Beaumont Site Visit: Mexican Rice Borer and Sugarcane Borer Sugarcane and Rice Research

Transcription

1 Beaumont Site Visit: Mexican Rice Borer and Sugarcane Borer Sugarcane and Rice Research Project Investigators: Post-Doctoral Researcher: Research Associate: Graduate Assistant: Gene Reagan, LSU AgCenter M.O. Way, Texas A&M AgriLife Research Julien Beuzelin Blake Wilson Matt VanWeelden Cooperators: Texas A&M AgriLife Research & Extension Ctr., Beaumont Ted Wilson, Professor and Center Director Lee Tarpley, Associate Professor Yubin Yang, Senior Systems Analyst Fugen Dou, Assistant Professor Mark Nunez and Rebecca Pearson, Research Associates USDA ARS Bill White, Research Scientist (Sugarcane Research Station at Houma, LA) Allan Showler, Research Scientist (Kika de la Garza Research Station at Weslaco, TX) LSU AgCenter Natalie Hummel, Associate Professor, Extension Entomology Louisiana Dept. of Agriculture and Forestry Tad Hardy, State Entomologist American Sugarcane League Rio Grande Valley Sugar Growers, Inc. 28 September, 2011 This work has been supported by grants from the USDA NIFA, Southern Region IPM, Crops at Risk IPM, NRI AFRI Sustainable Bioenergy, and US EPA Strategic Agricultural Initiative programs. We also thank the Texas Rice Research Foundation, the American Sugar Cane League and Rio Grande Valley Sugar Growers Inc, participating Agricultural Chemical Companies, the Texas Department of Agriculture and the Louisiana Department of Agriculture and Forestry for their support.

2 Comparison of Stem Borers Attacking Sugarcane and Rice (a) Adult female sugarcane borer (c) Adult female Mexican rice borer (e) Adult female rice stalk borer (b) Sugarcane borer larva (d) Mexican rice borer larva (f) Rice stalk borer larva Photos: (a) B. Castro; (b) J. Saichuk; (c) F. Reay-Jones; (d)(e)(f) A. Mészáros 2

3 TABLE OF CONTENTS Comparison of Stem Borers Attacking Sugarcane and Rice... 2 Table of Contents... 3 Field Research Site Visit Announcement... 4 Ten Years of Stem Borer Research Collaboration on Sugarcane and Rice... 5 Monitoring Mexican Rice Borer Movement: Range Expansion into Louisiana... 7 Evaluation of Commercial and Experimental Sugarcane Cultivars for Resistance to the Mexican Rice Borer, Beaumont, TX, 2010 and Feeding Behavior and Duration of Exposure of Mexican Rice Borer Larvae on Sugarcane Red Imported Fire Ant Predation on Mexican Rice Borer in Sugarcane at Beaumont, TX in Pheromone Trap Assisted Scouting and Aerial Insecticidal Control of the Mexican Rice Borer, 2009 and Comparison of Mexican Rice Borer Pest Pressure in Bioenergy and Conventional Sugarcane Small Plot Assessment of Insecticides Against the Sugarcane Borer, Field Assessment of Novaluron for Sugarcane Borer, Diatraea saccharalis (F.) (Lepidoptera: Crambidae), Management in Louisiana Sugarcane Seasonal Infestations of Two Stem Borers (Lepidoptera: Crambidae) in Non-Crop Grasses of Gulf Coast Rice Agroecosystems Harvest Cutting Height and Ratoon Crop Effects on Stem Borer Infestations in Rice Trapping for Mexican Rice Borer in the Texas Rice Belt, Rice Insecticide Evaluation Studies Beaumont Sugarcane and Energycane Variety Test, Beaumont Sugarcane and Energycane Variety Test, Example Data Sheet

4 2011 Field Research Site Visit Announcement To: Louisiana and Texas Sugarcane and Rice Consultants, Agricultural Extension Agents, and Industry Cooperators From: Gene Reagan and Mo Way LSU AgCenter and Texas A&M Entomologists Re: Texas AgriLife Research and Extension Center at Beaumont Mexican Rice Borer and Sugarcane Borer Field Research Observations LOCATION Please do not take any live insects from this location! Texas AgriLife Research and ITINERARY Extension Center at Beaumont 1509 Aggie Drive, Beaumont, TX Tuesday, 27 September 6:15 pm Wednesday, 28 September 8:00 am Meet in lobby of Holiday Inn and Suites to go to dinner probably at Papadeaux s (optional) Meet in front of Texas AgriLife Research and Extension Center: - Dr. Ted Wilson (Center Director): Welcome and introduction - Dr. Gene Reagan: Overview of planned activities, handouts, and instructions to go to the field ACTIVITIES 1. Tad Hardy (LA State Entomologist): Review of LA Dept. Ag & Forestry MRB pheromone trap monitoring 2. Dr. Bill White: Variety diversity in the test 3. Dr. Gene Reagan, Dr. Julien Beuzelin, and Mr. Blake Wilson: Hands-on sampling for Mexican rice borer ( MRB) and sugarcane borer (SCB) injury in sugarcane varieties 4. Observe MRB and SCB larvae in replicated test of LA sugarcane varieties (HoCP , Ho , L , HoCP , L 07-57,Ho , Ho , L , Ho , HoCP , L , HoL , Ho , HoCP , Ho , L , L08-075, HoCP , Ho ) 5. Mr. Blake Wilson: Use of MRB pheromone traps to help with scouting. 6. Dr. Julien Beuzelin and Mr. Matt VanWeelden: Multi-crop bioenergy research. 7. Dr. Mo Way: Observe MRB and SCB damage and discuss insecticides and cultural practices in rice or visit demonstration of sugarcane stalk splitter machine (Gene Reagan). Wednesday, 28 September 11:00 am Wednesday, 28 September Noon Sun grant/chevron/beaumont energy cane and high biomass sorghum research near main building, Texas AgriLife Research and Center at Beaumont, 1509 Aggie Dr., approx. 9 miles west of Beaumont on Hwy 90. Adjourn and return home RESERVATION AND HOTEL INFORMATION For hotel reservations call Any time prior to Tuesday, 20 September HOTEL ADDRESS: Holiday Inn and Suites 3950 I-10 South Beaumont, TX (hotel) (fax) Reservation Code: LSU Entomology You may reserve rooms with Samantha by at: samantha.richards@jqh.com $ tax reduced rate, Breakfast buffet (6:00 AM) included DIRECTIONS TO RESEARCH SITE: 9.5 miles west of Beaumont on Hwy 90, ~ 1 mile north on Aggie Drive 4

