared to the others (AT1: 178 flies and AT2: 253 flies). Another field trial conducted in the same mango orchard compared a density of 24 McPhail traps per ha to 36 traps/ha using the most cost-effective artisanal trap, and revealed that they were not different in number of fruit flies caught (AT2: 236 flies and MP: 239 flies). In addition, the capture rate of Anastrepha spp. in both trap densities had a similar increasing trend line throughout the mango fruiting season. To determine an optimal trap density for the artisanal trap (AT2) under mass trapping conditions, a field experiment assessed six different trap densities (4, 8, 12, 16, 24, and 36 traps/ha), and suggested that a density of 25 traps/ha could protect the mango orchard from the growing phase to the maturation phase of mango fruits. However, analysis of fruit fly data available throughout the year suggested that trapping density should be increased during the ripening phase, when the Anastrepha spp. density reach their peak in this orchard. These findings indicated that cost-effective artisanal trap models can be developed to substitute the expensive commercial traps for implementation of fruit fly control programs with mass trapping methods.

The fall armyworm (FAW), Spodoptera frugiperda, and corn earworm (CEW), Helicoverpa zea, are two target pests of pyramided Bt corn in the U.S. This study determined the susceptibility of the two pests to pyramided Bt corn and evaluated if a 95:5% seed mixture of Bt and non-Bt corn was an appropriate approach for providing refuge populations of CEW for resistance management. Firstly, susceptibility of 150 F2 two-parent families to three common pyramided Bt corn traits was examined using 7-day leaf tissue bioassays and whole plant tests. A few families survived the 7-day leaf tissue bioassays but progeny of the survivors from the leaf tissue bioassays could not survive in the whole plant tests, suggesting that the pyramided Bt corn products were effective against FAW. Secondly, occurrence and damage of CEW in three planting patterns of non-Bt and Bt corn plants containing SmartStaxTM and VipteraTM 3111 traits were evaluated in the fields. The studies showed that both Bt corn products were highly effective against CEW in both pure stand and mixed plantings. Relative to pure non-Bt corn plantings, larval occurrence at the early stages (3-4 instars) in a mixed planting of 96% Bt and 4% of non-Bt corn were similar, but the larval development was delayed. Finally, intensity of Bt protein contamination and its associated impacts on CEW populations in a mixed planting of 95% Bt and 5% non-Bt corn were assessed with the SmartStaxTM traits. The results of field and laboratory studies showed that cross-pollinations in the mixed planting caused majority (> 90%) of the refuge kernels to express ≥ one Bt protein, and the intensive Bt protein contamination in the refuge ears reduced neonate-to-adult survivorship of CEW to only 4.6%, a reduction of 88.1% relative to the larvae feeding on ears of pure non-Bt corn plantings. The results suggest that the 95:5% seed mixed approach cannot provide adequate refuge populations for CEW. Data generated from this study should provide useful information for developing appropriate resistance management strategies for the sustainable use of the Bt corn technology as a pest management tool.

The fall armyworm (Spodoptera frugiperda (J. E. Smith)) is a major pest targeted by transgenic corn expressing Bacillus thuringiensis (Bt) proteins in both North and South America. The objective of this study was to characterize the inheritance and fitness costs of the Cry1F resistance in two fall armyworm populations collected from Puerto Rico (RR-PR) and Florida (RR-FL). To determine the inheritance of the resistance, besides RR-PR, RR-FL, and a Cry1F-susceptible population (Bt-SS), 14 other populations were developed by reciprocal crosses, F1 by F1 crosses, backcrosses, and crosses between RR-PR and RR-FL. Diet-incorporated bioassays were conducted to determine the Cry1F susceptibility for all 17 populations. To assess the fitness costs of the resistance, seven insect populations were assayed on a non-toxic diet as well as on a combined rearing of non-Bt corn leaf tissue and non-Bt diet. The seven populations were RR-PR, RR-FL, Bt-SS, and four F1 populations that were developed from the reciprocal crosses between Bt-SS and the two resistant populations. Biological parameters measured in the fitness tests were neonate-to-adult survivorship, neonate-to-adult development time, 10-day larval mass on non-Bt corn leaf tissue, pupal mass, and sex ratios. The results showed that there might be a different genetic basis for the Cry1F resistance between the Puerto Rico and Florida populations. The Cry1F resistance in RR-PR was likely inherited in >1 recessive or incompletely recessive genes and the genes associated the resistance were sex-linked to the males of the insect. In contrast, the resistance in RR-FL was dominant and more likely controlled by autosomal genes. Cry1F resistance in both resistant populations was associated with considerable fitness costs, especially for the Florida population. The fitness costs in the Cry1F-resistant fall armyworm were revealed in reduced growth, increased mortality, and delayed development. Data generated from this study should be valuable in understanding the mechanisms of Cry1F resistance in fall armyworm and developing effective strategies for resistance management.

