THE BERMUDAGRASS STEM MAGGOT: AN EXOTIC PEST IN THE SOUTHEASTERN UNITED STATES LISA LEANNE BAXTER. (Under the Direction of Dennis W.

Transcription

1 THE BERMUDAGRASS STEM MAGGOT: AN EXOTIC PEST IN THE SOUTHEASTERN UNITED STATES by LISA LEANNE BAXTER (Under the Direction of Dennis W. Hancock) ABSTRACT The bermudagrass stem maggot (BSM; Atherigona reversura Villeneuve) has infested and damaged bermudagrass [Cynodon dactylon (L.) Pers.] hayfields throughout the southeastern United States. Eight Cynodon cultivars were tested and it was determined that the number and percent of tillers damaged depends on cultivar, but an average 7.7% decrease in total dry biomass was observed for all cultivars. Unfortunately, there is no information available on the reproductive potential of the BSM. Adult flies dissected from various regions in Georgia and Florida demonstrated that the reproductive morphology resembled A. soccata and the total number of ovarioles varied with region. To provide information on morphological differences among Cynodon cultivars, eight characteristics were compared, and it was concluded that C. dactylon have denser, lighter green, finer-textured canopies while C. nlemfuensis Vanderyst had less leaf density, were darker green, and more robust. Cultivar tolerance appears to be the best IPM strategy to implement for BSM control. INDEX WORDS: Bermudagrass, Cynodon, Bermudagrass Stem Maggot, BSM, Atherigona reversura, cultivar differences, ovariole number

2 THE BERMUDAGRASS STEM MAGGOT: AN EXOTIC PEST IN THE SOUTHEASTERN UNITED STATES by LISA LEANNE BAXTER BS, Berea College, 2012 A Thesis Submitted to the Graduate Faculty of The University of Georgia in Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE ATHENS, GEORGIA 2014

3 2014 Lisa Leanne Baxter All Rights Reserved

4 THE BERMUDAGRASS STEM MAGGOT: AN EXOTIC PEST IN THE SOUTHEASTERN UNITED STATES by LISA LEANNE BAXTER Major Professor: Committee: Dennis W. Hancock William G. Hudson William F. Anderson Brian M. Schwartz Electronic Version Approved: Maureen Grasso Dean of the Graduate School The University of Georgia May 2014

5 iv DEDICATION I dedicate this thesis to my advisors and professors who have guided me through not only my education but also many of life s important choices. Also, this thesis would not have been possible without the unconditionally support of my closest friends.

6 v ACKNOWLEDGEMENTS I would like to acknowledge my committee and collaborators on this project: Dr. Dennis Hancock (UGA-Athens), Dr. Will Hudson (UGA-Athens), Dr. Brian Schwartz (UGA-Tifton), Dr. Bill Anderson (USDA-ARS Coastal Plains Experiment Station), and Dr. Patricia Moore (UGA-Athens). I would also like to thank all of the producers who invited me to sample for the BSM in their pastures and hayfields and the Cooperative Extension Agents who have coordinated field visits: Tammy Cheely (Warren County, GA), Phillip Edwards (Irwin County, GA), Keith Fielder (Putnam County, GA), Raymond Joyce (Laurens County, GA), Peyton Sapp (Burke County, GA), Elena Toro (Suwannee County, FL), and Tim Varnedore (Jeff Davis County, GA).

7 vi TABLE OF CONTENTS Page ACKNOWLEDGEMENTS...v LIST OF TABLES... viii LIST OF FIGURES...x CHAPTER 1 INTRODUCTION...1 REFERENCES LITERATURE REVIEW...9 REFERENCES THE RESPONSE OF SELECTED BERMUDAGRASS CULTIVARS TO BERMUDAGRASS STEM MAGGOT DAMAGE...25 ABSTRACT...26 INTRODUCTION...26 MATERIALS AND METHODS...28 RESULTS AND DISCUSSION...33 CONCLUSIONS...37 REFERENCES EXPLORATION OF THE REPRODUCTIVE MORPHOLOGY AND POTENTIAL OF THE BERMUDAGRASS STEM MAGGOT...48 ABSTRACT...49

8 vii INTRODUCTION...49 MATERIALS AND METHODS...52 RESULTS AND DISCUSSION...53 CONCLUSIONS...55 REFERENCES MORPHOLOGICAL CHARACTERISTICS OF SELECT CYNODON CULTIVARS...64 ABSTRACT...65 INTRODUCTION...65 MATERIALS AND METHODS...68 RESULTS AND DISCUSSION...70 CONCLUSIONS...72 REFERENCES CONCLUSIONS, IMPLICATIONS, AND FUTURE RESEARCH DIRECTIONS.79 APPENDIX A SELECT TABLES...83

9 viii LIST OF TABLES Page Table 2.1: Summary of plant characteristics that reduce Atherigona damage in other field crops Table 2.2: Summary of plant compounds found to attract or deter A. soccata...24 Table 3.1: Location of field sampling sites...43 Table 3.2: Analysis of variance for the effects and interactions of year, fly treatment, and cultivar on number of tillers, number of infected tillers, and percent of infected tillers pot -1 during a 2-year period Table 3.3: Effect of cultivar on the mean number of tillers pot -1 during a 2-year period Table 3.4: Analysis of variance for the effects of year, fly treatment, and cultivar on undamaged, damaged, and total dry weight during a 2-year period Table 3.5: Analysis of variance for effects and interactions of year, treatment, and cultivar on phenotypic characteristics over a 2-year period Table 3.6: Effect of fly treatment on phenotypic plant characteristics over a 2-year period Table 3.7: Analysis of variance for the effects and interaction of yield and cultivar on the mean number of larvae excised pot -1 harvest -1 over a 2-year period Table 4.1: Location of field sampling sites Table 4.2: Analysis of variance for the effect of region on total number of ovarioles and total number of mature oocytes...63 Table 5.1: Summary of bermudagrass cultivar recommendations for the SE US....76

10 ix Table 5.2: Analysis of variance for the effects and interaction of year and cultivar on various phenotypic plant characteristics during a 2-year period Table 5.3: Effect of cultivar on phenotypic plant characteristics...78

11 x LIST OF FIGURES Page Figure 1.1: Large areas of bermudagrass with chlorotic tips indicate the bermudagrass stem maggot may be present, though a more thorough evaluation should be conducted....7 Figure 1.2: An individual shoot that has been damaged by the bermudagrass stem maggot as indicated by the restriction of the chlorosis to the top leaves (a). Damaged leaves that have been easily pulled from the pseudostem (b). Decaying plant material resulting from BSM larva feeding (c)...7 Figure 1.3: States where bermudagrass stem maggot damage has been verified and the growing season in which damage was first reported...8 Figure 2.1: The male abdomen is shorter and more rounded than the female due to the ovipositor on the end of the female s abdomen Figure 2.2: Adult flies are easily collected from infested fields utilizing an insect sweep net (a). Adults prefer dense forage canopies, rarely emerging more than a short distance above the sward (b)...23 Figure 2.3: The immature larva is somewhat difficult to find in the field (a). This is partially because of the small size of the larva (b). Also the damage symptoms persist after the larva matures and moves to the soil for pupation, leaving a visible hole in the bermudagrass tiller (c)...23 Figure 3.1: Acetate and mesh cylindrical enclosures were designed and constructed specifically for this project (a). Adult flies were swept from various locations around Georgia to be

12 xi introduced into the enclosures (b). Flies were introduced by tilting back the enclosure and opening the vial underneath (c)...41 Figure 3.2: This figure illustrates the timeline of an individual run...41 Figure 3.3: Effect of cultivar on the average number of infected tillers and the percent of infected tillers pot -1 for units in the fly treatment during a 2-year period. Points on the line sharing the same letter are similar (P < 0.05). Error bars illustrate LSD 0.05 values Figure 3.4: Effect of fly treatment on mean total dry biomass (g pot -1 ). Data pooled across year and treatment. Columns sharing the same letter are similar (P < 0.05) Figure 3.5: Effect of cultivar on mean undamaged and damaged dry biomass for units in the fly treatment over a 2-year period. Error bars illustrate LSD 0.05 values Figure 4.1: Map of field sampling sites Figure 4.2: The status of each fly s ovarioles was assigned to one the following categories: present (a), absent (b), or underdeveloped (c) Figure 4.3: Reproductive system of A. reversura female Figure 4.4: Comparison of the mean number of ovarioles and mature oocytes side -1. Error bars illustrate LSD 0.05 values Figure 4.5: Mature A. reversura oocytes Figure 4.6: Effect of region on the mean total number of ovarioles female -1 and mean total number of mature oocytes female -1 with ovarioles present. Error bars illustrate LSD 0.05 values Figure 4.7: Effect of region on the number of flies where ovarioles were present, underdeveloped, or not present. Region did not affect the number of flies in the group

13 xii where ovarioles were present or underdeveloped. Columns sharing the same letter are similar (P < 0.05) Figure 5.1: Canopy comparison of the cultivars selected for this study Figure 5.2: Relationship between mean number of tillers pot -1 and mean weight tiller -1 (g). Error bars illustrate LSD 0.05 values....76

14 1 CHAPTER 1 INTRODUCTION Bermudagrasses (Cynodon spp.) have developed into one of the most important forage crops in the United States (Taliaferro et al., 2004). This perennial, warm season forage is well suited for hay and pasture production in the hot and humid Southeast. Several improved hybrid bermudagrass cultivars have been developed in recent years that are higher yielding and superior in nutritive value than their predecessors. Bermudagrass is also resilient, persistent, and tolerant of grazing, drought, and pest pressure. However, in the summer of 2010 Georgia hay producers began noticing a bronzing of their bermudagrass hay fields generating damage similar to that of drought or frost damaged bermudagrass (Fig. 1.1; Hancock, 2012). The bronzing was the result of chlorosis in the top two to three leaves of the plant (Fig. 1.2a). Another characteristic symptom was that the damaged tillers could easily be pulled from the sheath and the perinodal end showed evidence of insect damage and/or obvious decay (Fig. 1.2b-c). Under controlled conditions, collected larvae were reared and the resulting adults were subsequently identified as Atherigona reversura Villeneuve, now commonly known as the bermudagrass stem maggot (BSM; W.G. Hudson, unpublished data, 2010). The BSM is thought to be native to SE Asia, where it was first collected during an exploratory expedition (Villeneuve, 1936). First reports of the BSM in the US followed diagnosis of BSM damage in turfgrass in Hawaii (Hardy, 1976). The first documented appearance of the BSM in North America was in 2009 when it was found in the Los Angeles

15 2 area of California (Holderbaum, 2009). Until the BSM was found in Georgia, it was mainly regarded as a turfgrass pest. Since the 2010 discovery in southern Georgia, BSM has spread throughout the southeast, infesting bermudagrass hayfields, pastures, and turf (Fig. 1.3). To date, BSM has been reported as far north as Kentucky (Townsend et al., 2013) and as far west as Texas (Vanessa Corriher, personal communication), although damage seems to be more severe in the hotter, more humid areas of the southeast. The BSM is now found throughout all of the areas of the SE US where bermudagrass is commonly grown for forage. Although the BSM has been reported in the literature many times, it has generally been considered of minor importance relative to other Atherigona species, such as the closely related A. soccata (sorghum shoot fly). There is a paucity of information about the lifecycle of A. reversura and how it can be managed or controlled, but some information is available on basic larval behavior, fly physiology, and the potential differences in tolerance among some Cynodon cultivars (Grzywacz et al., 2013; Ikeda et al., 1991; Pont et al., 1995). Integrated pest management (IPM) is a continuous, evolving process that uses field inspection and knowledge of pest biology to plan and implement techniques that minimize pest damage. IPM makes use of a variety of interventions including avoidance, tolerance, sanitation, habitat modification, mechanical suppression, chemical suppression, trapping, and biological/genetic modification of the pest. However, there are several challenges that face any IPM program. In general, Atherigona populations have proven to be difficult to fully control through mechanical or chemical means (Talati et al., 1978). The use of mechanical or chemical controls may, however, suppress the population to achieve an acceptable level of economic damage.