5 TEN YEARS OF STEM BORER RESEARCH COLLABORATION ON SUGARCANE AND RICE Gene Reagan and M.O. Way * LSU AgCenter and Texas A&M AgriLife Research The Mexican rice borer (MRB), Eoreuma loftini (Dyar), is the most destructive insect pest of sugarcane in North America. This invasive alien species entered the Lower Rio Grande Valley (LRGV) of Texas in 1980, and quickly caused such severe losses that sugarcane farmers were unable to harvest some of their fields. MRB continued to expand its geographical range throughout the Texas Gulf Coast rice producing area and into western Louisiana, now infesting rice in all of Calcasieu Parish (Lake Charles area). Causing as much as 50% yield loss in commercial Texas rice fields, projected economic loss to the Louisiana sugarcane and rice industries may be expected to reach as much as $220 million (sugarcane) and $45 million (rice) annually when MRB becomes fully established. The MRB is also a serious pest of sorghum and corn in Texas. The sugarcane borer (SCB), Diatraea saccharalis (F.), continues to be a serious pest of sugarcane in Louisiana and is also a key pest of rice and non-transgenic corn. MRB was first discovered in the Texas rice belt in 1988 and soon received attention from producer organizations and support industries. At this time, little was known about the biology and ecology of the pest, and even less was known about possible management techniques. After a several million dollar biological control program proved unsuccessful with LRGV sugarcane growers, we knew that control efforts would have to be much more comprehensive. This would require a far greater knowledge of MRB biology and how its life history relates to different host plants. On behalf of the LSU AgCenter and Texas A&M University, we initiated a national competitive grant effort in 2001 starting with $40,000 seed money from the USDA (CSREES) Critical Issues program and a project titled MRB identification of range and variety resistance assessment. In addition to three multi-year Strategic Agricultural Initiative grants from the US Environmental Protection Agency, we were successful in obtaining five years of support from two USDA Crops-at-Risk grants. All of these grants have been oriented toward building a system for sugarcane and rice that would not only help to manage stem borer problems, but also reduce area-wide pest populations. This year, our team was expanded to include L.T. Wilson and Yubin Yang (Texas A&M AgriLife), Allan Showler (USDA-ARS), and Jeff Hoy (LSU AgCenter Plant Pathology) for a sustainable biomass energy grant to further mitigate insect and disease pressures on conventional crops in interaction with potentially emerging bioenergy cropping systems. * Thomas E. (Gene) Reagan, Austin C. Thompson Endowed Professor of Entomology, Louisiana State University Agricultural Center; and M.O. Way, Professor of Entomology, Texas A&M AgriLife Research and Extension Center at Beaumont 5

6 During the ten years of our collaborative work, we have developed sampling approaches to monitor infestations and quantify pest populations, identified resistant varieties, and evaluated and helped label environmentally friendly insecticides. With colleagues, we have studied numerous plant-insect interactions involving crop and non-crop host preferences, and better defined the role of plant stress impacted by cultural practices, salt, water and nutrients. Techniques reducing scouting efforts and achieving better insecticide application timing were also developed to assist sugarcane consultants. With recently labeled insecticides having four different modes of action (Confirm, Diamond, Coragen / Belt, Besiege ), the potential for insecticide resistance is also reduced. In rice, a newly developed seed treatment, Dermacor X- 100, impacts stem borer management in addition to pyrethroid foliar applications. Thank you for participating in the 10 th stem borer research site visit training. We welcome you to the 2011 Beaumont Site Visit and hope you depart with good information to help you grow a more profitable crop. 6

7 MONITORING MEXICAN RICE BORER MOVEMENT: RANGE EXPANSION INTO LOUISIANA T. Hardy 1, T.E. Reagan 2, M.O. Way 3, R.A. Pearson 3, B.E. Wilson 2, and J.M. Beuzelin 2 1 Louisiana Department of Agriculture and Forestry; 2 Department of Entomology, LSU AgCenter 3 Texas A&M AgriLife Research and Extension Center at Beaumont Cooperative studies on the Mexican rice borer (MRB), Eoreuma loftini, between the LSU AgCenter, Texas A&M University AgriLife Research Center at Beaumont, the Texas Department of Agriculture, and the Louisiana Department of Agriculture and Forestry have been on-going since 1999 to monitor the movement of this devastating pest of sugarcane towards Louisiana. As previously anticipated, MRB spread into Louisiana by the end of 2008, and was collected in two traps near rice fields northwest of Vinton, LA on December 12. While no MRB specimens were detected in Louisiana in 2009, data from 2010 showed that this invasive pest had expanded its range into Cameron and Calcasieu parishes. Additional MRB moths captured in 2011 indicate the species has expanded its range farther north into southeastern Beauregard Parish. The first specimens trapped since 2008 were collected in non-crop habitat with wild grass hosts 6.8 miles southeast of Vinton, Calcasieu parish, LA, on 22 November Since that date, numerous specimens have been collected in traps from 36 different locations in Louisiana (Table 1, Fig. 1). Currently, the locations of positive traps have been in rice or wild-host areas; however, the eastern-most location is directly south of Lacassine, and it is anticipated the MRB will soon infest producing sugarcane in that region. More than 200 MRB have been trapped in Calcasieu parish so far in 2011 (Table 1), indicating the species has established a clear presence. Additionally, rice growers in this parish have begun to report MRB larval infestations in their fields. In August, traps were retrieved and/or re-deployed east of their previous locations in an attempt to stay ahead of the eastern MRB movement (Table 2). Continued monitoring of MRB populations will be conducted with additional traps at locations further east and north. Currently, LDAF has a total of 25 MRB pheromone traps in Calcasieu, Cameron and Jefferson Davis parishes, with 3 additional traps in Beauregard and Vermilion parishes. In late September, 12 traps will be added in St. Mary and Iberia parishes near sugarcane processing and off-loading facilities. As the pest s eastward expansion continues, effective management strategies such as the use of varietal resistance, improved chemical control tactics, and management of non-crop hosts are becoming critical to slow the spread of this devastating insect. Table Louisiana MRB Trap Captures Parish # Sites # + Sites # MRB Calcasieu Cameron Beauregard Jefferson Davis Table 2. Monthly Total MRB Captures in LA Month # MRB March 36 April 59 May 36 June 57 July 19 August 32 7

8 Fig. 1. Monitoring MRB movement in Louisiana, 2010 and Stars designate MRB positive trap locations. Two positive sites in Southwestern Cameron Parish are not shown. References: Hummel, N.A., T. Hardy, T.E. Reagan, D.K. Pollet, C.E. Carlton, M.J. Stout, J.M. Beuzelin, W. Akbar, W.H. White Monitoring and first discovery of the Mexican rice borer Eoreuma loftini (Lepidoptera: Crambidae) in Louisiana. Fla. Entomol. 93: Hummel, N., G. Reagan, D. Pollet, W. Akbar, J. Beuzelin, C. Carlton, J. Saichuk, T. Hardy, M. Way Mexican Rice Borer, Eoreuma loftini (Dyar). LSU AgCenter Pub Reay-Jones, F.P.F., L.T. Wilson, M.O. Way, T.E. Reagan, C.E. Carlton Movement of the Mexican rice borer (Lepidoptera: Crambidae) through the Texas rice belt. J. Econ. Entomol. 100: Reay-Jones, F.P.F., L.T. Wilson, T.E. Reagan, B.L. Legendre, and M.O. Way Predicting economic losses from the continued spread of the Mexican rice borer (Lepidoptera: Crambidae). J. Econ. Entomol. 101:

9 EVALUATION OF COMMERCIAL AND EXPERIMENTAL SUGARCANE CULTIVARS FOR RESISTANCE TO THE MEXICAN RICE BORER, BEAUMONT, TX, 2010 AND 2011 T.E. Reagan 1, B.E. Wilson 1, J.M. Beuzelin 1, W.H. White 2, M.O. Way 3, M. VanWeelden 1, and A.T. Showler 4 1 Department of Entomology, LSU AgCenter 2 USDA Sugarcane Research Unit at Houma, Louisiana 3 Texas A&M AgriLife Research and Extension Center at Beaumont, Texas 4 USDA-ARS, Kika de la Garza Agricultural Research Center at Weslaco, Texas Because of the limitations of chemical and biological control against the Mexican rice borer (MRB), Eoreuma loftini, host plant resistance is an important part of management. As a control tactic, host plant resistance can not only aid in reducing stalk borer injury, but can also reduce area-wide populations and potentially slow the spread of the MRB. The effect of cultivars on reducing area-wide populations is examined by comparing the number of adult emergence holes. In addition, recent research suggests resistant cultivars which impede stalk entry and prolong larval exposure on plant surfaces may enhance the efficacy of insecticide applications. Continued evaluation of stalk borer resistance is necessary as host plant resistance remains a valuable integrated pest management (IPM) tool. A 2-year field study was conducted at the Texas A&M AgriLife Research and Extension Center at Beaumont, TX, to assess resistance to MRB among commercial and experimental sugarcane cultivars. Thirty-eight cultivars were evaluated over both years. The tests included a wide variety of cultivars developed from breeding programs in St. Gabriel, LA; Houma, LA; Canal Point, FL; and Natal, South Africa. In addition, the tests examined resistance in 4 biomass energy cultivars. In both years, the tests had 1-row, 12-foot plots arranged in a randomized block design with 5 replications (See field maps pp ) evaluation The 25 varieties evaluated in 2010 include: 5 in commercial use (HoCP , HoCP , HoCP , L , and L ), 11 experimental clones (HoCP , HoCP , HoCP , Ho , Ho , Ho , Ho , Ho , Ho , L 07-68, and L 07-57), 3 clones bred for high fiber content (Ho , US 93-15, and US 01-40), 2 energy canes (US and US ), and 4 South African cultivars (N-17, N-21, N-24, N-27). The cultivars from the South African Sugar Research Institute in KwaZulu-Natal (N-cultivars) have potential resistance to MRB because they have demonstrated varying levels of resistance to African stalkborers, especially Eldana spp., which shares many characteristics with MRB. Differences were detected in percentages of bored internodes among cultivars (F=3.56, P<0.001). Results (Table 1) showed infestations ranging from 1.0% bored internodes (N-21 and HoCP ) to 20.4% (Ho ). Of the commercial cultivars, HoCP and L were the most resistant, while L and HoCP were the most susceptible. HoCP , currently the most widely planted cultivar in Louisiana, experienced nearly 8-fold more injury than the most resistant varieties. All of the South African cultivars showed some level of resistance with N-21 being the most resistant. Adult emergence data followed the same trend as percent bored internodes with moth production ranging from < 0.01 to 0.38 emergence holes/stalk (Table 1); however, differences in emergence among cultivars were not detected (F=1.57, P=0.065). 9

10 Table 1. MRB injury and moth production in the 2010 Beaumont sugarcane variety test Variety % Bored Internodes Emergence per Stalk Ho HoCP HoCP Ho L HoCP L Ho US N Ho Ho N HoCP US Ho HoCP L Ho US N <0.01 L US HoCP <0.01 N <0.01 *Means which share a line are not significantly different (LSD, α=0.05) 10

11 2011 evaluation The 2011 test evaluated resistance in 19 cultivars. Cultivars from the 2010 test which were reevaluated include: HoCP , HoCP , Ho , L 07-57, HoCP , and HoCP HoCP has been our resistant standard for several years. HoCP , which appears to have little resistance to the MRB, has recently been released to commercial growers. Experimental cultivars in the early stages of varietal development include: HoCP , Ho , L , L , Ho , Ho , HoL , L , L , Ho Two energy cane varieties, L and Ho , were also evaluated. Results showed significant differences (F=2.71, P= 0.002) in injury, ranging from 1.9 to 17.2% bored internodes (Table 2). The most resistant cultivars examined were HoCP and L Experimental cultivar L is potentially highly resistant as it demonstrated >8-fold reductions in MRB injury compared to susceptible cultivars. The most susceptible cultivars were HoCP , L , and HoCP Differences in adult emergence (F= 1.99, P =0.019) followed the same trend as injury data ranging from 0.02 to 0.45 emergence hole per stalk (Table 2). Results from the cultivars which were reevaluated were consistent with findings from Energy cane varieties showed intermediate levels of resistance. Table 2. MRB injury and moth production in the 2011 Beaumont sugarcane variety test Variety % Bored Internodes Emergence/stalk HoCP L HoCP HoL Ho Ho Ho Ho L L Ho L HoCP Ho L Ho HoCP HoCP L *Means which share a line are not significantly different (LSD α=0.05). 11

12 FEEDING BEHAVIOR AND DURATION OF EXPOSURE OF MEXICAN RICE BORER LARVAE ON SUGARCANE Blake E. Wilson 1, T.E. Reagan 1, J.M. Beuzelin 1, and A.T. Showler 2 1 Department of Entomology, LSU AgCenter 2 USDA-ARS Weslaco, Texas A greenhouse study was conducted at the USDA ARS Kika de La Garza Subtropical Agricultural Research Center (Weslaco, Hidalgo County, TX) to investigate Mexican rice borer (MRB), Eoreuma loftini, larval feeding behavior on immature (6 nodes) and mature (12 nodes) sugarcane stalks of a resistant (HoCP ) and susceptible (HoCP ) cultivar. Plants were arranged in a completely randomized design with each of the four treatments (cultivar by phenological stage) applied to 12 stalks. Strips of freshly laid MRB eggs were attached to the leaves at locations consistent with normal oviposition activity. Egg strips were removed after hatching, and position and feeding behavior of newly emerged larvae were recorded daily. Numerous entry holes into leaf midribs within one day of hatching indicated that many larvae were only briefly exposed on plant surfaces. The number of larvae to enter the midribs, duration of exposure, and larval survival were recorded. Over all treatments,feeding behavior and establishment of a total of 277 larvae was monitored (Table 1). More than half of newly hatched larvae on immature stalks of HoCP bored into the plant (midrib), where they would be protected from contact insecticides within one day. A greater percentage of larvae became established feeding on the susceptible HoCP than on HoCP Larval establishment was greater on mature than on immature sugarcane. However, larval survival to stalk entry was greater on immature than mature sugarcane, which may be related to increasing rind hardness as stalks mature. Duration of exposure was shortest on immature HoCP (3.4 d) and greatest on mature stalks of HoCP (6.4 d). This research demonstrates the short window of exposure of MRB larvae to control tactics. Because of the limited vulnerability of MRB larvae, improved application timing and residual activity of insecticides have potential to enhance efficacy of MRB chemical control. Additionally, resistant cultivars which impede larval establishment and prolong exposure would likely allow increased larval vulnerability to chemical or biological control tactics. Table 1. MRB larval behavior and exposure on sugarcane, Weslaco, TX, 2010 % of larvae to establish feeding % of larvae to enter midrib in 1 day Duration of exposure (day) % of established larvae surviving to stalk entry HoCP Nodes Nodes HoCP Nodes Nodes

13 RED IMPORTED FIRE ANT PREDATION ON MEXICAN RICE BORER IN SUGARCANE AT BEAUMONT, TX IN M.T. VanWeelden 1, J.M. Beuzelin 1, B.E. Wilson 1, T.E. Reagan 1, and M. O. Way 2 1 Department of Entomology, LSU AgCenter 2 Texas A&M AgriLife Research and Extension Center at Beaumont, Texas A study was initiated in the summer of 2011 at the Texas A&M AgriLife Center at Beaumont, TX to assess the effect of predation by the red imported fire ant (Solenopsis invicta) on Mexican rice borer (MRB) injury to sugarcane. The experiment was conducted in plots of the 2010 and 2011 sugarcane variety tests by establishing ant-suppressed and unsuppressed areas. Ant populations were suppressed using a granule bait formulation of hydramethylnon and S- methoprene applied to the rows and bases of plants. In each area of the variety tests, MRB injury was assessed in four sugarcane cultivars of interest; two conventional cultivars and two energy cultivars (Table 1). Bored internodes and emergence holes from MRB were counted on 10 randomly selected stalks from each plot using destructive sampling and a stalk-splitter machine borrowed from the Texas A&M Center at Weslaco. The percentage of bored internodes and number of emergence holes were analyzed using generalized linear models (Proc Glimmix, SAS Institute) with binomial and Poisson distributions, respectively. A 50% increase in the percentage of bored internodes was observed across all antsuppressed areas. However, statistical analysis did not detect differences (F=1.48, P=0.284) supporting the numerical trend (Table 1). A difference in emergence holes per stalk was associated with ant suppression (F=2.43, P=0.023). The mean number of emergence holes per stalk across all unsuppressed areas was 0.16, and increased to 0.36 in areas where ants were suppressed. This data suggests that predation of the MRB by the red imported fire ant decreases both injury and buildups of pest populations in sugarcane. Additional data collected from pitfall traps implemented throughout the summer to detect relative abundance of the red imported fire ant may help to better quantify the role of ant predation. MRB infestations in leaf sheaths recorded bi-weekly still need to be analyzed. Table 1. Mean percentage of bored internodes and emergence per stalk by sugarcane cultivar with ants suppressed and unsuppressed in Beaumont, TX, 2011 Variety Ants Suppressed Ants Not Suppressed % Bored internodes Emergence/stalk % Bored internodes Emergence/stalk HoCP (plant and ratoon) HoCP (plant and ratoon) Ho (plant) L (plant) Ho (ratoon) Ho (ratoon) This research is part of the Ph.D. dissertation program of Matt VanWeelden 13

