Cotton (Gossypium hirsutum L.) crop maturity is delayed by cotton fleahopper (Pseudatomoscelis seriatus Reuter) (fleahopper) feeding on early-season fruit forms which increases vulnerability to late-season pests such as Helicoverpa zea (Boddie) and Heliothis virescens (Fabricius). The objectives of this research were to evaluate methods of screening for resistance to fleahopper and to screen selected genotypes. Six fleahoppers were caged on plants in the insectary for 72 h. Numbers of live fleahoppers and percent square damage were determined 48 h following the removal of fleahoppers. Fleahopper numbers and percent square set were determined on randomly selected plants of 16 genotypes when grown under field conditions in 2002 and 2003. Across multiple sampling dates, the number of fleahoppers per plant was higher (p=0.05) in G. arboreum and Pilose (G. hirsutum), but no consistent differences were observed among the remaining 15 genotypes which represented several germplasm pools across the United States. Field and no-choice feeding tests suggested that Pilose, Lankart 142, Suregrow 747, and Stoneville 474 were more resistant hairy-leaf genotypes and not different (p=0.05) in resistance than the smooth-leaf genotypes, Deltapine 50 and TAM 96WD-69s. Pin-head, match-head, and one-third grown squares were removed from plants and placed on agar in petri-plates. Four fleahoppers were released per plate and allowed to feed for 48 h. Fleahopper damage, brown areas along the anthers and/or brown and shrunken pollen sacs was most evident in pin-head sized squares.

Cotton (Gossypium hirsutum L.) crop maturity is delayed by cotton fleahopper (Pseudatomoscelis seriatus Reuter) (fleahopper) feeding on early-season fruit forms which increases vulnerability to late-season pests such as Helicoverpa zea (Boddie) and Heliothis virescens (Fabricius). The objectives of this research were to evaluate methods of screening for resistance to fleahopper and to screen selected genotypes. Six fleahoppers were caged on plants in the insectary for 72 h. Numbers of live fleahoppers and percent square damage were determined 48 h following the removal of fleahoppers. Fleahopper numbers and percent square set were determined on randomly selected plants of 16 genotypes when grown under field conditions in 2002 and 2003. Across multiple sampling dates, the number of fleahoppers per plant was higher (p=0.05) in G. arboreum and Pilose (G. hirsutum), but no consistent differences were observed among the remaining 15 genotypes which represented several germplasm pools across the United States. Field and no-choice feeding tests suggested that Pilose, Lankart 142, Suregrow 747, and Stoneville 474 were more resistant hairy-leaf genotypes and not different (p=0.05) in resistance than the smooth-leaf genotypes, Deltapine 50 and TAM 96WD-69s. Pin-head, match-head, and one-third grown squares were removed from plants and placed on agar in petri-plates. Four fleahoppers were released per plate and allowed to feed for 48 h. Fleahopper damage, brown areas along the anthers and/or brown and shrunken pollen sacs was most evident in pin-head sized squares.

The cotton fleahopper, Pseudatomoscelis seriatus (Reuter) is a widely distributed insect across the United States. Although, it feeds on several native wild hosts, its agricultural importance lies as an economic pest of cotton in several states in the southern United States. No studies have addressed intraspecific genetic and phenotypic variation of this insect pest at a large geographic scale. I examined genetic variation among cotton fleahopper populations associated with cotton in different geographic locations across the southern United States (Chapter II). Using dominant, neutral, nuclear molecular markers (AFLP, amplified fragment length polymorphism) and mitochondrial DNA sequences, I found that overall genetic differentiation among different geographic populations, collected from cotton in eleven cotton growing states, was low but significant. AFLP revealed the presence of three regional groups representing western (Arizona), central (Texas, Oklahoma, Arkansas, Louisiana, Mississippi and Alabama), and eastern (Florida, Georgia, South Carolina and North Carolina) populations. I examined if there were distinct lineages of cotton fleahoppers associated with three of its host plant species: cotton (Gossypium hirsutum), horsemint (Monarda punctata) and woolly croton (Croton capitatus) in five different locations of Texas by using AFLP markers (Chapter III). I found two distinct host-associated lineages at three locations and local panmixia in the other two locations. I tested if host preference of cotton fleahoppers were affected by geographic variation and prior experience. Conducting choice tests with a Y-tube olfactometer, I found that host preference in cotton fleahoppers for horsemint (one of its native host plants) is conserved and unaffected by individual?s prior experience with cotton (Chapter IV). Finally, I explored the role of host-plant species in morphological differentiation of the cotton fleahopper in two locations that differ in presence of distinct host-associated lineages. Using a geometric-morphometric approach, I detected significant effect of host plant and geography on body morphology and wing shape of cotton fleahopper populations (Chapter V). Length of antenna and rostrum were two important traits associated with morphological divergence of cotton and horsemint associated insect populations. Cotton associated individuals had relatively longer antenna and rostrum compared to individuals associated with horsemint.