16 3 Another possible solution is cultural management. Cultural management, such as searching for or developing resistant cultivars, is a major part of integrated pest management. McNaughton (1983) discussed how the ability of a plant to resist or overcome damage is dependent on evolutionary responses. Examples include hormone reallocation, recycling of nutrients, and increase in photosynthetic capacity in remaining leaves (Detling et al., 1979; McNaughton, 1983). Both tolerance and avoidance can contribute to plant defenses (Ruiz et al., 2008). Tolerance is the plant s allocation of resources in stems and roots to be drawn upon if an insect attack occurs. It may also refer to the reallocation of resources within the plant once an attack occurs. There is extensive literature on avoidance mechanisms plants use to ward off insect populations including release of hormones, secondary metabolites, and toxins. Field observations indicate that bermudagrass relies more on tolerance than avoidance mechanisms to withstand BSM damage, although more research is needed. Previous research suggests that cultivars with a denser canopy and thinner tillers are more prone to BSM damage (Ikeda et al., 1991). The connection to more damage in fields with a higher number of tillers has also been observed throughout Georgia (L.L. Baxter and D.W. Hancock, unpublished data, 2012 and 2013). Before any in-depth research on chemical suppression begins, the life cycle and biology of the BSM that can be inferred from the literature and early research observations should be confirmed. Field observations show that BSM population density is variable. Understanding the reproduction potential of this pest will aid in understanding populations dynamics. Ovariole number can provide an indication of potential fecundity of a species, although many biotic and abiotic factors have been found to influence ovariole number in dipteran insects such as

17 4 environment, nutrition, and population density (Peters et al., 1977; Taylor et al., 2010; Bergland, 2011). Although the work by Ikeda et al. (1991) introduced interactions between bermudagrass characteristics and BSM injury, a more detailed understanding of the effects of the BSM on yield, yield components, and crop morphology, as well as confirmation of aspects of the BSM s life cycle are needed. The objectives of this study were to: 1. Compile any available literature on A. reversura or closely related species so that researchers and producers have a better understanding of this new, invasive pest 2. Compare the severity of damage among selected bermudagrass cultivars 3. Quantify the phenotypic variation in cultivar s response to the BSM 4. Assess the fecundity of the BSM in the selected bermudagrass cultivars 5. Study and describe the reproductive morphology of the BSM female 6. Determine if any differences exist in ovariole number between pooled BSM fly populations from Georgia and Florida 7. Compare key morphological characteristics among the most common and important Cynodon cultivars currently grown in the southeastern USA. References Bergland, A.O Mechanisms of nutrient-dependent reproduction in dipteran insects. In: T. Flatt and A. Heyland, editors, The Genetics and Physiology of Life History Traits and Trade-Offs. Oxford University Press, New York. p Detling, J.K., M.I. Dyer, and D.T. Winn Net photosynthesis, root respiration, and regrowth of Bouteloua gracilis following simulated grazing. Oecologia. 41:

18 5 Grzywacz, A., T. Pape, W.G. Hudson, and S. Gomez Morphology of immature stages of Atherigona reversura (Diptera: Muscidae), with notes on the recent invasion of North America. J. of Nat. History, 47:15-16, Hancock, D.W Bermudagrass stem maggot. Georgia Cattlemen. 40:20. Hardy, D.E Proceedings of the Hawaiian Entomological Society for Hawaiian Entomol. Soc. Proc., Honolulu, Hawaii, Hawaiian Entomol. Soc. 22(2). Holderbaum, B Orange Fly-Atherigona. IA State University Entomology. (accessed 15 May 2012). Ikeda, H., M. Oyamada, and H. Ando Varietal differences of bermudagrass in parasitic shoot ratio caused by bermudagrass stem maggot. (In Japanese). Jpn. Soc. Grassland Sci. 37(2): McNaughton, S.J Compensatory plant growth as a response of herbivory. Oikos. 40: Pont, A.C. and F.R. Magpayo Muscid shoot-flies of the Philippine Islands (Diptera: Muscidae, genus Atherigona Rondani ). Bull. Entomol. Res. Supp. Ser. 3: Peters, T.M. and P. Barbosa., Influence of population density on size, fecundity, and developmental rate of insects in culture. Ann. Rev. Entomol. 22: Ruiz, N., D. Ward, and D. Saltz Leaf compensatory growth as a tolerance strategy to resist herbivory in Pancratium Sickenbergeri. Plant Ecology. 198(1): Talati, G.M. and V.R. Upadhyay Status of shoot fly Atherigona approximata Malloch as a pest of Bajra Pennisetum tyyphoides crop in Gujarat State. GAU. Res. J. 4(1):

19 6 Taliaferro, C.M., F.M. Rouquette, and P.Mislevy Bermudagrass and stargrass. In: L.E. Moser, B.L. Burson, and L.E. Sollenberger, editors, Warm-season (C4) grasses, Agronomy Monograph No. 45. American Society of Agronomy, Madison. p Taylor, B.J. and D.W. Whitman A test of three hypotheses for ovariole number determination in the grasshopper Romalea microptera. Physiological Entomology. 35: Townsend, L. and S. Osborne Bermudagrass stem maggot found in Allen County. Kentucky Pest News No University of Kentucky, Lexington, KY. Villeneuve, J Schwedisch-chinesische wissenschaftliche expedition nach den nordwestlichen Provinzen Chinas, unter Leitung von Dr. Sven Hedin und Prof. Su Pingchang. Insekten gesammelt vom schwedischen Arzt der Expedition Dr. David Hummerl Diptera. 16. Muscidae. Ark. Zool. 27 A 34: 1-13.

20 7 Figure 1.1. Large areas of bermudagrass with chlorotic tips indicate the bermudagrass stem maggot may be present, though a more thorough evaluation should be conducted. Figure 1.2. An individual shoot that has been damaged by the bermudagrass stem maggot as indicated by the restriction of the chlorosis to the top leaves (a). Damaged leaves that have been easily pulled from the pseudostem (b). Decaying plant material resulting from BSM larva feeding (c).

21 Figure 1.3. States where bermudagrass stem maggot damage has been verified and the growing season in which damage was first reported. 8

22 9 CHAPTER 2 LITERATURE REVIEW In the summer of 2010, bermudagrass [Cynodon dactylon (L.) Pers.] hay producers began noticing a bronzing of their hay fields (Hancock, 2012). It began in low-lying areas, such as wheel depressions and in the uncut residual around the perimeter of the field. Soon the bronzing spread throughout entire fields, generating damage similar to that of drought or frost damaged bermudagrass (Fig. 1.1). The bronzing was the result of chlorosis in the top two to three leaves of the plant (Fig. 1.2a). Another characteristic symptom was that the damaged tillers could easily be pulled from the sheath and the perinodal end showed evidence of larval feeding and/or decay consistent with that which follows maceration of vascular tissue (Fig. 1.2b-c). Under controlled conditions, collected larvae were reared and allowed to pupate and mature. The resulting adults were subsequently identified as Atherigona reversura Villeneuve, now commonly known as the bermudagrass stem maggot (BSM; W.G. Hudson, unpublished data, 2010). Initial Discovery through Invasion to the United States The BSM was first collected in 1936 during a Swedish-Chinese expedition and documented in the Arkiv för Zoologi where Villeneuve noted that the species appeared to be an undescribed member of the genus Atherigona. During the late twenty-first century, various entomological researchers have found evidence of adult A. reversura throughout SE Asia while they were investigating other Atherigona species, such as the closely-related sorghum shoot fly, A. soccata (Davies et al., 1981; Pont, 1981; Pont et al., 1995). Generally, the BSM appears to be more common in warm, humid climates. A Japanese study on the BSM found that humidity

23 10 improved fecundity of the females and viability of the eggs (Ikeda et al., 1991). Ideal conditions for sorghum shoot fly egg viability and larval development is 30 C and between 60-90% humidity (Doharey et al., 1977). The BSM was first discovered in the US in Hawaiian turfgrass in the 1970 s. In January 1974, Dr. J.W. Beardsley captured an unknown fly in light traps on Oahu Island (Hardy, 1976; Hardy, 1981). The flies were sent to A.C. Pont for identification and in May 1974 the invasion of the BSM was officially documented by the Hawaiian Entomological Society (Hardy, 1976). Severe damage was seen in the months following, particularly in turfgrass situations (Hardy, 1976). The first documented appearance of the BSM in North America was not until 2009 when it was found in the Los Angeles area of California (Holderbaum, 2009). In the summer of 2010, BSM was documented in the SE USA throughout Pierce, Jeff Davis, and Tift Counties in Georgia (Hancock, 2012). Until the BSM was found in Georgia, it was mainly regarded as a turfgrass pest. Since the 2010 discovery in southern Georgia, BSM has spread throughout the southeast, infesting bermudagrass hayfields, pastures, and turf. To date, BSM has been reported as far north as Kentucky (Townsend et al., 2013) and as far west as Texas (Vanessa Corriher, personal communication), although damage seems to be more severe in the hotter, more humid areas of the southeast. The BSM is now found throughout all of the areas of the SE US where bermudagrass is commonly grown for forage. The exact time, place, and method of any of the aforementioned BSM introductions are still unknown. A list of other BSM reports from around the US can be found in Grzywacz et al. (2013). However, actual specimens are not available for authentication of many of these reports

24 11 (Grzywacz et al., 2013). An illustration of the population spread in terms of substantial agricultural damage is found in Fig Identifying the BSM The BSM adult is easier to find and identify than the larva or pupa because it is outside of the pseudostem and has distinct coloration (Fig. 2.1). Sweep net collections and identification of BSM within the field can be made relatively easily (Fig. 2.2a). One challenge with identifying the BSM fly in the field is that they tend to stay low in dense forage stands, rarely emerging more than a short distance above the canopy (Fig. 2.2b). Adult males and females have transparent wings, a gray thorax, and a yellow abdomen with at least one pair of black spots (Fig. 2.1a-b). Adult BSM range between 3.0 and 3.5 mm in length, and females are typically larger than the males. The female abdomen is longer and more pointed while the male s is shorter and more rounded. A thorough and detailed anatomical description of the male and female has been recorded by Pont et al. (1995). While there is no literature available on the internal morphology or physiology of the BSM, Unnithan (1981) describes and illustrates the internal reproductive system of the sorghum shootfly. Unnithan (1981) found that the females possessed a variable number of polytrophic ovarioles (33.4 ± 3.9), egg maturation occurred synchronously, and most females had an uneven number of ovarioles in each ovary. While the eggs have yet to be seen in a field setting, they were recently discovered during a micro-dissection of the BSM (Baxter et al., 2014). Grzywacz et al. (2013) provided a detailed description of the larva and puparium. Finding the larva is more challenging than catching the flies. Tillers may be carefully dissected using a sharp knife or razor blades. Slowly and carefully split the pseudostem until the center of the shoot is revealed. Because of the small size of the larva, it is best to work over a solid, dark colored surface so that the larva is not lost during the

25 12 procedure. Larvae are more frequently found in stems that show only initial symptoms of BSM damage. If tillers show extensive damage, it is likely the larva has already left the shoot to pupate. Mature larvae are whitish, cylindrical, and about 3 mm in length (Fig. 2.3a-b). As they mature, the color gradually darkens. The mouth-hooks are barely visible to the naked eye. It is presumed that these mouth-hooks enable the BSM to macerate the walls of the pseudostem. The larvae also have poorly developed oral ridges (Grzywacz et al., 2013). The puparium is orange to dark red and barrel shaped, which is similar to that of other Atherigona species (Grzywacz et al., 2013). As of yet, no protocols have been developed to search for the pupae in the soil of a damaged field. Predicted Biology and Damage Mechanism There have yet to be any reports published specifically on the biology of the BSM. Based on literature for the sorghum shoot fly and preliminary field observations, the life cycle appears to be approximately three weeks (Blum, 1967). The adult sorghum shoot fly lays its eggs on the underside of the sorghum leaf (Raina, 1981). The maggots hatch approximately 2.5 days postoviposition (Talati et al., 1978). Upon hatching, sorghum shoot fly larva travels along the underside of the leaf to the central whorl where it bores into the central shoot and macerates and feeds on the contents of the tiller at the node and the subsequent decaying plant material (Pont, 1981). It is critical that the larva penetrates the central shoot for survival. If the larva succeeds, visible damage is seen within 1-3 days (Blum, 1967). From here the larva exits the stem (Fig. 2.3c) and moves to the soil for pupation. After pupation, the adult emerges. There is very limited information on these later stages. Each of these known steps in the sorghum shoot fly s life cycle coincides with observations of BSM maturation.

26 13 Host Range Atherigona spp. are found on a wide range of hosts but are most commonly found on decaying plant material (Pont, 1981). According to Soto et al. (1971), insects in the order Diptera, such as Atherigona, utilize carbohydrates for energy. They primarily infest Gramineae (Pont, 1981). Each species within the genus Atherigona has a preferred host. While the BSM prefers bermudagrass, they have also been found in fields of jungle rice (Echinochloa colona), tropical cupgrass (Erichola procera), sain grass (Schima nervosum), finger millet (Eleusine coracana), sorghum (Sorghum bicolor), and corn (Zea mays; Davies et al., 1981; Pont, 1981). Many of these hosts have only been confirmed in SE Asia (Pont, 1981). Thus far, the BSM has only been collected from Cynodon grasses in the USA. The source of nutrition for the adult BSM flies remains unresolved. It has been observed that adult flies die quickly in cages without water. However, flies reared and kept in enclosures lived for approximately d when provided sugar water or live grass. It is known that grasses, through the process of guttation, exude low concentrations of sugar and other ions and Hancock et al. (2014) have hypothesized that the BSM flies may feed on sugar exudates from the grass (Duell et al., 1977). Fly populations and damage are often observed coincident with areas of higher N fertilization and/or bermudagrass leaf spot (Bipolaris cynodontis), and/or leaf rust (Puccinia cynodontis; Hancock et al., 2014). High N fertility and leaf diseases are associated with substantial amounts of guttation. This supports the sugar-exudate hypothesis for the source of nutrition for adult BSM flies, but additional research is needed to confirm it. Comparison of BSM Damage to Signs of Other Stresses Damage to bermudagrass by BSM is frequently mistaken for other abiotic and biotic stresses. Drought-stress, nutrient deficiency, bermudagrass leaf spot, and/or leaf rust can result in

27 14 discolored and/or senescent plant material that is similar to the damage done by the BSM (Andrae et al., 2012; Havlin et al., 2004; Martinez et al., 2012; Read et al., 2012; Robinson 1985). However, these can be distinguished from BSM damage by the location of the chlorosis (Hancock, 2012). Damage from the BSM only produces chlorosis in the top 2-3 leaves of the plant. Since the larvae feeds at the uppermost node near the top of the tiller, the portion above the node will die as a result of the larva s feeding while everything below the node will remain intact. Furthermore, if the leaves slide easily from the sheath when the discolored material is gently pulled, then the damage is likely from the BSM. Suppression Efforts In general, Atherigona populations cannot be fully controlled through mechanical or chemical means (Talati et al., 1978). The use of mechanical or chemical controls may, however, suppress the population to achieve an acceptable level of economic damage. Mechanical suppression (i.e. mowing) has been shown to prevent large populations of the sorghum shoot fly from building in sorghum fields (Young, 1981). For BSM in hayfields, the current recommendation is that bermudagrass hay producer should harvest the field as soon as conditions become favorable if they begin to see signs of damage within a week or so of the anticipated harvest date (Hancock, 2012). Damage seen earlier in the growth cycle will very likely reduce agronomic performance of the crop substantially. Once a stand that is 6 inches (15 cm) or taller has been damaged by BSM feeding, the only option is to cut and/or graze the stand to a height of 3-4 inches ( cm) and encourage regrowth to occur because the bermudagrass crop is unlikely to further develop. It is better to cut the field extremely early and accept the loss than to have a low-yielding, severely damaged crop that harbors a large fly population and leads to a further buildup. Ideally, the infected material would be removed from the field (Hancock,