14 PHEROMONE TRAP ASSISTED SCOUTING AND AERIAL INSECTICIDAL CONTROL OF THE MEXICAN RICE BORER, 2009 AND 2010 Blake E. Wilson 1, T.E. Reagan 1, J.M. Beuzelin 1, and A.T. Showler 2 1 Department of Entomology and 2 USDA-ARS Weslaco, Texas A 2-year field study was conducted to evaluate the use of pheromone traps to enhance scouting and improve chemical control of the Mexican rice borer (MRB), Eoreuma loftini, in commercial sugarcane fields in the Lower Rio Grande Valley (Cameron County, Texas). Evaluation of aerial insecticide applications for control of MRB was conducted in a large area randomized block design (RBD) with 5 replications. Insecticide treatments were assigned randomly to plots (10 acres/plot) in fields ranging from acres of variety CP (ratoon) in 2009 and Pheromone traps were used to help with scouting and better monitor MRB population densities to more effectively time the need for insecticide applications. Trap catches of >20 moths/trap/week were used as a scouting threshold to initiate monitoring for larval infestations in (Fig. 1A). Treatable larval infestations (on plant surfaces) were determined by examining two ten stalk samples per plot. In 2009, one incident of larval scouting was necessary to determine that infestations exceeded the threshold of 5% of stalks with larvae on plant surfaces. Weekly larval scouting was conducted in 2010 throughout the growing season, and a direct correlation was observed between pheromone trap catches and larval infestations (Fig. 1B). A single aerial application was made in both 2009 and 2010 on mornings of 21 Aug and 13 Aug, respectively, by fixed wing aircraft at 10 GPA with less than 5 mph wind. At the end of the growing season, injury data were collected from 30 stalks/plot. Yield data in 2010 were collected with the core sampling method with each 10 acre plot harvested completely. In both 2009 and 2010, the recently labeled (Section 3 for sugarcane) environmentally friendly insecticide, novaluron (Diamond ), showed the best control with 7.6% bored internodes, which was significantly less than the untreated plots (19.1% bored) averaged over both years. β- cyfluthrin (Baythroid ) provided intermediate control (Table 1). Differences in moth emergence followed the same trend as percent bored internodes, with significant differences detected among treatments (Table 1). Yield data from 2010 indicate that the novaluron treatment led to a 14% increase in sugar production over untreated controls, while β-cyfluthrin treated plots were only significantly different from controls in terms of sugar/ton of cane (Table 1). Based on the current price of sugar (~$695.60/ton), the novaluron application reduced revenue losses by $276/acre. This study demonstrates the potential of pheromone trap-assisted-scouting to reduce scouting effort and optimally time insecticide applications. Additionally, the economics of MRB insecticidal control could be greatly improved if sugar production can be increased with a single, well-timed insecticide application. Table 1. MRB injury and sugar yield from aerial insecticide tests, LRGV, 2009 and 2010 Rate (fl Sugar(lbs)/ton Sugar Treatment % Bored Emergence/Stalk oz/acre) of cane (tons)/acre Diamond a 0.26 a a 3.16 a Baythroid a 0.39 ab b 2.59 b Untreated NA 19.1 b 0.62 b c 2.91 b *Means which are followed by the same letter are not significantly different (P > 0.05). 14

15 Fig. 1. Pheromone trap monitoring of MRB in Hidalgo and Cameron Counties, TX. (A) Average no. of MRB/trap/week throughout the 2009 growing season. (B) Relationship between adult population densities (no. of MRB/trap/week) and larval infestation (percent of stalks infested with treatable larvae feeding in leaf sheaths),

16 COMPARISON OF MEXICAN RICE BORER PEST PRESSURE IN BIOENERGY AND CONVENTIONAL SUGARCANE 1 T.E. Reagan 1, B.E. Wilson 1, M.T. VanWeelden 1, J.M. Beuzelin 1, W.H. White 2, and M.O. Way 3 1 Department of Entomology, LSU AgCenter; 2 USDA Sugarcane Research Unit at Houma 3 Texas A&M AgriLife Research and Extension Center at Beaumont, Texas A study conducted at the Texas A&M AgriLife Center at Beaumont, TX compared the effects of Mexican rice borer (MRB), Eoreuma loftini, infestations in energycane cultivar L and two conventional sugarcane cultivars, HoCP (resistant) and HoCP (susceptible). The experiment was set up in a randomized block design arrangement with 4 replications. Each 1-row 12-ft-long plot was split into two 6-ft sub-plots. Sub-plots were either protected from MRB infestations or left unprotected. Protected sub-plots received two applications of tebufenozide (Confirm) applied at 15.0 oz/a in Jul and Aug with a back-pack sprayer containing 2 gal of water. From late Jun to late Aug, MRB larval feeding signs in leaf sheaths were monitored every other week. In early Sep, stand counts were taken from each subplot and10 stalk samples were collected and weighed. For each stalk, the numbers of bored internodes, total internodes, and emergence holes were recorded. Total juice volume and Brix value were recorded from 4 stalks. Juice volume/6 row-ft was calculated multiplying volume/stalk by the no. stalks/sub-plot. Untreated MRB larval feeding injury in leaf sheaths of energycane L ranged between 60 and 90% of injured stalks during the initial sampling periods, and averaged 20.3 and 12.5% in HoCP and HoCP , respectively (Table 1). Insecticide applications reduced the percentage of bored internodes (F=23.8, P<0.001) and emergence per stalk (F=5.7, P=0.024), with unprotected HoCP and protected HoCP sustaining the greatest and lowest levels of injury, respectively (Table 2). Energycane L sustained intermediate levels of injury. Differences between cultivars were detected for weight of 10 stalks (F=3.8, P= ), juice volume (F=13.1, P<0.001), and Brix (F=273.6, P<0.001). Although insecticidal protection decreased MRB injury for all cultivars, increases in yield parameters were only detected for susceptible sugarcane HoCP (Table 3). These data suggest that HoCP and L are more tolerant to MRB injury. Future quantification of the impact of MRB infestations and associated injury on yield components will be critical to determine the need for management actions in energycane. Table 1. MRB injury in leaf sheaths of sugarcane and energycane, Beaumont, TX, 2011 % Injured stalks Cultivar Treatment* 8 Jul (pre-treatment) 22 Jul 3 Aug HoCP Protected NA 0 5 (resistant) Unprotected HoCP Protected NA 0 10 (susceptible) Unprotected L Protected NA (energycane) Unprotected *Protected = Confirm applied on July 10 and August 3 1 A portion of this study is anticipated to be part of the Ph.D. dissertation program of Matt VanWeelden 16