My thesis research includes three components: (1) a methodology project, (2) a systematic project comprising a taxonomic revision, and (3) a morphology-based, cladistic analysis. 1. The methodological study compared the Winkler (Moczarski) eclector and the Berlese-Tullgren funnel (referred to hereafter as Winkler and Berlese funnels, respectively) as methods for extracting forest litter-dwelling Coleoptera from sifted substrate, with a focus on relevance for taxonomic studies. A general review of the history and methodologies of leaf litter arthropod extraction was provided, in conjunction with repeated, timed trials to compare the relative effectiveness of the two methods using litter samples from various geographic regions. Extraction rates for Winkler and Berlese funnels were evaluated based on temporal and taxonomic criteria by comparing accumulations of Coleoptera specimens and species at timed intervals. In addition, the Curculionidae and Staphylinidae, as well as rare species were targeted and evaluated, individually. Results did not indicate clear differences in extraction efficiency, either temporally or taxonomically, between Winklers and Berleses. However, extraction patterns for both Winklers and Berleses suggested that intervals at the beginning of trials (3-18 hours) and later in the trials (60-144 hours) yielded information most relevant for taxonomic studies. An extraction protocol for litter-dwelling Coleoptera is suggested based on these results. A checklist of all species collected was provided. 2. A taxonomic revision of the North American trichonychine genus Eutyphlus LeConte reviewed the current status of described species. These were redescribed along with documentation and drawings of external variations, male genitalia, biometrics, and range maps for each species. One undescribed species was described from museum specimens based on morphological and phylogenetic criteria. 3. Monophyly of the genus Eutyphlus was tested and the phylogenetic placement of this genus within the subtribe Panaphantina Jeannel was evaluated based on morphological cladistic analysis of trichonychine genera, including numerous relevant outgroup taxa. Eutyphlus was recovered as a monophyletic genus within the tribe Trichonychini Reitter, while many other higher-level taxa were not grouped together and merit redefinition or further examination.

The Formosan subterranean termite, Coptotermes formosanus Shiraki, is a worldwide distributed pest of wooden structures and living plants that causes huge economic losses. Compared to chemical pesticides, biological control may provide a more environmentally friendly and persistent method for the control of C. formosanus. In this research, a series of studies were conducted to understand the termite-pathogen interaction and to develop a feasible biological control strategy. In the first part of the research, the toxicity of Bt toxins expressed by genetically modified maize to termites was tested. Plant tissues or extracts of three commercially planted Bt maize and two non-Bt maize were provided to termites as food. The results revealed no significant difference in survival rate, food consumption or length of tunnels among termites feeding on Bt and non-Bt maize. The following experiments show that maize cob can be used as a termite bait matrix. In the second part of the research, the susceptibility of C. formosanus to MosquitoDunks®, which contains about 10% of the entomopathogenic bacterium, Bacillus thuringiensis subspecies israelensis, was tested. No-choice and choice bioassays did not show a promising lethal effect of MosquitoDunks® on termites. Furthermore it was shown that C. formosanus can suppress the growth of B. thuringiensis. Also, clay was tested for its potential to be a termite bait matrix that can be used to encapsulate biological control agents. Choice tests showed that significantly more termites aggregated in chambers where clay was provided, indicating the possibility of clay to be used as a termite attractant. In the third part of the research, the potential to combine a biological control agent and a chemical pesticide against termites was investigated. The effect of low concentrations of lufenuron, a chitin synthesis inhibitor, on termite physiology and behavior was tested. Results showed that lufenuron significantly reduced vigor and disease resistance of termites. In the following experiments, termite mortality was significantly higher and synergistic in the combination of lufenuron and Pseudomonas aeruginosa (Schroeter) compared to treatment of lufenuron or P. aeruginosa alone. To combine lufenuron and a termite pathogen may bring a successful IPM strategy for the control of termites.

For many years the study of insect pathogens was not regarded as one of the duties of the entomologist. When insects became numerous and caused economic damage, they were combatted either with insecticides or by the introduction of parasitic or predaceous insects. The use of insecticides in the forest is not economical. Forest entomologists therefore resorted largely to the introduction and propagation of parasitic insects. Unfortunately, experience has shown, in many cases, that introduced parasites multiply too slowly to check current insect infestations. Insect pathogens multiply rapidly and at times contribute more to the control of insect pests than all other combined control factors. The control of the European spruce sawfly by disease created new interest in the possible use of insect pathogens in biological control. Relatively few insect diseases have been discovered. Those which are now known are the obvious cases. A study of any insect population probably would reveal one or more pathogens affecting it. This is being demonstrated as the result of the newly created interest and research in insect pathology. The discovery of an organism pathogenic to insects does not necessarily imply that it can be propagated and immediately disseminated in nature and effect control. This has been attempted too often in the past with disappointing results. Any program of artificial dissemination must be based on a thorough knowledge of the nature of the organism [...]