28 ; Talati et al., 1978). The larvae do not appear to remain in cut stems. Within hours of cutting, larvae will exit damaged stems and travel to the soil. Those larvae that are mature enough to pupate will do so. Adult flies in fields that have been harvested escape to field margins and neighboring bermudagrass fields. Suppression of the BSM fly is challenging in part because the flies are mobile, though it is unclear to what degree the flies travel from one field to another or escape from a treated area. In our experience, the flies do not fly far (no more than 10 feet or meters) in any single instance of flight, even after being disturbed. Further, properly timing the control tactic requires some knowledge of how the effort will overlap with the life cycle, which remains vaguely understood at best in the case of the BSM. In addition, one must consider the limits of a chemical application in canopy penetration. Observations suggest the BSM flies tend to remain deep in the canopy except to move from one location to another or in response to a disturbance. According to Delobel (1982), sorghum shoot fly infestations are most easily reduced when the crop is in the lag and early exponential growth phases. Potentially, this could be true for BSM suppression, as well. During these phases, pesticide applications can penetrate more deeply into the canopy. Moreover, suppressing flies at this time could serve to protect the crop until the bermudagrass gets tall enough and/or produces adequate tillers to reduce the effects of the damage on yield. The current recommended chemical suppression technique is to apply a recommended rate of an inexpensive pyrethroid insecticide after the bermudagrass has begun to regrow (7-10 days after cutting) following an affected harvest (Hudson et al., 2013). A second application may be made 7-10 days later to suppress any flies that have emerged or arrived since the first application (Hudson et al., 2013). Chemical actions should be taken if there is a known history of

29 16 BSM damage to the bermudagrass and the expense of the two applications (usually less than $15/acre [$40/ha] for both applications) is justified by the forage yield saved. To date, economic thresholds for when action is warranted have not been established. Based on observations in Georgia (Baxter et al., unpublished data, 2014), BSM populations are usually not high enough to warrant chemical suppression prior to the first bermudagrass hay cutting (or equivalent timing if the crop is to be grazed). Population buildup may not occur until late May, June, or July in USDA Hardiness Zones 9, 8b, and 8a of the SE US, respectively (roughly, late into the regrowth cycle for the second cutting or during the third cutting). Such a population buildup may not occur until late summer for more northern areas where bermudagrass is grown. Cultivar Tolerance Blum et al. (1967) and Soto (1974) noted that Atherigona spp. demonstrated preference for some varieties over others. For example, the sorghum shoot fly attacks plants with certain morphological characteristics such as wide, dark green leaves more frequently than narrower, lighter leaves. As such, it is recommended to plant sorghum varieties that had been screened for physical and/or chemical attributes that decreased sorghum shoot fly preference (Table 2.1; Table 2.2; Blum et al., 1967; Soto, 1974). Ikeda et al. (1991) noted that the proportion of tillers damaged by the BSM was positively correlated to shoot density and negatively correlated to tiller diameter, length of internode, and width of leaf blade. The connection to more damage in fields with a higher number of shoots has also been observed throughout Georgia and confirmed in recent greenhouse trials (Baxter et al., 2013; Baxter et al., 2014).

30 17 Current and Future Research Objectives Much remains unknown about the BSM. Ongoing work to compare the severity of BSM damage among selected cultivars, determine if there are phenotypic differences in the cultivars that are preferred by BSM, and assess the fecundity of BSM on different cultivars will provide guidance for the appropriate cultivars upon which future work should be conducted. There is a great need to better understand the life cycle and reproductive potential of the BSM. Further, quantifying the severity of BSM damage on bermudagrass yield, quality, and aesthetics is necessary to develop economic thresholds for treatment. Finally, and most crucially, much research is needed to determine how best to manage and/or control BSM larvae and/or adults so that producers make more informed decisions when considering management options. References Andrae. J., A. Martinez, and R. Morgan Leafspot diagnosis and management in bermudagrass forages. C 877. University of Georgia Cooperative Extension Service, Athens. Baxter, L.L., D.W. Hancock, and W.G. Hudson Bermudagrass stem maggot: An exotic pest in the Southeastern United States. Paper presented at: ASA-CSSA-SSA Annual Meetings, Tampa, FL. 4 Nov. Baxter, L.L., D.W. Hancock, W.G. Hudson, and P.J. Moore The difference in response of selected bermudagrass cultivars (Cynodon dactylon) to bermudagrass stem maggot (BSM, Atherigona reversura Villeneuve) damage. Paper presented at: American Forage and Grassland Conference, Memphis, TN. 13 Jan.

31 18 Blum, A Varietal resistance of sorghum to the sorghum-shoot-fly (Atherigona varia var. soccata). Crop Sci. 7: Blum, A Anatomical phenomena in seedlings of sorghum varieties resistant to the sorghum shoot fly (Atherigona varia soccata). Crop Sci. 8: Chamarthi, S.K., H.C. Sharma, P.M. Vijay, and M.L. Narasu Leaf surface chemistry of sorghum seedlings influencing expression of resistance to sorghum shoot fly, Atherigona soccata. J. Plant. Biochem. Biotechnol. 20(2): Coleman, J.S. and C.G. Jones A phytocentric perspective of phytochemical induction by herbivores. In: D.W. Tallamy and M.J. Raupp, editors, Phytochemical Induction by Herbivores. John Wiley and Sons, Inc., New York. p Davies, J.C. and K.V. Seshu Reddy Shootfly species and their graminaceous hosts in Andhra Pradesh, India. Insect Sci. Application. 2(2): Delobel, A.G.L Oviposition and larval survival of the sorghum shoot fly Atherigona soccata Rond. as influenced by the size of its host plant (Diptera, Muscidae). Zeitschrift für Angewandte Entomologi. 93: Doharey, K.L., B.G. Srivastava, M.G. Jotwani, K. Dang Effect of temperature and humidity on the development of Atherigona soccata Rondani. Indian J. Entomol. 39(3): Duell, R.W., and D.K. Markus Guttation deposits on turfgrass. Agron. J. 69: Grzywacz, A., T. Pape, W.G. Hudson, and S. Gomez Morphology of immature stages of Atherigona reversura (Diptera: Muscidae), with notes on the recent invasion of North America. J. of Nat. History, 47:15-16, Hancock, D.W Bermudagrass stem maggot. Georgia Cattlemen. 40:20.

32 19 Hancock, D.W., W.G. Hudson, L.L. Baxter, and J.T. McCullers What we have learned about the bermudagrass stem maggot. Paper presented at: American Forage and Grassland Conference, Memphis, TN. 14 Jan. Hardy, D.E Proceedings of the Hawaiian Entomological Society for Hawaiian Entomol. Soc. Proc., Honolulu, Hawaii, Hawaiian Entomol. Soc. 22(2). Hardy, D.E Genus Atherigona Rondani. In: D.E. Hardy and W. Henning (eds.) Insects of Hawaii, Volume 14, Diptera: Cyclorrhapha IV. University Press Hawaii, Honolulu. p Havlin, J.L., S.L. Tisdale, W.L. Nelson, and J.D. Beaton Soil Fertility and Fertilizers: An Introduction to Nutrient Management 7th Edition. Prentice Hall. USA. Holderbaum, B Orange Fly-Atherigona. IA State University Entomology. (accessed 15 May 2012). Hudson, W., D. Hancock, K. Flanders, and H. Dorough Biology and management of bermudagrass stem maggot. Auburn Cooperative Extension System. Circular ANR Auburn, AL. Ikeda, H., M. Oyamada, and H. Ando Varietal differences of bermudagrass in parasitic shoot ratio caused by bermudagrass stem maggot. (In Japanese). Jpn. Soc. Grassland Sci. 37(2): International Crop Research Institute for the Semi-Arid Tropics ICRISAT Annual Report, Hyderbad, India. p Khurana, A.D., and A.N. Verma Some biochemical plant characters in relation to susceptibility of sorghum to stemborer and shootfly. Indian J. Entomol. 45(1):

33 20 Maiti, R. K. and P.T. Gibson Trichomes in segregating generations of sorghum matings. II. Association with shootfly resistance. Crop Sci. 23: Martinez, A., M. Pearce amd L. Burpee Turfgrass diseases in Georgia: Identification and control. Bul University of Georgia Cooperative Extension Service, Athens Mattson, Jr., W.J Herbivory in relation to plant nitrogen content. Annu. Rev. Ecol. Syst. 11: Nwanze, K.F., F.E. Nwilene and Y.V.R. Reddy Evidence of shoot fly Atherigona soccata Rondani (Dipt., Muscidae) oviposition response to sorghum seedling volatiles. J. Appl. Entomol. 122: Nwanze, K.F., Y.V.R. Reddy, and P. Soman The role of leaf surface wetness in larval behavior of the sorghum shoot fly, Atherigona. soccata. Entomol. exp. appl. 56: Ponniaya, B.W.X Studies in the genus Sorghum: II. The cause of resistance in Sorghum to the insect pest Atherigona indica M. MS Thesis, Univ. of Madras, Madras. Pont, A.C Some new Oriental shootflies (Diptera: Muscidae, genus Atherigona) of actual or suspected economic importance. Bull. Entomol. Res. 71: 371:393. Pont, A.C. and F.R. Magpayo Muscid shoot-flies of the Philippine Islands (Diptera: Muscidae, genus Atherigona Rondani ). Bull. Entomol. Res. Supp. Ser. 3: Raina, A.K Movement, feeding behavior, and growth of larvae of the sorghum shootfly, Atherigona soccata. Insect Sci. Application. 2(2): Raina, A.K Fecundity and oviposition behavior of the sorghum shootfly, Atherigona soccata. Entomol. Exp. Appl. 31:

34 21 Read, J.J. and R.G. Pratt Potassium influences forage bermudagrass yield and fungal leaf disease severity in Mississippi. Forage and Grazinglands. doi: /FG RS. Robinson, D.L Potassium nutrition of forage grasses. In: Robert Munson, editor, Potassium in Agriculture. pgs Schoonhoven, L.M Chemosensory bases of host plant selection. Annu. Rev. Entomol. 13: Singh, S.K., S.S. Yazdani, S.F. Hameed, and D.N. Mehto Bio-efficacy of spray applications of some newer insecticides against shoot fly Atherigona spp. on proso millet (Panicum miliaceum L.) in North Bihar. Entomon. 8(3): Soto, P.E. and K. Laxminarayana A method for rearing the sorghum shoot fly. Journal of Econ. Entomol. 64(2): 553. Soto, P.E Ovipositional preference and antibiosis in relation to resistance to a sorghum shoot fly. Journal of Econ. Entomol. 67(2): Staedler, E Sensory aspects of insect plant interactions. Proc. Int. Congr. Entomol. 15: Swarup, V. and D.S. Chaugale A preliminary study on resistance to stem-borer (Chilo Zonellus Swinh. Infestation in Sorghum (Sorghum vulgare Pers.). Curr. Sci. 4: Talati, G.M. and V.R. Upadhyay Status of shoot fly Atherigona approximata Malloch as a pest of Bajra Pennisetum tyyphoides crop in Gujarat State. GAU. Res. J. 4(1): Townsend, L. and S. Osborne Bermudagrass stem maggot found in Allen County. Kentucky Pest News No University of Kentucky, Lexington, KY.

35 22 Unnithan, G. C Aspects of sorghum shootfly reproduction. Insect Sci. Application. 2(1): Villeneuve, J Schwedisch-chinesische wissenschaftliche expedition nach den nordwestlichen Provinzen Chinas, unter Leitung von Dr. Sven Hedin und Prof. Su Pingchang. Insekten gesammelt vom schwedischen Arzt der Expedition Dr. David Hummerl Diptera. 16. Muscidae. Ark. Zool. 27 A 34: Young, W.R Fifty-five years of research on the sorghum shootfly. Insect Sci. Application. 2(2): 3-9. Figure 2.1. The male abdomen is shorter and more rounded than the female due to the ovipositor on the end of the female s abdomen.

36 23 Figure 2.2. Adult flies are easily collected from infested fields utilizing an insect sweep net (a). Adults prefer dense forage canopies, rarely emerging more than a short distance above the sward (b). Figure 2.3. The immature larva is somewhat difficult to find in the field (a). This is partially because of the small size of the larva (b). Also, the damage symptoms persist after the larva matures and moves to the soil for pupation, leaving a visible hole in the bermudagrass tiller (c).