17 Table 2. MRB injury and emergence in sugarcane and energycane, Beaumont, TX, 2011 Cultivar Treatment % Bored internodes Emergence/Stalk HoCP Unprotected L Unprotected HoCP Unprotected L Protected HoCP Protected HoCP Protected *Means sharing a line are not significantly different (LSD, α=0.05); Protected = Confirm applied on July 10 and August 3 Table 3. Yield parameters as affected by cultivar and insecticide applications Cultivar Treatment # Stalks/ Weight of 10 Juice volume 6 row ft stalks (Kg) (L / 6 row ft) Brix HoCP Protected Unprotected HoCP Protected Unprotected L Protected Unprotected *Means sharing a line are not significantly different (LSD, α=0.05) 17

18 SMALL PLOT ASSESSMENT OF INSECTICIDES AGAINST THE SUGARCANE BORER, 2011 B.E. Wilson 1, J.M. Beuzelin 1, M.T. VanWeelden 1, M.O. Way 2, and T.E. Reagan 1 1 Department of Entomology, LSU AgCenter 2 Texas A&M AgriLife Research and Extension Center at Beaumont, Texas Seven insecticide treatments (Table 1), in addition to an untreated check, are being assessed for season-long control of the sugarcane borer (SCB), Diatraea saccharalis (F.), in a RBD with five replications in a field of variety HoCP stubble cane at Burns Point in St. Mary Parish. Lorsban and Extinguish were applied on June 16 for suppression of red imported fire ants and other predatory arthropods. Insecticides for SCB were applied to 3-row plots (24 ft) on August 4 and 30, The treatments were mixed in water and applied with the nonionic surfactant Induce at 0.25% v/v using a Solo back pack sprayer delivering 10 gpa at 14 psi. SCB internode boring and larvae infesting leaf sheaths were observed on August 25 in selected plots. These preliminary observations helped to verify possible differences in control residual among treatments prior to the second insecticide application. SCB injury will be assessed by recording the number of bored internodes and the total number of internodes from 15 stalks per plot in early October. Table 1. Treatments applied to manage SCB in sugarcane in 2011, Burns Point, LA Treatment Trade name Common name Rate (oz/a) A Besiege Rynaxypyr + λ- 9.0 Cyhalothrin B Belt Flubendiamide 3.0 C Control NA NA D Diamond Novaluron 12.0 E Confirm Tebufenozide 8.0 F Coragen Rynaxypyr 3.0 G Prevathon Rynaxypyr 20 H Prevathon Rynaxypyr 12 Table 2. Plot map, 2011, Burns Point, LA Rep 5 Rep 4 Rep 3 Rep 2 Rep 1 D5 B4 A3 D2 H1 H5 C4 H3 G2 G1 E5 F4 D3 B2 F1 B5 A4 G3 E2 E1 G5 H4 B3 F2 D1 A5 D4 E3 C2 C1 F5 E4 C3 A2 B1 C5 G4 F3 H2 A1 Quarter Drain 1 Border- row 18

19 Author's personal copy Crop Protection 29 (2010) 1168e1176 Contents lists available at ScienceDirect Crop Protection journal homepage: Field assessment of novaluron for sugarcane borer, Diatraea saccharalis (F.) (Lepidoptera: Crambidae), management in Louisiana sugarcane J.M. Beuzelin a, *, W. Akbar b, A. Mészáros a, F.P.F. Reay-Jones c, T.E. Reagan a a Department of Entomology, Louisiana Agricultural Experiment Station, Louisiana State University Agricultural Center, 404 Life Sciences Bldg, Baton Rouge, LA 70803, USA b Monsanto Company, 700 Chesterfield Pkwy West GG3E, Chesterfield, MO 63017, USA c Department of Entomology, Soils and Plant Sciences, Clemson University, Pee Dee Research and Education Center, 2200 Pocket Rd., Florence, SC 29506, USA article info abstract Article history: Received 28 January 2010 Received in revised form 24 May 2010 Accepted 1 June 2010 Keywords: Diatraea saccharalis (F.) Sugarcane Biorational insecticide Chitin synthesis inhibitor Integrated pest management On-farm field experiments were conducted in 2004 and 2007 to assess the suitability of novaluron, a chitin synthesis inhibitor, for sugarcane borer, Diatraea saccharalis (F.), management in Louisiana sugarcane (Saccharum spp. hybrids). Aerial insecticide applications reproducing commercial production practices were made when D. saccharalis infestation levels exceeded a recommended action threshold. In addition to decreased D. saccharalis infestations, 6.3 e 14.5-fold reductions in end of season injury, expressed as the percentage of bored sugarcane internodes, were observed in plots treated with novaluron. D. saccharalis control in novaluron plots was equivalent to (P > 0.05) or better (P < 0.05) than that achieved with tebufenozide, an ecdysone agonist that has been extensively used for over a decade on sugarcane. With a numerical trend of a 3.1-fold decrease in percent bored internodes, the pyrethroid gamma-cyhalothrin seemed less effective than the biorational insecticides in protecting sugarcane against D. saccharalis. Using continuous pitfall trap sampling, no measurable (P > 0.05) decreases in predaceous and non-predaceous soil-dwelling non-target arthropods were associated with insecticides. However, numerical trends for decreases in immature crickets associated with novaluron and gammacyhalothrin were recorded in Our data suggest that novaluron will fit well in Louisiana sugarcane integrated pest management. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction The sugarcane borer, Diatraea saccharalis (F.), is a lepidopteran pest that has historically been the most damaging arthropod in Louisiana sugarcane (hybrids of Saccharum L. spp.) (Reagan et al., 1972; Reagan, 2001). Management recommendations for D. saccharalis emphasize the importance of cultivar resistance, scouting, properly timed insecticide applications, and conservation of beneficial arthropods (Reagan and Posey, 2001; Posey et al., 2006). However, resistant cultivars have been underexploited for the past decade due to widespread use of susceptible high-yielding cultivars, and adequate D. saccharalis control with narrow-range insecticides and associated conservation of natural enemies (Reay- Jones et al., 2005). The red imported fire ant, Solenopsis invicta Buren, is the dominant natural enemy of D. saccharalis in Louisiana sugarcane (Reagan,1986), contributing an estimated savings of as much as two insecticide applications per year for D. saccharalis management (Sauer et al., * Corresponding author. Tel.: þ ; fax: þ address: jbeuzelin@agcenter.lsu.edu (J.M. Beuzelin). 1982). Spiders (Araneae) are the primary D. saccharalis egg predators and are probably second in importance in the natural enemy complex (Negm and Hensley, 1969; Ali and Reagan, 1986). Ground beetles (Coleoptera: Carabidae), tiger beetles (Coleoptera: Carabidae: Cicindelinae), rove beetles (Coleoptera: Staphylinidae), click beetles (Coleoptera: Elateridae), and earwigs (Dermaptera) have also been cited as important components of the D. saccharalis natural enemy complex in Louisiana (Negm and Hensley, 1967, 1969). Natural enemies of D. saccharalis are largely protected in Louisiana sugarcane by the widespread use of tebufenozide, which represented 90% of the foliar applications in 2007 (Pollet, 2008). This biorational insecticide belonging to the diacylhydrazine class is an ecdysone agonist that causes larvae to produce a malformed cuticle (Dhadialla et al., 1998). This compound is very specific to certain lepidopterans (Dhadialla et al., 1998) and has shown little to no toxicity to D. saccharalis natural enemies (Reagan and Posey, 2001). In addition to tebufenozide, the pyrethroids esfenvalerate, cyfluthrin, zeta-cypermethrin, lambda-cyhalothrin, and gammacyhalothrin are labeled but seldom used (Pollet, 2008). Because the development of resistance to different classes of insecticides in D. saccharalis populations has been a recurring problem in Louisiana /$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi: /j.cropro

20 Author's personal copy J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e sugarcane (Vines et al., 1984; Akbar et al., 2008), over-reliance on tebufenozide has raised concerns. Depending on cultivar and agricultural consultant recommendations, growers apply insecticides when the level of stalks infested with at least one live larva feeding in the leaf sheaths exceeds a 5e10% threshold (Schexnayder et al., 2001; Posey et al., 2006). After field management failures were reported, Reay-Jones et al. (2005) documented reductions in susceptibility to tebufenozide among D. saccharalis populations in Louisiana. Akbar et al. (2008) obtained a 27.1-fold increase in LC 50 after 12 generations of selection with tebufenozide in the laboratory. Appropriate insecticide resistance management strategies are therefore needed to preserve a balance of D. saccharalis control tactics for the Louisiana sugarcane industry. Among potential alternatives to tebufenozide, novaluron is a biorational insecticide belonging to the benzoylphenyl urea class that was initially registered in the USA in 2001 (Ishaaya and Horowitz, 1998; US EPA, 2001). Benzoylphenyl ureas inhibit chitin polymerization, thus disrupting cuticle formation in immature insects (Oberlander and Silhacek, 1998). Novaluron is therefore not directly toxic to adult insects, but exerts insecticidal activity on egg and larval stages (Barzani, 2001). By 2008, this insecticide had been granted permanent federal labels in the USA for use on cotton, potato, apple, Brassica vegetables, and ornamentals to control or suppress caterpillars (Lepidoptera: Gracillariidae, Noctuidae, Plutellidae, Pyralidae, Tortricidae), hemipterans (Hemiptera: Aleyrodidae, Miridae, Pentatomidae), beetles (Coleoptera: Chrysomelidae, Curculionidae), thrips (Thysanoptera: Thripidae), and leafminers (Diptera: Agromyzidae) (CPR, 2008; T&OR, 2008). Additionally, novaluron has a relatively low mammalian toxicity (Barzani, 2001). In sugarcane, preliminary small-plot studies showed that novaluron reduced D. saccharalis infestations below economic levels (Posey et al., 2003; Akbar et al., 2004). Targeting immature stages, novaluron is expected to have limited non-target effects on adult natural enemies that are present in the sugarcane agroecosystem (Ishaaya et al., 2001, 2002). Thus, this biorational pesticide has the potential to become a major component of Louisiana sugarcane integrated pest management (IPM). In addition, having a different mode of action from other labeled insecticides, novaluron represents an alternative that would reduce the selection pressure on D. saccharalis from other classes of insecticides, mitigating the potential development of insecticide resistance. Before novaluron was granted a permanent federal label in 2009 for use on sugarcane in the USA ( 2009), two aerial application field studies were conducted in 2004 and These studies reported in this paper were conducted on commercial farms to assess the efficacy and non-target arthropod impacts of novaluron for D. saccharalis management in Louisiana sugarcane. 2. Material and methods 2.1. Experimental plots and D. saccharalis pest severity assessment e 2004 A study was conducted during the summer of 2004 near Cheneyville, Rapides Parish, LA (N , W ) in commercial fields planted during the summer of 2003 with sugarcane cultivar LCP Portions of fields were divided into 16 plots of 2 ha (30 rows, 1.83-m row spacing) in a randomized complete block design arrangement with four blocks. Each plot was assigned one of four treatments. In addition to an untreated control, insecticide treatments were tebufenozide (Confirm Ò 2F) at 140 g(ai)/ha, and novaluron (Diamond Ò 0.83EC) at 58 g(ai)/ha and 87 g(ai)/ha. From mid-june, pre-treatment D. saccharalis infestation levels were determined by weekly examinations of 25 randomly selected stalks from each block, observing for live larvae (1ste3rd instars) infesting leaf sheaths. The 5% threshold was exceeded on July 15 when 10% of the stalks were infested and the first insecticide application was made on July 16. All insecticide treatments were applied in water with the surfactant Latron Ò CS-7 at the rate of 0.25% vol/vol. A Turbo Thrush Commander aircraft equipped with 38 CP-09-3P nozzles (0.125 orifice, 30 deflector, kpa pressure, CP Products Inc., Tempe, AZ) and delivering 46.7 L per hectare of finished formulation was used to spray swaths of 18.3 m at a speed of approximately 210 km/h. Subsequently, post-treatment infestation levels were assessed in each plot on July 25, 30, August 5, 13, 21, 26, and September 2. All insecticides were applied again on August 13 when a 10% threshold was exceeded in the high rate novaluron plots. Later infestation levels did not warrant a third insecticide application. At the end of the growing season, D. saccharalis injury (no. bored internodes/total no. internodes) and moth production (no. adult emergence holes) were recorded from 25 stalks randomly selected in each plot on September Non-target arthropod pitfall trap sampling e 2004 Three pitfall traps were used to determine relative soil-associated arthropod abundance in each plot. Traps consisted of wide mouth 0.47-L glass jars (Ball Corp., Broomfield, CO) filled with 150 ml of ethylene glycol and 2 ml of liquid soap to reduce surface tension. Traps were placed on the 15th, 16th, and 15th row of each plot, respectively 30, 60, and 90 m from the unplowed front. Pitfall traps were imbedded to the soil surface and were covered by a 15 by 15 cm metal plate, which was supported by a tripod and elevated 3 cm above the jar to exclude rain, debris, and larger animals. Pitfall traps were initially placed in the experimental plots on June 11. For pre-treatment sampling, traps were collected and replaced on July 2 (21 days) and July 20 (18 days). For treatment assessment, traps were collected and replaced on August 4 (15 days) and August 17 (13 days). All traps were collected after a fifth sampling period on September 2 (16 days). For each sampling period, the non-target arthropods collected were counted after being sorted to the following 13 groups: S. invicta, spiders, earwigs (Dermaptera: Anisolabididae, Forficulidae), ground beetles, tiger beetles, click beetles, rove beetles, scarab beetles (Coleoptera: Scarabaeidae), other Coleoptera, field crickets (Orthoptera: Gryllidae), Orthoptera other than field crickets (Orthoptera: Gryllotalpidae, Tridactylidae), leafhoppers (Hemiptera: Cicadellidae), and other ground-dwelling arthropods. Predator abundance was determined considering four groups of predators: S. invicta, spiders, pooled predaceous beetles (ground, tiger, click, and rove), and earwigs. Non-predator abundance was determined considering four groups: field crickets, pooled non-predaceous beetles (scarab and others), leafhoppers, and pooled other arthropods (Orthoptera other than field crickets and other ground-dwelling arthropods) Experimental plots and D. saccharalis pest severity assessment e 2007 A study was conducted during the summer of 2007, near Broussard, Iberia Parish, LA (N , W ) in commercial fields planted during the summer of 2006 with sugarcane cultivar HoCP Portions of fields were divided into 20 plots of 0.4 ha (12 rows, 1.83-m row spacing) in a randomized complete block design arrangement with five blocks. Each plot was assigned one of four treatments. In addition to an untreated control, insecticide treatments were tebufenozide (Confirm Ò 2F) at 140 g(ai)/ha, novaluron (Diamond Ò 0.83EC) at 65 g(ai)/ha, and gamma-cyhalothrin (Prolex Ò 1.25EC) at 20 g(ai)/ha. From mid-june, weekly examinations of 20 stalks per block indicated that the 5% threshold 20