37 24 Table 2.1. Summary of plant characteristics that reduce Atherigona damage in other field crops. Characteristic Plant Pest Source(s) Thicker Stems (more lignin, etc.) Sorghum A. soccata Blum, 1967; Blum, 1968; Delobel, 1982; Khurana et al., 1984; Nwanze et al., 1990 Millet Atherigona spp. Singh et al., 1983 Tillering Ability Sorghum A. soccata Blum, 1967; Young, 1981 Trichomes Sorghum A. soccata ICRISAT, 1978; Young, 1981; Maiti et al., 1983; Chamarthi et al., 2011 Glossy Leaves Sorghum A. soccata Young, 1981; Maiti et al., 1983; Nwanze et al., 1990 Coarser Texture Leaves Sorghum A. soccata Raina, 1982 Darker Leaves Sorghum A. soccata Raina, 1982 Lighter Leaves Sorghum A. soccata Site of Damage Sorghum A. soccata Delobel, 1982 Wider Leaves Sorghum A. soccata Blum, 1968 Irregular Shaped Silica Bodies Sorghum A. soccata Blum, 1968 Nwanze et al., 1998; Chamarthi et al., 2011 A. indica Ponniaya, 1951 Less Mature Sorghum A. indica Swarup et al., 1962 Table 2.2. Summary of plant compounds found to attract or deter A. soccata. Compound + or - Source(s) Undecane 5- methyl + Chamarthi et al., 2011 Decane 4- methyl + Chamarthi et al., 2011 Hexane 2,4- dimethyl + Chamarthi et al., 2011 Pentadecane 8- hexyl + Chamarthi et al., 2011 Dodecane 2, 6, 11- trimethyl + Chamarthi et al., , 4- dimethylcyclooctene - Chamarthi et al., 2011 Phenols + Khurana et al., 1983 Phosphorus + Khurana et al., 1983 Tannins, Alkaloids - Mattson, 1980; Khurana et al., 1987 Phenols - Coleman et al., 1991 Carbohydrates - Schoonhoven, 1968 Amino Acids - Schoonhoven, 1968 Vitamins - Schoonhoven, 1968 Nitrogen + Mattson, 1980; Coleman et al., 1991 Aromatics + Staedlar, 1977

38 25 CHAPTER 3 THE RESPONSE OF SELECTED BERMUDAGRASS CULTIVARS TO BERMUDAGRASS STEM MAGGOT DAMAGE 1 1 L. Baxter, D.W. Hancock, W.G. Hudson, W.F. Anderson, and B.M. Schwartz To be submitted to Agronomy Journal

39 26 Abstract Information regarding the susceptibility of currently grown bermudagrass [Cynodon dactylon (L.) Pers.] cultivars to the bermudagrass stem maggot (BSM; Atherigona reversura Villeneuve) will aid forage producers with the implementation of integrated pest management (IPM) strategies to manage this exotic pest. The objectives of this research were to compare the severity of damage among selected cultivars, quantify the phenotypic variation in cultivar response to the BSM, and assess the fecundity of the BSM on selected bermudagrass cultivars. Eight Cynodon cultivars were used in this study. Flies collected from infested fields were introduced six times throughout the 4-wk growing period to the cultivars, which were grown in the greenhouse and contained in acetate and mesh enclosures. At the end of each growth period, the forage was harvested and morphological characteristics were analyzed. The number and percent of tillers damaged depended upon cultivar, with cultivars with higher tiller density exhibiting the greatest damage. An average 7.7% decrease in total dry biomass was observed for all cultivars in this study. Presence of the BSM was coincident with a lower tiller count, increased tiller diameter, and darker leaf color, yet no difference in weight tiller -1 was observed. The results showed that stargrass and Tifton 68 (C. nlemfuensis Vanderyst) and hybrids of C. dactylon with C. nlemfuensis, including Coastcross-II and Tifton 85 are less susceptible to damage by the BSM and should be employed in IPM strategies wherever these cultivars are adapted. Introduction In the summer of 2010, Georgia hay producers began noticing a bronzing of their bermudagrass hay fields generating damage similar to that of drought or frost damaged bermudagrass (Fig. 1.1; Hancock et al., 2012). The bronzing was the result of chlorosis in the top

40 27 two to three leaves of the plant (Fig. 1.2a). Another characteristic symptom was that the damaged tillers could easily be pulled from the sheath and the end inside the sheath showed evidence of either insect damage or obvious decay (Fig. 1.2b-c). Under controlled conditions, collected larvae were reared and the resulting adults were subsequently identified as Atherigona reversura Villeneuve, now commonly known as the bermudagrass stem maggot (BSM; W.G. Hudson, unpublished data, 2010). The BSM is thought to be native to SE Asia, where it was first collected during an exploratory expedition (Villeneuve, 1936). First reports of BSM in the US followed diagnosis of BSM damage in turfgrass in Hawaii (Hardy, 1976). The first documented appearance of the BSM in North America was not until 2009 when it was found in the Los Angeles area of California (Holderbaum, 2009). Until the BSM was found in Georgia, it was mainly regarded as a turfgrass pest. Since the 2010 discovery in southern Georgia, BSM has spread throughout the southeast, infesting bermudagrass hayfields, pastures, and turf (Fig. 1.3). To date, BSM has been reported as far north as Kentucky (Townsend et al., 2013) and as far west as Texas (Vanessa Corriher, personal communication), although damage seems to be more severe in the hotter, more humid areas of the southeast. The BSM is now found throughout all of the areas of the SE US where bermudagrass is commonly grown for forage. Although the BSM has been reported in literature many times, it has generally been considered secondary to other Atherigona species, such as the closely related A. soccata (sorghum shoot fly). There is a paucity of information about the lifecycle of A. reversura and how it can be managed or controlled, but some information is available on basic larval behavior, fly physiology, and the potential differences in tolerance among some Cynodon cultivars. It is predicted that upon hatching, the BSM larva migrates down the leaf and penetrates the

41 28 pseudostem at the first node where it then feeds on the subsequent decaying plant material before exiting the pseudostem to pupate and emerging as an adult fly (Hancock et al., 2014). The result of this feeding injury causes senescence and necrosis of the terminal leaves on the affected shoots (Hancock et al., 2014). In severe infestations, over 80% of the shoots in a given area may be affected (Hancock et al., 2014; D.W. Hancock, unpublished data, 2014). Cultural management, such as employing resistant or more tolerant cultivars, is a major part of integrated pest management. Previous research suggests that cultivars with a denser canopy and thinner tillers are more prone to BSM damage (Ikeda et al., 1991). Those conclusions are consistent with general observations in the southeastern USA that stands of bermudagrass cultivars having the highest tiller density seem to be most severely damaged by the BSM (L.L. Baxter and D.W. Hancock, unpublished data, 2012 and 2013). Although the work by Ikeda et al. (1991) introduced interactions between physical characteristic of bermudagrass and BSM damage, a more detailed understanding of the effects of the BSM on yield, yield components, and crop morphology, as well as confirmation of aspects of the BSM s life cycle are needed. The objectives of this research were to compare the severity of damage among selected cultivars, quantify the phenotypic variation in cultivar response to the BSM, and assess the fecundity of the BSM on the cultivars. Materials and Methods Cultivar Selection Eight Cynodon cultivars were selected to represent a range of canopy densities from very dense to relatively open and to include newer C. dactylon cultivars and C. dactylon x C. nlemfuensis hybrids that were not included in the previous research (Ikeda et al., 1991). Selected

42 29 cultivars included four C. dactylon cultivars (common [ecotype: Tifton, GA], Coastal, Alicia, and Russell ), two C. nlemfuensis cultivars ( Tifton 68 and stargrass), and two C. dactylon x C. nlemfuensis hybrids ( Tifton 85 and Coastcross-II ). All cultivars were propagated from the USDA-Agricultural Research Service s Cynodon germplasm nursery at the Coastal Plain Experiment Station in Tifton, GA. Forage Management Bermudagrass sprigs were transplanted from the field into 15.5 cm diameter pots with Fafard 3B Potting Mix (Sun Gro Horticulture, Agawam, MA) before they were transported to the Crop and Soil Science Greenhouses in Athens, GA. After transport, all pots were fertilized with a greenhouse fertilizer at a rate equivalent to 2.2 kg N ha -1 d -1. Seventy-eight grams of fertilizer were dissolved into 15 liters of water and ~120 ml of the solution was hand applied to each experimental unit. A micro-irrigation system was installed to supply water directly to the soil surface. The irrigation line was controlled by a battery-operated, programmable timer set to deliver water for at least 2 minutes at 12-hour intervals. The duration was adjusted as needed to regulate soil moisture. The grasses were allowed one month to acclimate before the trial began. Fly Treatments From the 12 experimental units for each set of the eight cultivars, one experimental unit receiving flies and one receiving no flies were randomly assigned to one of six blocks. Blocks were then randomly assigned to an area on the benches along a slight shade and temperature gradient that existed within the greenhouse. Therefore, the experiment was a randomized

43 30 complete block design with treatments being complete combinations of eight cultivars and two levels of fly treatments (with or without flies) in six replicated blocks. Each experimental unit, regardless of treatment, was topped with a clear acetate and mesh cylindrical enclosure designed and constructed specifically for this project (Fig. 3.1a). Durable, clear acetate sheets (70 cm x 43 cm) were coiled to make a tube. A 6-cm strip of fine mesh fabric was glued on to each end to attach them together to form the cylinder, and a 20-cm circle cut from the mesh was glued on to enclose the top. The clear structure allowed light to reach the plant and the mesh permitted airflow to prevent excess moisture buildup on the acetate and the foliage. The enclosures ensured that flies were restricted to the experimental unit they were introduced to and, in the instance a fly may have escaped into the greenhouse, protected those units in the treatment without flies from damage. Three EL-USB-1 Data Loggers (Lascar electronics, Erie, PA) were placed at equal distances on the greenhouse bench along the slight shade and temperature gradient and set to record temperature and humidity every 5 minutes during the growing period. Flies were collected in several bermudagrass fields around Georgia using an insect sweep net (Fig. 3.1b). Locations of the sites where the BSM flies were collected are listed in Table 3.1. The sweep samples were transferred to a mesh enclosure with a small dish of damp sand (Soto et al., 1971). The enclosure was then placed on ice packs in a cooler to be transported back to Athens. Adult BSM flies were separated from the sweep sample and placed into small plastic vials. A pair of flies, one male and one female, was then introduced into each of the enclosures of the 48 experimental units under the fly treatment. Flies were introduced by slightly tilting back the enclosure and opening the vial inside (Fig. 3.1c). This technique took advantage of the

44 31 tendency of the flies to travel towards light since they would move towards the top of the enclosure while the bottom was secured back into the soil to prevent escape. Experimental Design and Project Timeline Four runs were completed during the 2012 growing season (7/2/2012 to 10/25/2012) but, as a result of unusual weather, only three runs were completed in 2013 (7/8/2013 to 9/30/2013). Runs consisted of 4-wk growth periods, simulating a recommended bermudagrass hay cutting schedule (Fig. 3.2). At the beginning of each run, all units were clipped to a 5-cm stubble height. After a 1-week initial growth stage, the first pair of flies was introduced and another pair of flies was introduced every 3-4 days thereafter for a total of 6 introductions during the 4-week growing period for each run. Data were averaged across runs within the respective year unless the difference between years was determined to be one of magnitude, and then the data were averaged across all seven runs. Canopy Evaluation At the conclusion of the growing period, all units were moved from the greenhouse and brought into a neighboring laboratory for evaluation. First, canopy height (± 1 cm) and number of tillers were recorded. A SPAD meter (Konica Minolta, Ramsey, NJ) was used to record and calculate the average measure of color (to an accuracy of ± 1.0 SPAD units) of five randomly selected leaves. Then, five tillers were randomly selected and clipped off at the soil surface. A micrometer was used to measure the width of the leaf blade that emerged from the second node below the terminal end, as well as the width of the pseudostem at this node (both measurements recorded to an accuracy of ± 0.01 mm). Measurements were taken at the second node for consistency and because damage occurs at the uppermost node. During the first year of this study

45 32 internode length was also evaluated by measuring the distance between the five nodes from the basal end on each of the five tillers to an accuracy of ± 0.01 mm. The remaining sample was then hand clipped to 5 cm and, including the five tillers previously harvested, meticulously analyzed to determine the number of damaged and undamaged tillers. Tillers were designated as damaged if the top two to three leaves could be easily slipped out of sheath of the pseudostem. All tillers determined to exhibit signs of BSM damage were dissected to search for the BSM larva. The green weight of the harvested sample was then recorded to an accuracy of ± 0.01 g. Samples were dried at 60 C for 48 hr to correct for moisture content. This data collection protocol resulted in the following response variables: number of tillers pot -1, number of infected tillers pot -1, leaf color, tiller diameter, leaf width, canopy height, internode length, number of larvae excised, and total dry biomass pot -1. These variables were also used to calculate other response variables, including percent of infected tillers pot -1 (Eq. [3.1]), weight tiller -1 (Eq. [3.2]), damaged biomass pot -1 (Eq. [3.3]), undamaged biomass pot -1 (Eq. [3.4]), and mean number of larvae excised pot -1 harvest -1 (Eq. [3.5]).

46 33 Statistics Statistical analyses were performed for each response variable using PROC MIXED in SAS 9.1 (SAS Institute, 2001). Treatment, cultivar, and year were all designated as fixed effects, while replication was set as a random effect. Unless otherwise indicated, differences were considered significant at α = 0.05 level. Results and Discussion Infected Tillers The effect of individual cultivar response to fly treatment differed by year (P < 0.05) with respect to the total number of tillers pot -1, the number of infected tillers pot -1, and the percent of tillers damaged (Table 3.2). However, these were interactions of magnitude and likely the result of differences between years. Therefore, to determine the impact on the BSM on the selected cultivars, data were pooled across the two years to focus on the interacting effects (P < ) of fly treatment and cultivar on the number of infected tillers and the percent of tillers damaged. The number of infected tillers pot -1 when flies were present was greater in the finer, C. dactylon cultivars than the coarser stargrass or those with stargrass-influence (P < ; Fig. 3.3), with common bermudagrass having the highest mean number of infected tillers (12.7 infected tillers pot -1 ), more than even the other fine-textured cultivars. The stargrass cultivar had the lowest number of infected tillers (1.93 infected tillers pot -1 ), but it was not significantly different from the other three coarse textured cultivars (P < ). It should be noted that the number of tillers differs among these eight cultivars (P < ; Table 3.2 and 3.3), but when damage is compared using the percent of tillers damaged, the C. dactylon cultivars still exhibit greater damage than those with stargrass-influence. These results are consistent with previous

47 34 research (Ikeda et al., 1991) that showed finer texture bermudagrass cultivars tend to be more heavily damaged than those that are coarser textured. Ikeda et al. (1991) posited that BSM damage was positively correlated with canopy density in that Coastal (C. dactylon) bermudagrass had a higher percent of tillers damaged than Tifton 68 (C. nlemfuensis; Ikeda et al., 1991). Cultivars of C. dactylon have denser canopies with a high number of fine-textured tillers whereas C. nlemfuensis canopies are more open with a lower number of very large, coarse tillers (Table 3.3). The trend that BSM damage is more prevalent in bermudagrass with dense forage canopies is also consistent with observations in hay fields and pastures throughout the Southeast US. It is thought that the denser canopies of the C. dactylon cultivars, such as Coastal and common, provide a more suitable habitat for the BSM. The density of a forage canopy is the product of many interacting plant characteristics, including tiller diameter and number of tillers. Effect on Biomass Production Though differences between years caused an interaction in how much was produced by individual cultivars, no other significant interaction of factors was observed in regards to total biomass production pot -1 (Table 3.4). Since the selected cultivars are well known for their differing yield potential, the influence of cultivar (P < ) on total biomass pot -1 was expected. Yet, the fact that the presence of the fly reduced (P = ) total biomass production by 7.7% (Fig. 3.4) without significantly reducing some cultivars more than others (P = ) seems to contrast with the data showing that tiller damage was worse in the fine-textured cultivars compared to the coarser, C. nlemfuensis influenced cultivars. This also does not correspond with field observations that suggest that fine-textured cultivars tend to exhibit greater yield loss in response to BSM damage than do the coarser, C. nlemfuensis-influenced cultivars.