21 Author's personal copy 1170 J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e1176 was exceeded on July 24 when 6.6% of the stalks were infested with at least one live D. saccharalis larva in the leaf sheaths. Insecticides were applied in water with the surfactant Latron Ò CS- 7 (0.25% v/v) on July 26. A Robinson R44 helicopter equipped with 36 TeeJet D6-46 nozzles directed 90 back (TeeJet Technologies, Wheaton, IL) was used to spray swaths of m. The helicopter was equipped with a flow meter calibrated to deliver 28.1 L per hectare of finished formulation regardless of ground speed. Posttreatment infestation levels were assessed in each plot on August 8, 16, 24, and 31. Because the threshold was not exceeded in the novaluron treated plots, there was no second insecticide application. At the end of the growing season, D. saccharalis injury (no. bored internodes/total no. internodes) and moth production (no. adult emergence holes) were recorded from 20 stalks randomly selected in each plot on October Non-target arthropod pitfall trap sampling e 2007 Relative soil-associated arthropod abundance was determined using two pitfall traps per plot. The two pitfall traps were placed 38 and 76 m from the front of each plot, on the 6th and 7th row, respectively. Traps were placed in plots on June 27, with pretreatment sampling conducted from July 17 to 25 (8 days). Trap assessment of treatments was conducted from July 25 to August 8 (14 days), August 8 to 31 (23 days), and August 31 to September 21 (21 days). For each sampling period, the non-target arthropods collected were counted after being sorted to 17 groups: S. invicta, ants other than S. invicta, spiders, earwigs, ground beetles, tiger beetles, click beetles, rove beetles, scarab beetles, other Coleoptera, field crickets, non-field cricket Orthoptera, leafhoppers, plant-hoppers (Hemiptera: Delphacidae), other Hemiptera (including Cercopidae), centipedes (class Chilopoda), and other ground-dwelling arthropods. Predator abundance was determined considering the same four groups of predators as in the 2004 experiment. Non-predator abundance was determined considering four groups: field crickets, pooled non-predaceous beetles (scarab and others), pooled hemipterans (leafhoppers, planthoppers, and other Hemiptera), and pooled other arthropods (ants other than S. invicta, Orthoptera other than field crickets, centipedes, and other ground-dwelling arthropods) Data analyses Each experiment was analyzed separately using Proc GLIMMIX (SAS Institute, 2008). Proportions of D. saccharalis infested stalks and bored internodes were analyzed using generalized linear mixed models with a binomial distribution and a logit link function. The number of moth emergence holes was analyzed using a oneway analysis of variance (ANOVA) with treatment as factor. Nontarget arthropod count data, including pre-treatment observations, were divided by pitfall trap sampling period duration in days and analyzed using a two-way ANOVA with treatment and sampling period as factors. Each pitfall trap was considered a sampling unit. Prior to ANOVA, moth pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi production and non-target arthropod data were transformed ð x þ 0:5Þ to normalize their distribution. A variance component covariance structure was used to model the effects of repeated measures for infestation levels and non-target arthropod counts. The KenwardeRoger adjustment for denominator degrees of freedom was used in all the models to correct for inexact F distributions. Least square means are reported for treatment effects, and were separated with Tukey s HSD (a ¼ 0.05) when differences among treatments were detected. For the 2004 experiment, contrasts were also used to compare D. saccharalis injury (proportion of bored internodes) means from novaluron (low and high rates combined) vs. tebufenozide plots. 3. Results 3.1. D. saccharalis control e 2004 and 2007 Post-treatment D. saccharalis larval infestations were lower (P < 0.05) in insecticide treated plots relative to untreated plots in both 2004 and 2007 (Table 1, Fig. 1). In 2004, differences among tebufenozide and novaluron treated plots were not detected. In treated plots, D. saccharalis infestations above the action threshold of 5e10% of infested stalks were observed 28 days after the first insecticide application, warranting the second application on August 13. Infestations in untreated plots were above the threshold from July 16, date of the first insecticide application, until the end of the season. Infestations changed over time (P < 0.05, Table 1), with a general increase observed over the growing season in untreated plots, attaining a maximum of 25.6% of infested stalks on August 31. In insecticide treated plots, reduced D. saccharalis infestations were observed 8 and 13e15 days after each insecticide application (Fig. 1). In 2007, differences in post-treatment D. saccharalis infestations among tebufenozide, novaluron, and gamma-cyhalothrin treated plots were not detected, with infestations remaining below the action threshold of 5e10% after the first insecticide applications. Infestations in untreated plots were near or above the action threshold of 5e10% from July 26, date of the first insecticide applications, until the end of the season. Post-treatment D. saccharalis infestations did not differ in time (P > 0.05, Table 1); however, a trend for an increase was observed over the growing season in untreated plots (Fig. 1). Untreated plots had the highest end of season D. saccharalis injury with 12.6 and 7.8% bored internodes in 2004 and 2007, respectively (Table 1, Fig. 2). In 2004, a reduction (8.6-fold) in injury was observed in plots treated with the low rate of novaluron. A numerical trend for a decrease in bored internodes was observed in plots treated with tebufenozide (2.9-fold) and novaluron high rate (6.3-fold) (Fig. 2). In addition, contrasts comparing novaluron treated plots with those treated with tebufenozide (F ¼ 6.56; df ¼ 1, 8.24; P ¼ 0.033) showed that novaluron was associated with lower D. saccharalis injury than tebufenozide. Differences in D. saccharalis moth production associated with insecticide treatments (Fig. 2) were not detected (Table 1). In 2007, reductions in injury were observed in plots treated with tebufenozide (8.0-fold) and novaluron (14.5-fold). Only a numerical trend for a decrease (3.1-fold) in D. saccharalis bored internodes was observed in plots treated with gamma-cyhalothrin. A numerical trend for a decrease (3.4-fold) in D. saccharalis moth emergence holes was also recorded in plots treated with gamma-cyhalothrin (Fig. 2). Whereas moth emergence holes averaged 0.37 per stalk in untreated plots, moth production was reduced (P < 0.05) in plots treated with tebufenozide (9.3-fold). Moth emergence holes were not observed in stalk samples from novaluron treated plots (Fig. 2) Non-target arthropod assessment, 2004 Non-target arthropod abundances did not differ (P > 0.05) among insecticide treated and untreated plots (Table 2). However, differences among sampling periods (P < 0.05) were detected for several soil-associated arthropod groups, as well as significant treatment by sampling period interactions (P < 0.05) (Table 2). Spider abundance differed among sampling periods extending from mid-june to early September, with no treatment by sampling period interactions detected. Prior to insecticide applications, spider abundance decreased (1.4-fold) between the first and second sampling periods. After the first insecticide applications, spider abundance increased (1.5-fold) between the second and 21