48 35 Here it should be noted that these very controlled conditions within the greenhouse may not fully represent the potential yield loss observed in the field, mainly because of the aspects of timing and control of BSM fly populations. Under the conditions of this experiment, the BSM population did not differ between the cultivars, whereas any demonstrable preference by the BSM for one or more cultivars in the field may cause yield loss to be greater for those cultivars. In contrast, the cultivar and fly treatment did interact to influence the amount of undamaged biomass pot -1 (P < ) and damaged biomass production (P = 0.07; Table 3.4). Since the experimental units that were not exposed to flies always had damaged dry biomass values equal to zero, comparisons of the biomass production of the different cultivars when exposed to the flies provides insight into the effects of the BSM (Fig. 3.5). The weights of the damaged tillers from the fine-textured C. dactylon cultivars (common, Coastal, Alicia, and Russell ) were not different from each other, and those of the coarse-textured C. nlemfuensisinfluenced cultivars (stargrass, Coastcross-II, Tifton 68, and Tifton 85 ) were not different from each other. However, the fine-textured cultivars generally had more damaged biomass than the coarse-textured cultivars. Even when accounting for total yield differences among the selected cultivars, the damaged biomass represented a greater proportion of the total biomass in the fine-textured cultivars compared to the coarser C. nlemfuensis-influenced cultivars (26.1% vs. 8.4%, respectively). Physiological Changes in Response to the BSM Cultivars are known for having different phenotypes, and the cultivars exhibited significant differences in each of the physiological characteristics assessed (Table 3.5). However, the presence of the BSM affected leaf color, tiller diameter, and tiller count but did not affect leaf width, weight tiller -1, canopy height, or internode length (Table 3.5). Moreover, the cultivars did

49 36 not respond differently to the presence of the BSM (i.e., no fly treatment x cultivar interaction) on any of the phenotypic characteristics assessed. Consequently, to determine the impact of the BSM on the physiological changes, data were pooled across the cultivars and the two years (Table 3.6). Bermudagrass subjected to fly damage had darker leaves and the tillers tended to be thicker. It should be noted that leaf color assessments were only made on leaves that were either on undamaged tillers or below the leaves that may have been damaged by the BSM. The darkening of the leaves is most likely explained by an increase in light saturation for photosynthesis in the remaining leaves (Detling et al., 1979). However, compensatory photosynthesis may also have occurred (Nowak et al., 1984). Greening in response to damage may also be in response to a reallocation of nutrients and/or photosynthates, which can be easily moved within a plant assuming the vascular tissues and the interconnections between them allow for such translocation (Dunford, 2013). Since the BSM larvae bore into the pseudostem and macerate and feed on the vascular tissue, they are essentially cutting through many of these connections. Furthermore, while the rate of nutrient flow in bermudagrass or time required for the larva to penetrate the shoot is not known, Fischer (2007) found that young leaves import more nutrients than they export to fuel exponential plant growth and some nutrients are highly mobile while others are relatively immobile in the phloem. Full recovery of nutrients would not be expected because of the possible lack of time to export this quantity of nutrients, loss of highly mobile nutrients via leakage from the bored hole in the pseudostem, and loss of immobile nutrients trapped in the decaying leaves. Cultivars which show more tolerance may have a greater capacity to relocate nutrients before all vascular tissues are compromised. At this point in time, it is not known if the darkening of bermudagrass leaves and the enlargement of tillers in the

50 37 presence of the BSM is the result of increased light saturation or a reallocation of nutrients. Without supporting evidence, it would be premature to assume that nutrients from the damaged portions of the plant are reallocated. Since the leaves became darker green and the diameter of each tiller increased, it was expected that the weight tiller -1 would have increased. However, the weight tiller -1 did not significantly change (P = ; Table 3.5). Presumably, the stress of the larval feeding hindered the ability of the bermudagrass to take advantage of these opportunities, despite having conditions conducive for compensatory plant growth created by the BSM. One other explanation may be that a significant difference may have been detected had a more accurate scale been used (e.g., accurate to ± g instead of our measuring to ± 0.01 g). Fecundity of the BSM The mean number of larvae excised pot -1 harvest -1 differed (P = ) between 2012 and 2013 (0.130 vs larvae pot -1 harvest -1, respectively) but was not affected by cultivar in either year (Table 3.7). It is important to note that the tillers were only examined every four weeks. It is possible that the BSM larvae may have been present but not found, either because they had already dropped to the soil and begun to pupate or had emerged as a mature fly at the time of inspection. Conclusions The number and percent of tillers damaged by the BSM were significantly influenced by bermudagrass cultivars. Trends from this research indicate that coarse-textured, stargrassinfluenced cultivars exhibit more tolerance to BSM damage compared to the finer leaf textured bermudagrass cultivars. Overall, the presence of the BSM resulted in an average 7.7% decrease

51 38 in total dry biomass under the very controlled conditions of this study. Actual damage in uncontrolled conditions in the field would likely be substantially greater. While the impact on yield was somewhat expected, the findings that the BSM elicited physiological changes in the bermudagrass forage canopy was an exciting discovery. When the BSM is present, there is a decrease in tiller number and an increase in tiller diameter and leaf greenness. The increase in size and color of tillers of affected plants may be attributed to a reallocation of nutrients within the plant and/or increased light penetration into the remaining canopy. Despite this increase in diameter and color, a lack of apparent change in the average weight stem -1 indicates that the plant cannot overcome the stress of the larval feeding to exploit these opportunities for compensatory growth, especially given that the tiller count is reduced in response to BSM activity. No differences were observed for leaf width, canopy height, or internode length when the forage was exposed to the BSM, suggesting these factors are genetically predetermined and unlikely to change as a result of BSM larvae feeding damage. Despite evidence of greater damage to the fine-textured cultivars compared to the coarsetextured cultivars, no difference in the fecundity of the BSM on the bermudagrass cultivars evaluated in this research was detected. However, it is possible that larvae may have been present, but not found, as an artifact of the introduction or sampling timing of this study relative to the life cycle of the BSM. While many questions remain unanswered regarding the life cycle or control of the BSM, this research does provide valuable information that will serve as the groundwork for future investigation. Further efforts to better understand the life cycle of this species are warranted. This will enable researchers to develop appropriate and economical suppression efforts for producers. Based on the results of this study, tolerance appears to be a useful IPM strategy for producers to

52 39 implement for BSM control. Assuming the cultivar is adapted to the site, producers should choose coarse-stemmed cultivars (e.g., Tifton 85 ) that are not as severely damaged by the BSM, particularly in regions where BSM populations and fecundity seems to be highest. References Detling, J.K., M.I. Dyer, and D.T. Winn Net photosynthesis, root respiration, and regrowth of Bouteloua gracilis following simulated grazing. Oecologia. 41: Dunford, S Sieve element as the transport cells between sources and sinks. In: L. Taiz and B.B. Gollapudi (eds.), A Companion to Plant Physiology, Fifth Ed. Sinauer Associates, Inc. Publishers, Sunderland, MA. [2013 is date accessed] Fischer, A.M Nutrient remobilization during leaf senescence. In: S. Gan (ed.), Senescence Processes in Plants: Annual Plant Reviews, Volume 26. Blackwell Publishing Limited, Oxford, UK. p Hancock, D.W Bermudagrass stem maggot. Georgia Cattlemen. 40:20. Hancock, D.W., W.G. Hudson, L.L. Baxter, and J.T. McCullers What we have learned about the bermudagrass stem maggot. Paper presented at: American Forage and Grassland Conference, Memphis, TN. 14 Jan. Hardy, D.E Proceedings of the Hawaiian Entomological Society for Hawaiian Entomol. Soc. Proc., Honolulu, Hawaii, Hawaiian Entomol. Soc. 22(2). Holderbaum, B Orange Fly-Atherigona. IA State University Entomology. (accessed 15 May 2012).

53 40 Ikeda, H., M. Oyamada, and H. Ando Varietal differences of bermudagrass in parasitic shoot ratio caused by bermudagrass stem maggot. (In Japanese). Jpn. Soc. Grassland Sci. 37(2): Nowak, R.S. and M.M. Caldwell A test of compensatory photosynthesis in the field: Implications for herbivory tolerance. Oecologia. 61(3): SAS Institute SAS/STAT user's guide, v 9.1. SAS Institute Inc, Cary, NC. Soto, P.E. and K. Laxminarayana A method for rearing the sorghum shoot fly. Journal of Econ. Entomol. 64(2): 553. Townsend, L. and S. Osborne Bermudagrass stem maggot found in Allen County. Kentucky Pest News No University of Kentucky, Lexington, KY. Villeneuve, J Schwedisch-chinesische wissenschaftliche expedition nach den nordwestlichen Provinzen Chinas, unter Leitung von Dr. Sven Hedin und Prof. Su Pingchang. Insekten gesammelt vom schwedischen Arzt der Expedition Dr. David Hummerl Diptera. 16. Muscidae. Ark. Zool. 27 A 34: 1-13.

54 41 Figure 3.1. Acetate and mesh cylindrical enclosures were designed and constructed specifically for this project (a). Adult flies were swept from various locations around Georgia to be introduced into the enclosures (b). Flies were introduced by tilting back the enclosure and opening the vial underneath (c). Figure 3.2. This figure illustrates the timeline of an individual run.

55 42 a a a a Mean Number of Infected Tillers Pot -1 Mean Percent of Infected Tillers Pot -1 b b b b Figure 3.3. Effect of cultivar on the average number of infected tillers and the percent of infected tillers pot -1 for units in the fly treatment during a 2-year period. Points on the line sharing the same letter are similar (P < 0.05). Error bars illustrate LSD 0.05 values. a b Figure 3.4. Effect of fly treatment on mean total dry biomass (g pot -1 ). Data pooled across year and treatment. Columns sharing the same letter are similar (P < 0.05).

56 43 Mean Undamaged Dry Biomass Pot -1 Mean Damaged Dry Biomass Pot -1 Figure 3.5. Effect of cultivar on mean undamaged and damaged dry biomass for units in the fly treatment over a 2-year period. Error bars illustrate LSD 0.05 values. Table 3.1. Location of field sampling sites Producer/Farm City, State Latitude, Longitude Plant Science Farm (University of Georgia) Bogart, GA , Central Georgia Experiment Station Eatonton, GA , Pat Weems Eatonton, GA , Milo Hege Millen, GA , Jerry Martin East Dublin, GA , Sim Yoder East Dublin, GA , Sim Yoder East Dublin, GA , Chip Roche East Dublin, GA , Chip Roche East Dublin, GA , Joseph Studstill Abbeville, GA , Abraham Baldwin Agricultural College Tifton, GA ,

57 44 Table 3.2 Analysis of variance for the effects and interactions of year, fly treatment, and cultivar on number of tillers, number of infected tillers, and percent of infected tillers pot -1 during a 2- year period. Effect Number of Tillers Number of Infected Percent of Tillers Tillers Damaged F-value P > F F-value P > F F-value P > F Cultivar < < Fly trt < < Fly trt*cultivar < < Year < < Year*Cultivar < Fly trt*year < < Fly trt*year*cultivar Table 3.3. Effect of cultivar on the mean number of tillers pot -1 during a 2-year period. Cultivar Mean number of tillers pot -1 Common 87 a Russell 58 b Alicia 57 b Tifton bc Coastcross-II 46 bc Coastal 38 cd Tifton cd Stargrass 27 d Mean 50 SE LSD LSD Means followed by the same letter are similar at the P < 0.05 level.

58 45 Table 3.4. Analysis of variance for the effects of year, fly treatment, and cultivar on undamaged, damaged, and total dry weight during a 2-year period. Effect Total Dry Biomass Pot -1 Undamaged Dry Biomass Pot -1 Damaged Dry Biomass Pot -1 F-value P > F F-value P > F F-value P > F Cultivar 9.26 < < < Fly trt < < Fly trt*cultivar < Year < Year*Cultivar 5.34 < Fly trt*year Fly trt*year* Cultivar

59 46 Table 3.5. Analysis of variance for effects and interactions of year, treatment, and cultivar on phenotypic characteristics over a 2-year period. F-value P > F F-value P > F F-value P > F Leaf Color Leaf Width Tiller Diameter Cultivar < < < Fly trt Fly trt*cultivar Year < Year*Cultivar < Fly trt*year Fly trt*yr*cult Number of Tillers Weight Tiller -1 Canopy Height Cultivar < < < Fly trt Fly trt*cultivar Year < < Year*Cultivar < < Fly trt*year Fly trt*yr*cult Length Between Ground and First Node Length Between First and Second Nodes Length Between Second and Third Nodes Cultivar 7.15 < < < Fly trt Fly trt*cultivar Year n/a n/a n/a n/a n/a n/a Year*Cultivar n/a n/a n/a n/a n/a n/a Fly trt*year n/a n/a n/a n/a n/a n/a Fly trt*yr*cult n/a n/a n/a n/a n/a n/a Length Between Third and Fourth Nodes Length Between Fourth and Fifth Nodes Cultivar 5.89 < Fly trt Fly trt*cultivar Year n/a n/a n/a n/a Year*Cultivar n/a n/a n/a n/a Fly trt*year n/a n/a n/a n/a Fly trt*yr*cult n/a n/a n/a n/a N/A indicates data only collected during year one of this study

60 47 Table 3.6. Effect of fly treatment on phenotypic plant characteristics over a 2-year period. Without Flies With Flies SE LSD 0.05 Leaf Color (SPAD units) 18.8 a 20.2 b Leaf Width (mm) 2.98 a 2.98 a n.s. Tiller Diameter (mm) 1.14 a 1.22 a n.s. Number of Tillers 50.1 a 44.2 b Weight Tiller -1 (g) 0.09 a 0.10 a n.s. Canopy Height (cm) 26.6 a 26.2 a n.s. Length Between Ground and First Node (mm) 47.1 a 44.6 a n.s. Length Between First and Second Nodes (mm) 37.5 a 35.9 a n.s. Length Between Second and Third Nodes (mm) 41.5 a 41.8 a n.s. Length Between Third and Fourth Nodes (mm) 46.2 a 42.9 a n.s. Length Between Fourth and Fifth Nodes (mm) 46.1 a 44.5 a n.s. Means followed by the same letter are similar at the P < 0.05 level. Significantly different at the P < 0.10 level. Table 3.7. Analysis of variance for the effects and interaction of yield and cultivar on the mean number of larvae excised pot-1 harvest-1 over a 2-year period. Effect F-value P > F Cultivar Year Year*Cultivar

61 48 CHAPTER 4 EXPLORATION OF THE REPRODUCTIVE POTENTIAL AND MORPHOLOGY OF THE BERMUDAGRASS STEM MAGGOT 1 1 L.L. Baxter, D.W. Hancock, W.G. Hudson, and P.J. Moore

62 49 Abstract The bermudagrass stem maggot (BSM; Atherigona reversura Villeneuve) has infested and damaged bermudagrass [Cynodon dactylon (L.) Pers.] hayfields throughout the southeastern United States. Unfortunately, there is no information available on the reproductive potential and morphology of this exotic, invasive species. The objectives of this research were to study and describe the reproductive potential and morphology of the BSM female and determine if any differences exist in ovariole number between BSM fly populations from different regions in Georgia and Florida. Twelve flies from three locations in four different regions in Georgia and Florida were dissected to describe the reproductive morphology of the BSM female and determine if any differences existed in ovariole number among the regions. The reproductive morphology of A. reversura resembled that of A. soccata, and the oocytes were white, elongated, and approximately 1 mm in length. The total number of ovarioles varied by region, with flies from Middle Georgia having greater ovarioles than flies from any of the other regions, flies from South Georgia having more ovarioles than East Georgia, and ovariole numbers in flies from Florida being intermediary to but not different from East and South Georgia. A more thorough investigation should take place to identify differences in the regions or management techniques producers can implement that may reduce ovariole number. Introduction The bermudagrass stem maggot (BSM) was first verified in Georgia during the summer of 2010 when producers began to notice a bronzing across their bermudagrass hayfields and pastures (Hancock, 2012). The bronzed bermudagrass tips were a consequence of larva penetrating the pseudostem and feeding on the subsequent decaying plant material. To date,

63 50 BSM has been reported as far north as Kentucky (Townsend et al., 2013) and as far west as Texas (Vanessa Corriher, personal communication), although damage seems to be more severe in the hotter, more humid areas of the southeast. There is a paucity of information available on the BSM. A physical description may be found in Pont et al. (1995) but there is no literature available specific to the reproductive morphology or physiology of the BSM. However, there has been limited research on the morphology and fecundity of A. soccata, the sorghum shootfly. Unnithan (1981) describes and illustrates the internal reproductive system of the sorghum shootfly. It was found that the females possessed a variable number of polytrophic ovarioles (33.4 ± 3.9), egg maturation occurred synchronously, and most females had an unlike number of ovarioles in each ovary (Unnithan, 1981). Fecundity is a summation of multiple aspects of reproduction including rate and extent of egg production, fertility, hatchability, and characteristics of the progeny (Peters et al., 1977). Ovariole number can provide an indication of potential fecundity of a species. Many biotic and abiotic factors have been found to influence ovariole number in insects (Peters et al., 1977; Taylor et al., 2010; Bergland, 2011). The three most prevalent in available literature are environment, nutrition, and population density. Larger populations of Drosophila melanogaster have been found in more tropical regions where the climate and resource availability are more stable (Bouletreau-Merle et al., 1982). Seasonal differences can affect development of dipteran insects (Bergland, 2011). Sexual traits, such as ovariole number, are highly prone to developmental instability (Vishalakshi et al., 2008). Environmental differences such a temperature and rainfall have been shown to affect ovariole number in several species (Bouletreau-Merle et al., 1982; Taylor, et al. 2010; Bergland, 2011).

64 51 Some species, such as D. melanogaster, can respond plastically to changing environmental conditions, shutting down or starting up ovarioles depending on environmental cues. Species including D. melanogaster and D. simulans were also found to vary latitudinally (Taylor et al., 2010). The nutritional status of the progeny and the parent can influence ovariole number (Vishalakshi et al., 2008; Taylor et al. 2010). Larval nutrition affects the rate and duration of larval growth (Bergland, 2011). The size at which larval growth terminates was found to directly influence the number of ovarioles in adult Sarcophaga and Phormia flies (Bennettova et al., 1981). Bergland (2011) reported a similar trend in other dipteran insects and noted the potential consequences on future reproductive productivity. Larval nutrition has also been shown to influence the body size of the adult (Bergland, 2011). Body size creates limitations for meal size in adult dipteran insects which influences resource allocation within the body and thus the available resources for reproduction (Bergland, 2011). Population density has been shown to influence ovariole number in D. ananassae (Vishalakshi et al., 2008). However, this may be associated with poor nutrition due to the increased competition for finite resources. Field observations show the BSM population density is variable. Understanding the reproduction potential of this pest will aid in understanding populations dynamics. The objectives of this research were to study and describe the reproductive morphology of the BSM female and determine if any differences exist in ovariole number between pooled BSM fly populations from Georgia and Florida.

65 52 Materials and Methods Collection and Storage Flies were collected from 12 locations in Georgia and Florida (Table 4.1). The locations were pooled into four different regions (central GA, east GA, south GA, and north FL) based on geographical location (Fig. 4.1). Flies were collected using an insect sweep net. The sweep samples were transferred to a mesh enclosure with a small dish of damp sand (Soto et al., 1971). The enclosure was then placed on ice packs in a cooler to be transported back to Athens. Upon arrival, the enclosure was put in a freezer at -18 C for at least 30 minutes. Adult BSM flies were separated from the sweep sample and placed into glass vials filled with ethanol. The flies were stored in ethanol in a freezer until time of dissection in the UGA Entomology Laboratory. Dissection Flies were rehydrated in a phosphate buffered saline solution and dissected under a stereomicroscope (Fisher Scientific, Inc., Waltham, MA) using micro-dissection tools. Twelve flies from each of the 12 locations were dissected and the number of ovarioles in each of the paired ovaries and the number of mature eggs within the ovaries were counted to estimate the reproductive potential of the BSM. If ovarioles were not able to be counted they were determined to be either undeveloped or not present. Figure 4.2a-c illustrates examples from each of these categories of ovariole status. This data collection protocol resulted in the following response variables: status of ovarioles, number of ovarioles (total and side -1 ), and number of mature eggs (total and side -1 ).

66 53 Statistics Statistical analyses were performed for each response variable using PROC MIXED in SAS 9.1 (SAS Institute, 2001). Differences were considered significant at α = 0.05 level. Region was set as a fixed effect. Results and Discussion Reproductive Morphology of Female A. reversura The reproductive morphology of the female A. reversura is similar to that of A. soccata (Fig. 4.3; Unnithan, 1981). The paired ovaries were similar and comprised 15.6 ± 0.4 ovarioles side -1 (Fig. 4.4; P = 0.545). The total number of ovarioles female -1 ranged from 24 to 45 with a mean of 32.1 ± 3.4. Each ovariole contained oocytes in various stages of development. Egg maturation appeared to be synchronous but no more than one mature oocyte was observed per ovariole. Oocytes are white, cylindrical, and become more elongated as development progresses. Mature oocytes are small and are difficult to see without magnification (Fig. 4.5). The paired ovaries contained a similar number of mature oocytes (13.4 ± 0.5 side -1 ; P = 0.523; Fig. 4.4). The total number of mature oocytes female -1 ranged from 11 to 40 with a mean of 27.6 ± 5.7. Lateral oviducts were found at the dorsal end of the ovaries and joined into the median oviduct. The median oviduct was connected to the external ovipositor. The entire reproductive system was approximately 1 mm in length. Regional Differences in Ovariole Number Region was found to influence ovariole number in BSM females (Table 4.2; P = ). The number of total ovarioles was found to differ by region such that females from middle Georgia had a higher number of ovarioles than those from the other regions (P < ; Fig.

67 54 4.6). Flies from east Georgia were found to have the lowest number of ovarioles. However, not all regions possessed the same number of flies with ovarioles. While there were no significant differences between the regions in terms of the number of flies with ovarioles present or underdeveloped, the number of absent ovaries was not independent of the region where flies were collected (P= ; Fig. 4.7). To connect these figures, while there were fewer BSM females in middle Georgia with functioning ovarioles they tended to have a higher number of ovarioles compared to flies from the other regions where the number of flies with functioning ovaries was higher. The analysis of variance also showed that region affected the number of mature oocytes female -1 (Table 4.2; P = ). Trends in total number of mature oocytes were similar to those for total number of ovarioles. Flies from middle Georgia had the highest number of mature oocytes though not different from those flies in south Georgia and Florida (P = ; Fig. 4.6). Alternatively, flies from east Georgia were found to have the lowest number of mature oocytes yet were not different from those samples from south Georgia and Florida (P = ; Fig. 4.6). These results are not surprising given the differences in ovariole number previously discussed. Based on the literature, it was originally hypothesized that region was the primary influence on ovariole number in BSM females. A Japanese study on the BSM found that humidity improved fecundity of the females and viability of the eggs (Ikeda et al., 1991). Ideal conditions for egg viability and larval development of the closely related sorghum shoot fly is 30 C and between 60-90% humidity (Doharey et al., 1977). BSM sampled from the more humid regions of south Georgia and Florida had the highest number of females with ovarioles fully developed and present indicating more reproductive potential. While this finding was not significant at α = 0.05, the trend does follow observations of BSM damage in the field (L.L.

68 55 Baxter, unpublished data, 2013). Collecting more data points should corroborate this observation. Upon conclusion of this trial it was thought that the maturity of the forage is likely an influence on presence of the ovarioles and BSM maturity. Fields sampled in east Georgia were still in the initial lag phase of forage growth following a harvest and the majority of the females dissected from this region were found to have underdeveloped ovarioles (23 of 36 flies). In contrast, the fields in middle Georgia were very mature due to the summer s unusually wet weather. In this region, 32 of the 36 females sampled did not have ovarioles present. Previous research has suggested that some species, such as D. melanogaster, could respond plastically to changing environmental conditions, shutting down or starting up ovarioles depending upon the environmental cues (Bouletreau-Merle et al., 1982; Taylor, et al. 2010; Bergland, 2011). As the forage matures, the quality declines as more lignin and indigestible cellulose are laid down to support cell wall formation and the more upright growth of the grass. Higher lignin and cell wall concentrations translate to a thicker pseudostem, which is thought to inhibit BSM larval penetration (Baxter et al., 2013; 2014). It is hypothesized that the BSM females shut down ovarioles and/or reabsorb the oocytes since the environmental conditions are not conducive for larval development, but more work is needed to better understand this physiological process. Conclusions The reproductive morphology of A. reversura appeared to resemble that of A. soccata as described by Unnithan (1981). The oocytes were white, cylindrical, and become more elongated as development progresses. The number of total and mature ovarioles was found to differ by region such that females from middle Georgia had more total and mature ovarioles than those from the other regions and flies from east Georgia were found to have the lowest number of

69 56 ovarioles. While there were no significant differences between the regions in terms of the number of flies with ovarioles present or underdeveloped, the number of absent ovaries is not independent of the region where flies were collected. It was originally hypothesized that region was the primary influence on ovariole number in BSM females, but observations from this trial indicate that the maturity of the forage may be important. It is predicted that the BSM females shut down ovarioles and/or reabsorb the oocytes when the environmental conditions are not conducive for larval development, but more work is needed to better understand this physiological process. Further research could elucidate other factors that may be driving the regional effect seen in total ovariole number. A more thorough investigation should take place to identify differences in the regions or management techniques that producers can implement that may lower ovariole number. References Baxter, L.L., D.W. Hancock, and W.G. Hudson Bermudagrass stem maggot: An exotic pest in the Southeastern United States. Paper presented at: ASA-CSSA-SSA Annual Meetings, Tampa, FL. 4 Nov. Baxter, L.L., D.W. Hancock, W.G. Hudson, and P.J. Moore The difference in response of selected bermudagrass cultivars (Cynodon dactylon) to bermudagrass stem maggot (BSM, Atherigona reversura Villeneuve) damage. Paper presented at: American Forage and Grassland Conference, Memphis, TN. 13 Jan. Bennettova, B. and G. Fraenkel What determines the number of ovarioles in a fly ovary?. J. Insect Physiol. 27(6):

70 57 Bergland, A.O Mechanisms of nutrient-dependent reproduction in dipteran insects. In: T. Flatt and A. Heyland, editors, The Genetics and Physiology of Life History Traits and Trade-Offs. Oxford University Press, New York. p Bouletreau-Merle, J., R. Alleemand, Y.Cohet, and J.R. David Reproductive strategy in Drosophila melanogaster: Significance of a genetic divergence between temperate and tropical populations. Oecologia. 53(3): Hancock, D.W Bermudagrass stem maggot. Georgia Cattlemen. 40:20. Peters, T.M. and P. Barbosa., Influence of population density on size, fecundity, and developmental rate of insects in culture. Ann. Rev. Entomol. 22: Pont, A.C. and F.R. Magpayo Muscid shoot-flies of the Philippine Islands (Diptera: Muscidae, genus Atherigona Rondani ). Bull. Entomol. Res. Supp. Ser. 3: SAS Institute SAS/STAT user's guide, v 9.1. SAS Institute Inc., Cary, NC. Soto, P.E. and K. Laxminarayana A method for rearing the sorghum shoot fly. Journal of Econ. Entomol. 64(2): 553. Taylor, B.J. and D.W. Whitman A test of three hypotheses for ovariole number determination in the grasshopper Romalea microptera. Physiological Entomology. 35: Townsend, L. and S. Osborne Bermudagrass stem maggot found in Allen County. Kentucky Pest News No University of Kentucky, Lexington, KY. Unnithan, G. C Aspects of sorghum shootfly reproduction. Insect Sci. Application. 2(1):

71 58 Vishalakshi, C. and B.N. Singh Effect of environmental stress on fluctuating asymmetry in certain morphological traits in Drosophila ananassae: nutrition and larval crowding. Can. J. Zool. 86:

72 59 East GA Middle GA South GA FL Figure 4.1. Map of field sampling sites.

73 60 Figure 4.2. The status of each fly s ovarioles was assigned to one the following categories: present (a), absent (b), or underdeveloped (c). Figure 4.3. Reproductive system of A. reversura female.

74 61 Figure 4.4. Comparison of the mean number of ovarioles and mature oocytes side -1. Error bars illustrate LSD 0.05 values. Figure 4.5. Mature A. reversura oocyte.

75 62 Figure 4.6. Effect of region on the mean total number of ovarioles female -1 and mean total number of mature oocytes female -1 with ovarioles present. Error bars illustrate LSD 0.05 values. a b b b Figure 4.7. Effect of region on the number of flies where ovarioles were present, underdeveloped, or not present. Region did not affect the number of flies in the group where ovarioles were present or underdeveloped. Columns sharing the same letter are similar (P < 0.05).

76 63 Table 4.1. Location of field sampling sites. Producer/Farm Dean Johnson Phil Walden Phil Walden Milo Hege Jerry Martin Sim Yoder Sim Yoder Joseph Studstill James Emory Tate Abraham Baldwin Agricultural College Shenandoah Dairy Boston Farm-Santa Fe River Ranch (University of Florida) Plant Science Research and Education Unit (University of Florida) City, State Waynesboro, GA Waynesboro, GA Waynesboro, GA Millen, GA East Dublin, GA East Dublin, GA East Dublin, GA Abbeville, GA Hazlehurst, GA Tifton, GA Live Oak, FL Alachua, FL Citra, FL Latitude, Longitude , , , , , , , , , , , , , Region Assigned East GA East GA East GA East GA Middle GA Middle GA Middle GA South GA South GA South GA FL FL FL Table 4.2. Analysis of variance for the effect of region on total number of ovarioles and total number of mature oocytes. F-value P > F F-value P > F Number of Ovarioles Number of Mature Oocytes Region Region

77 64 CHAPTER 5 MORPHOLOGICAL CHARACTERISTICS OF SELECT CYNODON CULTIVARS 1 1 L. Baxter, D.W. Hancock, W.G. Hudson, W.F. Anderson, and B.M. Schwartz To be submitted to Forage and Grazinglands

78 65 Abstract Bermudagrass [Cynodon dactylon (L.) Pers.] is one of the most important and widely used forages crops for livestock and hay production in the Southeast US. While there has been extensive research comparing yields among bermudagrass cultivars, there is limited information on the morphological differences between them. The objective of this study was to compare selected morphological characteristics among popular Cynodon cultivars presently grown in the southeast. Eight morphological characteristics were analyzed: number of tillers pot -1, leaf color, tiller diameter, leaf width, canopy height, internode length, total dry biomass pot -1, and weight tiller -1. The C. dactylon cultivars tended to have denser forage canopies with lighter green, finer tillers. Alternatively, the cultivars with stargrass (C. nlemfuensis Vanderyst) influence were more open-canopied and possessed darker, more robust leaves and tillers. This summary will benefit researchers and producers alike who are interested in selecting a cultivar based on canopy characteristics. Introduction Most pastures in the southeastern USA are planted in improved, warm season forage cultivars (Redfearn et al., 2003). Cynodon species are deeper rooted and more drought tolerant than most other warm season grasses (Redfearn et al., 2003). They vary widely in distribution, agronomic value, and economic importance (Taliaferro et al., 2004). These persistent, perennial, sod-forming grasses are good for pasture and hay production, turfgrass, soil conservation, and spoil site remediation (Ball et al., 2007; Taliaferro et al., 2004). Bermudagrass (C. dactylon) has developed into one of the most important forage crops for livestock in the United States (Taliaferro et al., 2004). Bermudagrass is estimated to be grown on million hectares in the United States for forage purposes (Taliaferro et al., 2004). This

79 66 perennial, warm season forage is frequently used for hay and pasture production throughout the southeastern US (Ball et al., 2007). The sod-forming grass spreads by stolons and rhizomes, generating high yields during the hot and humid summers. Although differences exist among the cultivars, bermudagrass tends to be tolerant of drought, saline soils, and temporary flooding (Skerman et al., 1990). Stargrass (C. nlemfuensis) is another warm season, perennial forage that is similar to bermudagrass but grows mostly in the subtropical regions of Florida (Taliaferro et al., 2004). While bermudagrass is finer and rhizomatous, stargrass tends to be more robust and lacks rhizomes (Skerman et al., 1990; Taliaferro et al., 2004). Most Cynodon grasses are native to East Africa, but stargrass was restricted to the more tropical African regions (Taliaferro et al., 2004). Henry Ellis, the second royal governor of Georgia, first introduced bermudagrass into North America in 1751 and the first documented use of stargrass was in 1937 by G.W. Burton (Taliaferro et al., 2004). Many researchers have collected germplasm from these grasses to use for the development and subsequent improvement of bermudagrass cultivars grown in the southeastern USA today. Cynodon grasses may be divided into two different groups: seeded and vegetatively propagated. The majority of the forage bermudagrass cultivars that have been released must be vegetatively propagated because they either heterozygous, produce too little viable seed, or are sterile as a result of the interspecific hybridization of C. dactylon and C. nlemfuensis (Redfearn et al., 2003; Sleper et al., 2006; Taliaferro et al., 2004). Although bermudagrass was once known as a ubiquitous, cosmopolitan weed, several improved bermudagrass cultivars have been developed in the past several decades that are higher yielding and tend to have superior nutritive value than their predecessors (Harlan and de Wet, 1969). Coastal was the first improved bermudagrass cultivar to be released (Taliaferro et al.,

80 ). Increasing yield potential has been the primary objective when developing bermudagrass cultivars, but the improvement of winter hardiness, quality, and disease resistance has also been important. Bermudagrass was generally restricted to temperate zones with mild winters until new cultivars selected for tolerance to frost and colder temperatures pushed the adaptation area northward (Ball et al., 2007; Redfearn et al., 2003). Since bermudagrass is such an important component for SE livestock production, selections have been made to improve digestibility of these warm-season forages to ultimately improve quality (Redfearn et al., 2003). Bipolaris (formerly Helminthosporium) leaf spot is a major fungal disease for southern forages, especially in potassium-stressed bermudagrass. Resistance to pathogens is now a selection criterion for improved cultivars (Redfearn et al., 2003; Taliaferro et al., 2004). Following recent developments, selecting for or developing resistance to the bermudagrass stem maggot (BSM; Atherigona reversura) could possibly be added to the list. Table 5.1 summarizes the cultivar recommendations for southeastern states where improved bermudagrass cultivars are commonly grown for forage. While there has been extensive research comparing yields among bermudagrass cultivars, there is limited information on the morphological differences among them. Bermudagrass is highly variable but tends to be fine-leaved and rhizomatous with leaves, culm, and stolons that are rarely as large as stargrass (Skerman et al., 1990; Taliaferro et al., 2004). Stargrass tends to be more robust with stiffer, coarser foliage (Taliaferro et al., 2004). Distinguishing bermudagrass and stargrass within a field is possible, but differentiating between cultivars in the field is exceptionally difficult in the absence of a detailed morphological key. Genetic testing can sometimes be employed to distinguish the cultivars from one another, but a morphological key could help producers, extension personnel, and service providers to identify some of the individual cultivars more

81 68 easily. Therefore, the objective of this study was to compare key morphological characteristics among the Cynodon cultivars most commonly grown in the southeastern USA. Materials and Methods Cultivar Selection Eight of the most common and important Cynodon cultivars were selected for assessment. These were also chosen to represent a range of canopy densities from very dense to relatively open. Selected cultivars included four C. dactylon cultivars (common [ecotype: Tifton, GA], Coastal, Alicia, and Russell ), two C. nlemfuensis cultivars ( Tifton 68 and stargrass), and two C. dactylon x C. nlemfuensis hybrids ( Tifton 85 and Coastcross-II ). All cultivars were selected from the USDA-Agricultural Research Service s Cynodon germplasm nursery at the Coastal Plain Experiment Station in Tifton, GA. Forage Management Bermudagrass sprigs were transplanted from the field and planted into 15.5 cm diameter pots with Fafard 3B Potting Mix (Sun Gro Horticulture, Agawam, MA) before they were transported to the Crop and Soil Science Greenhouses in Athens, GA. After transport, all pots were fertilized with a greenhouse fertilizer at a rate equivalent to 2.2 kg N ha -1 d grams of fertilizer was dissolved into 15 liters of water and ~120 ml of the solution was hand applied to each unit. Each unit was topped with a clear acetate and mesh cylindrical enclosure (Fig. 3.1a). Durable, clear acetate sheets (70 cm x 43 cm) were coiled to make a tube. A 6-cm strip of fine mesh fabric was glued on to each end to attach them together to form the cylinder, and a 20-cm

82 69 circle cut from the mesh was glued on to enclose the top. The clear structure allowed light to reach the plant and the mesh permitted airflow to prevent excess moisture buildup on the acetate and the foliage. The enclosures ensured that forage was restricted to the unit s growing area. A micro-irrigation system was installed to supply water directly to the soil surface. The irrigation line was controlled by a battery-operated, programmable timer set to deliver water for at least 2 minutes at 12-hour intervals. The duration was adjusted as needed to regulate soil moisture. The grasses were allowed a one month acclamation period before the trial began. Experimental Design and Project Timeline The experiment was a randomized complete block design. Four runs were completed during the 2012 growing season (7/2/2012 to 10/25/2012) but, as a result of unusual weather, only three runs were completed in 2013 (7/8/2013 to 9/30/2013). Runs consisted of 4-wk growth periods, simulating a recommended bermudagrass hay cutting schedule (Fig. 3.2). At the beginning of each run, all units were clipped to a 5-cm stubble height. Data were averaged across runs within the respective year unless the difference between years was one of magnitude, then the data were averaged among all seven runs. Canopy Evaluation At the conclusion of the growing period, all units were moved from the greenhouse and brought into a neighboring laboratory for evaluation. First, canopy height (± 1 cm) and number of tillers were recorded. A SPAD meter (Konica Minolta, Ramsey, NJ) was used to record and calculate the average measure of color (to an accuracy of ± 1.0 SPAD units) of five randomly selected leaves. Then, five tillers were randomly selected and clipped off at the soil surface. A micrometer was used to measure the width of the leaf blade that emerged from the second node

83 70 below the terminal end, as well as the width of the pseudostem at this node (both measurements recorded to an accuracy of ± 0.01 mm). Measurements were taken at the second node for consistency. During the first year of this study internode length was also evaluated by measuring the distance between the five nodes from the basal end on each of the five tillers to an accuracy of ± 0.01 mm. The remaining sample was then hand clipped to 5-cm and, including the five tillers previously harvested, the green weight of the harvested sample was then recorded to an accuracy of ± 0.01 g. Samples were dried at 60 C for 48 hr to correct for moisture content. This data collection protocol resulted in the following response variables: number of tillers pot -1, leaf color, tiller diameter, leaf width, canopy height, internode length, and total dry biomass pot -1. These variables were also used to calculate weight tiller -1 (Eq. [3.2]). Statistics Statistical analyses were performed for each response variable using PROC MIXED in SAS 9.1 (SAS Institute, 2001). Differences were considered significant at α = 0.05 level. Cultivar and year were designated as fixed effects, while replication was set as a random effect. Results and Discussion Figure 5.1 illustrates the canopies of the eight cultivars selected for this study. Table 5.2 contains the ANOVA analyses for the phenotypic plant characteristics examined in the study. In most response variables, there was an interaction between year and cultivar. These interactions were generally a result of magnitude differences observed in the two growing seasons. Consequently, data are pooled across the two years.

84 71 Leaf Morphology The cultivars with stargrass influence had significantly darker leaves than the four C. dactylon cultivars. Leaves from Stargrass and Tifton 68 were similar and measured darker than those of Tifton 85 and Coastcross-II (Table 5.3). There was no difference (P > 0.05) between the four C. dactylon cultivars (Table 5.3). Darker leaves can indicate more photosynthetic potential and nitrogen content, potentially translating to higher yields (Taiz et al., 2006). Stargrass had the widest leaves in this study, followed by Tifton 68 (Table 5.3). The four C. dactylon cultivars had the narrowest leaves with no significant difference between them (Table 5.3). Tifton 85 and Coastcross-II had comparable leaf widths and fell intermediate to these two groups (Table 5.3). Tiller Morphology and Number Stargrass and Tifton 68 had the largest diameters, followed by Tifton 85 (Table 5.3). Surprisingly, Coastcross-II and Russell followed next and were not significantly different from each other (Table 5.3). Presumably, this is an artifact of the greenhouse growing conditions since the two cultivars are not genetically similar (Anderson et al., 2009). Common bermudagrass had the highest number of tillers pot -1 in this study, followed by Alicia and Russell (Table 5.3). Stargrass was found to have the lowest number of tillers pot -1 (Table 5.3). The other cultivars fell intermediary, including Coastal, which had a lower number of tillers than expected based on field observations. In general, C. dactylon cultivars tended to have a higher number of tillers than those with C. nlemfuensis influence. The outlier in this study is Coastal as it had a lower number of tillers than expected based on field observations. Otherwise, the trend followed what was expected from the available literature. In general, the

85 72 number of tillers pot -1 was inversely proportional to the weight tiller -1 (Fig. 5.2; Table 5.3). Stargrass had the highest weight tiller -1 while Alicia and Coastal had the lowest (Table 5.3). Canopy Height and Internode Length Coastcross-II had the tallest canopy height, while Stargrass had the shortest in this study (Table 5.3). There was not a substantial difference in the mean height of the cultivars as was expected based on field observations. This is most likely an artifact of the grass being grown inside the greenhouse. This restriction forced the tillers to branch and expand vertically rather than along the soil surface as it would in the field. Based on the results in Table 5.3, trends in internode length appear to follow those for weight tiller -1 in that those cultivars with more robust tillers (i.e. stargrass) tend to have a larger distance between nodes than the finer-textured cultivars (i.e. Alicia, common, and Russell ). There were few statistically important differences for this characteristic, which may either be a consequence of only one year of data collection or a lack of genetic variation for internode length in these selected cultivars. Conclusions There is a substantial amount of phenotypic variation in Cynodon cultivars that are currently used in the southeastern USA. Cynodon dactylon cultivars tended to have denser forage canopies with a higher number of tillers and lighter green, finer-textured, narrower leaves. In contrast, C. nlemfuensis cultivars and hybrids were more open-canopied with fewer yet darker, coarser, and more robust leaves and tillers. Since a detailed discussion of popular Cynodon cultivars was previously unavailable, this summary will serve as a morphological key to assist producers, extension personnel, and service providers in identifying some of the individual Cynodon cultivars.

86 73 References Anderson, W.F., Maas, A., Ozias-Akins, P Genetic variability of a forage bermudagrass core collection. Crop Sci. 49: Ball, D Varieties of bermudagrass. ANR Alabama Cooperative Extension System, Auburn. Ball, D.M., C.S. Hoveland, and G.D. Lacefield Southern Forages. 4 th ed. Potash and Phosphate Institute, Norcross. Hancock, D.W., N.R. Edwards, T.W. Green, and D.M. Rehberg Selecting a forage bermudagrass variety. Cic. 919.Univ. of Georgia Cooperative Extension, Athens. Harlan, J.R., and J.M.J. de Wet Sources of variation in Cynodon dactylon (L.) Pers. Crop Sci. 9: Lang, D. and M.L. Broome Forage: Bermudagrass. Information Sheet 820. Mississippi State University Extension Service, Starkville. Newman, Y.C., J.M.B. Vendramini and F. A. Johnson Bermudagrass production in Florida. SS-AGR-60. University of Florida Extension, Gainesville. Redfearn, D.D. and C.J. Nelson Grasses for Southern Areas. In: R.F. Barnes, C.J. Nelson, M. Collins, and K.J. Moore. Forages: An Introduction to Grassland Agriculture. 6th edition, Vol. 1. Iowa State University Press, Ames. p SAS Institute SAS/STAT user's guide, v 9.1. SAS Institute Inc., Cary, NC. Skerman, P.J. and F. Riveros Tropical Grasses-Food and Agriculture Organization of the United Nations. No. 23. FAO, Rome. p Sleper, D.A. and J.M. Poehlman Breeding Forage Crops. 5 th ed. Blackwell Publishing, Ames.

87 74 Taiz, L. and E. Zeiger Plant Physiology. 4 th ed. Sinauer Associates, Inc., Sunderland. Taliaferro, C.M., F.M. Rouquette, and P.Mislevy Bermudagrass and stargrass. In: L.E. Moser, B.L. Burson, and L.E. Sollenberger, editors, Warm-season (C4) grasses, Agronomy Monograph No. 45. American Society of Agronomy, Madison. p Tidwell, E.K Bermudagrass: Varieties for hay and pasture in Louisiana. Pub Louisiana State University Ag Center, Baton Rouge. Undersander, D.J. and B. W. Pinkerton Cultivars of bermudagrass. Forage Leaflet 4. Clemson University, Clemson.

88 Figure 5.1. Canopy comparison of the cultivars selected for this study. 75

89 76 Mean Number of Tillers Pot -1 Mean Weight Tiller -1 Figure 5.2. Relationship between mean number of tillers pot -1 and mean weight tiller -1 (g). Error bars illustrate LSD 0.05 values. Table 5.1. Summary of bermudagrass cultivar recommendations for the SE US. State Cultivars Recommended Source Alabama Coastal, Midland, Russell, Tifton 44, Georgia Florida Louisiana Tifton 78, Tifton 85 Coastal, Tifton 44, Russell, Tifton 85 Callie, Coastal, Coastcross-I, Florakirk, Jiggs, Suwannee, Tifton 44, Tifton 78, Tifton 85 Alicia, Brazos, Coastal, Common, Grazer, Russell, Tifton 44, Tifton 85 Ball, 2002 Hancock et al., 2010 Newman et al., 2013 Tidwell, 2010 Mississippi Alicia, Coastal, Callie, Tifton 44 Lang, 2011 South Carolina Coastal, Tifton 44 Undersander et al., 1988

90 77 Table 5.2. Analysis of variance for the effects and interaction of year and cultivar on various phenotypic plant characteristics during a 2-year period. F-value P > F F-value P > F F-value P > F Cultivar Year Year*Cultivar Leaf Color < Leaf Width < < < Tiller Diameter 0.01 < Number of Tillers < < <.0001 Weight Tiller < < Canopy Height < Length Between Ground n/a n/a n/a n/a and First Node Length Between First and 8.11 < n/a n/a n/a n/a Second Nodes Length Between Second n/a n/a n/a n/a and Third Nodes Length Between Third n/a n/a n/a n/a and Fourth Nodes Length Between Fourth and Fifth Nodes n/a n/a n/a n/a N/A indicates data only collected during year one of this study.

91 78 Table 5.3. Effect of cultivar on phenotypic plant characteristics. Mean Leaf Color (SPAD units) Mean Leaf Width (mm) Mean Tiller Diameter (mm) Mean Number of Tillers Pot -1 Mean Weight Tiller -1 (g) Canopy Height (cm) Ground and 1st (mm) Mean Length Between Nodes st and 2nd (mm) 2nd and 3rd (mm) 3rd and 4th (mm) 4th and 5th (mm) Alicia c 2.08 d 0.84 d b 0.07 de de c d ab d bc Coastal c 2.09 d 0.81 d cd 0.07 cde ab a ab ab ab bc Coastcross-II b 3.35 c 1.17 c bc 0.10 bcd a ab a a a a Common c 1.86 d 0.74 d a 0.05 c de bc bc b abc c Russell c 2.21 d 1.04 c b 0.08 cde abc abc cd b cd c Stargrass a 4.49 a 1.60 a d 0.16 a e bc bc ab bcd bc Tifton a 4.11 b 1.62 a cd 0.12 ab cde ab ab ab abcd bc Tifton b 3.67 c 1.34 b bc 0.10 bc bcd ab a a abc ab Mean SE LSD LSD Means followed by the same letter are similar at the P < 0.05 level.

92 79 CHAPTER 6 CONCLUSIONS, IMPLICATIONS, AND FUTURE RESEARCH DIRECTIONS The bermudagrass stem maggot (BSM; Atherigona reversura Villeneuve) was first reported in Georgia during the summer of 2010 and quickly spread throughout the southeast, causing significant damage to bermudagrass [Cynodon dactylon (L.) Pers.] hayfields and pastures. Prior to this project, there was very little information on the BSM or its control. Though previous research and field observations provided good preliminary guidance, a more detailed understanding of the effects of the BSM on yield, yield components, and crop morphology is needed, as well as evaluations of newer C. dactylon cultivars and C. dactylon x C. nlemfuensis Vanderyst hybrids. Bermudagrass forage producers needed to know if there are cultivars that are more or less susceptible to the BSM in order to employ that knowledge as part of an integrated pest management (IPM) strategy. The objectives of this research were to compare the severity of damage among selected cultivars, quantify the phenotypic variation in cultivar response to the BSM, and assess the fecundity of the BSM on selected bermudagrass cultivars. By introducing flies to eight different cultivars and comparing the results to their respective controls it was determined that the number and percent of tillers damaged by the BSM depends on cultivar although tiller diameter and number of tillers appear to be the phenotypic characteristics most indicative of tolerance to BSM damage. When the BSM is present, there is an increase in tiller diameter and leaf greenness and a decrease in the number of tillers, though no apparent change in

93 80 the average weight stem -1, leaf width, canopy height, or internode length. The results showed that some cultivars may serve as better host for the BSM larva; however, more research is needed to quantify these results. Since the attempt to determine fecundity of the BSM was not clear from the first study, a different approach was taken. Literature reported that population density and region can affect ovariole number in some dipteran insects. Field observations suggest the BSM population density is variable. Understanding the reproduction potential of this pest will aid in understanding populations dynamics. The objectives were to study and describe the reproductive morphology of the BSM female and determine if any differences exist in ovariole number between pooled BSM fly populations from Georgia and Florida. After dissecting 36 female BSM from 4 different regions it was determined that the paired ovaries were similar and the total number of ovarioles female -1 ranged from 24 to 45 with a mean of 32.2 ± 3.5. The total number of ovarioles varied by region, with flies from middle Georgia having greater ovarioles than flies from any of the other regions, flies from south Georgia having more ovarioles than east Georgia, and ovariole numbers in flies from Florida being intermediary to East and South Georgia. While it was originally hypothesized that region was the primary factor driving differences in ovariole number it now appears as if forage quality may be a major influence. It is hypothesized that the BSM females shut down ovarioles and/or reabsorb the oocytes when conditions are not conducive for larval development, but more work is needed to better understand this physiological process. After analyzing the canopy characteristics for the first trial it became evident that bermudagrass is highly variable. The objective of this final study was to compare selected morphological characteristics among popular Cynodon cultivars grown today in the southeast.

94 81 Eight morphological characteristics were analyzed: number of tillers pot -1, leaf color, tiller diameter, leaf width, canopy height, internode length, total dry biomass pot -1, and weight tiller -1. C. dactylon cultivars tended to have denser forage canopies with lighter green, finer-textured leaves and tillers while C. nlemfuensis cultivars and hybrids were more open-canopied with fewer yet darker and more robust leaves and tillers. The research in all three of these projects will be instrumental in laying the groundwork for future research endeavors as well as providing better guidance for bermudagrass producers. By comparing BSM damage among the selected cultivars it is clear that economic thresholds should be assessed in the field using a fine-textured cultivar (i.e. common, Alicia, Coastal, or Russell ) and a coarser, more tolerant control (i.e. Tifton 85 or Coastcross II ). Assuming the cultivar is adapted to the site; producers should choose coarse-stemmed cultivars (e.g., Tifton 85 ) that are not as severely damaged by the BSM, particularly in areas where BSM fecundity seems to be highest. The preliminary work on reproductive morphology will serve not only as a foundation for understanding the pest s biology but also for estimating the population and damage potential of the BSM. Since a detailed discussion of popular Cynodon cultivars was previously unavailable, this summary will serve as a morphological key to assist producers, extension personnel, and service providers in identifying some of the individual Cynodon cultivars. Despite the many findings in these projects, there are many questions which remain unanswered about the bermudagrass stem maggot. The next step should be to better understand the life cycle of this species. This will enable researchers to develop appropriate and economical suppression efforts for producers. Other endeavors should include quantifying the economic impact of the damage, determine which plant characteristics are most strongly correlated to BSM

95 82 damage (preliminary data found in Tables A.1 and A.2), and elucidating any factors that may be driving the regional effect seen in total ovariole number. It is important in any IPM program to have a damage threshold to make economically sound decisions for when to treat. A more thorough investigation should take place to identify differences in the regions or management techniques so that producers can implement those found to lower ovariole number as another IPM strategy. Based on the results of this study, tolerance appears to be the best IPM strategy for producers to implement for BSM control.

96 83 APPENDIX SELECT TABLES

97 84 (a) (b) (c) (d) (e) Figure A.1. Comparison of mean values of phenotypic plant characteritics and mean percent of infected tillers pot -1 over a 2-year period. Phenotypic characteristcs include leaf color (a), leaf width (b), tiller diameter (c), number of tillers (d), weight tiller -1 (e), and canopy height (f). Error bars illustrate LSD 0.05 values. Mean separation on the line is found in Fig (f)

98 85 (a) (b) (c) (d) (e) Figure A.2. Comparison of mean length between nodes and mean percent of infected tillers pot -1 during the first year of this trial. Data represented include between ground and first (a), first and second (b), second and third (c), third and fourth (d), and fourth and fifth nodes (e). Error bars illustrate LSD 0.05 values. Mean separation on the line is found in Fig. 3.3.