22 Author's personal copy J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e Table 1 Statistical comparisons of insecticide efficacy from on-farm aerial application experiments on sugarcane in Louisiana, 2004 and F df P > F F df P > F Post-treatment D. saccharalis larval infestations Treatment , 7.18 < , 63 <0.001 Date , , Treatment Date , , End of season D. saccharalis injury Treatment , , D. saccharalis moth production Treatment , , fourth sampling period, and then decreased (1.5-fold) during the fifth period. Predaceous beetle abundance decreased over the five pitfall trap sampling periods. However, as shown by the significant treatment by sampling period interaction, abundances were stable in plots treated with tebufenozide and novaluron high rate, whereas a decrease was observed in novaluron low rate and untreated plots (Fig. 3). When considering each sampling period separately, predaceous beetle abundances among treatments were not different. For other beetles, field crickets, and leafhoppers, abundances differed among sampling periods, with no treatment by sampling period interactions detected (Table 2). Non-predaceous beetle abundance decreased (3.3-fold) prior to insecticide applications between the first and second sampling periods, and was Fig. 1. Post-treatment levels of live D. saccharalis larval infestations (LSMeans SEM) in the leaf sheath of sugarcane from insecticide aerial application experiments in Louisiana, 2004 and stable during the remaining sampling periods. Overall field cricket and leafhopper abundances increased (6.6-fold and 33.7-fold, respectively) from mid-june to early September, with lowest abundances observed prior to the first insecticide applications. Abundance of other arthropods increased (4.7-fold) over the five pitfall trap sampling periods. However, as shown by the significant treatment by sampling period interaction, changes in abundance between the second, third, and fourth sampling periods in novaluron low rate plots (increase followed by decrease) were different from those observed in novaluron high rate plots (decrease followed by increase). When considering each sampling period separately, the abundance of other arthropods among treatments was not significantly different Non-target arthropod assessment e 2007 Except for predaceous beetles, non-target arthropod abundances did not differ (P > 0.05) among untreated and insecticide treated plots (Table 2). In comparison to untreated plots, a 1.5-fold lower predaceous beetle abundance was observed in plots treated with gamma-cyhalothrin. Predaceous beetle abundances in plots treated with novaluron and tebufenozide were not different from those in either untreated or gamma-cyhalothrin treated plots. An overall decrease in abundance over the growing season was also observed (Fig. 3). For several other soil-associated arthropod groups, differences among sampling periods were detected (P < 0.05), as well as significant treatment by period interactions (P < 0.05) (Table 2). For S. invicta and spiders, abundances decreased throughout the four pitfall trap sampling periods extending from mid-july to mid- September, 2.4-fold and 6.9-fold, respectively. Adult cricket abundance differed among sampling periods, with more adult crickets collected during the fourth sampling period than during the second. However, a significant treatment by sampling period interaction was detected. Whereas adult cricket abundance remained relatively stable in tebufenozide and novaluron treated plots, a numerical trend for a decrease (11.9-fold) between the first and second sampling, and a significant increase (13.5-fold) from the second to the fourth sampling were observed in gamma-cyhalothrin treated plots. A similar pattern was observed in untreated plots although differences among periods were not detected. When considering each sampling period separately, adult cricket abundances among treatments were not different. Immature cricket abundance increased (1.7-fold) between the first and third pitfall trap sampling periods (Fig. 4). However, a significant treatment by sampling period interaction was detected. From the first to the third sampling period, a significant increase (4.9-fold) in immature crickets was observed in untreated plots, whereas there was a trend for a decrease from the second to the third sampling in novaluron (1.5-fold) and gamma-cyhalothrin (1.3-fold) treated plots. The total number of field crickets in pitfall traps had the same pattern among sampling dates and treatments as immatures, which represented 81, 91, 90, and 77% of the crickets collected over the four sampling periods, respectively. For hemipterans, pitfall trap catches 22

23 Author's personal copy 1172 J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e1176 A B Fig. 2. (A) End of season D. saccharalis bored internodes (LSMeans SEM) and (B) adult emergence (LSMeans SEM) from insecticide aerial application experiments on sugarcane in Louisiana, 2004 and Bars within each chart followed by the same letter are not significantly different (P > 0.05, Tukey s HSD). increased (10.5-fold) between the first and third sampling periods, before decreasing (3.1-fold) during the fourth period. 4. Discussion 4.1. Insecticide efficacy Aerial applications of the biorational insecticides tebufenozide and novaluron effectively reduced D. saccharalis larval infestations in sugarcane. Both insecticides also reduced end of season injury based on D. saccharalis bored internodes, although not always significant at a ¼ Our on-farm study showed that under conditions consistent with Louisiana production practices, the chitin synthesis inhibitor novaluron decreased D. saccharalis infestations and injury, with efficacy levels equal to or better than those of tebufenozide. Sugarcane growers traditionally accept D. saccharalis injury levels at harvest below 10% bored internodes. Although sugarcane tolerance to injury differs with cultivar, White et al. (2008) determined that each 1% bored internode injury resulted in an average loss of 0.6% in sugar produced per hectare. Because a strong association exists between yield losses and D. saccharalis injury expressed as % bored internodes (White et al., 2008), yield data were not collected in our study. Whereas bored internodes represent injury causing yield losses, moth emergence holes estimate D. saccharalis adult production (Bessin et al., 1990) and document the efficacy of insecticides in decreasing pest populations produced by the infested crop. Although tebufenozide and novaluron had no measurable effects on D. saccharalis adult production in 2004, moth emergence hole data collected in 2007 provided some evidence that the biorational insecticides could decrease areawide pest populations in addition to protecting yields. Data on post-treatment larval infestations, end of season injury, and moth emergence holes collectively suggest that the pyrethroid, gamma-cyhalothrin, was less efficacious than the biorational insecticides in protecting sugarcane from D. saccharalis. However, this pyrethroid was studied only during one growing season. Gamma-cyhalothrin is the active insecticidal isomer of lambdacyhalothrin. Lambda-cyhalothrin (Karate Ò 1EC or Karate Ò Z 2.08EC) was compared to tebufenozide in five previous studies assessing the efficacy of aerially applied insecticides for D. saccharalis 23

24 Author's personal copy J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e Table 2 Statistical comparisons of the abundance of selected non-target arthropods from continuous pitfall trap sampling in sugarcane plots from on-farm insecticide aerial application experiments in Louisiana, 2004 and F df P > F F df P > F S. invicta Treatment , , Period , , Treatment Period , , Spiders Treatment , , Period , 176 < , 105 <0.001 Treatment Period , , Predaceous beetles Treatment , , Period , 176 < , 106 <0.001 Treatment Period , , Earwigs Treatment , , Period , , Treatment Period , , Adult field crickets Treatment e e e , Period e e e , Treatment Period e e e , Immature field crickets Treatment e e e , Period e e e , Treatment Period e e e , Field crickets Treatment , , Period , 176 < , Treatment Period , , Non-predaceous beetles Treatment , , Period , 176 < , Treatment Period , , Hemipterans Treatment e e e , Period e e e , 104 <0.001 Treatment Period e e e , Leafhoppers Treatment , e e e Period , 176 <0.001 e e e Treatment Period , e e e Other arthropods Treatment , , Period , 208 < , Treatment Period , , management in sugarcane (Rodriguez et al., 1995, 1998; Schexnayder et al., 1999; Posey and Reagan, 2000; McAllister et al., 2002). In all five studies, decreases in percent bored internodes below economic levels were associated with lambda-cyhalothrin, showing that this pyrethroid was suitable for D. saccharalis management. D. saccharalis injury reductions associated with lambda-cyhalothrin were not different (P > 0.05) from those associated with tebufenozide although numerically lower in all but one study (McAllister et al., 2002). Although it deserves further study, gamma-cyhalothrin also seems suitable for managing D. saccharalis below economic levels despite a possible lower efficacy compared to the biorational insecticides Non-target arthropod impact S. invicta plays a central role in Louisiana sugarcane IPM by suppressing D. saccharalis populations (Negm and Hensley, 1967, 1969; Beuzelin et al., 2009). In our study, no disruptive effects on S. invicta were observed in association with insecticide applications. Spiders are also key D. saccharalis predators (Ali and Reagan, 1986), and because their pitfall trap samples have limited spatial and temporal variability, these arthropods have been used as an indicator group in insecticide non-target assessment (Reagan and Posey, 2001). No major disruptive effects on spiders and other predaceous non-target arthropods were observed in association with insecticide applications reported in our study. Nevertheless, predaceous beetles may have been affected by aerial applications of gamma-cyhalothrin in However, because a numerical trend for more abundant (2.3-fold) predaceous beetles was observed in control plots during the pre-treatment sampling period, differential abundances might have been caused by the initial distribution of beetles among plots. Non-predaceous arthropods are also involved in the balance of the sugarcane agroecosystem. For instance, crickets, which have been used as an indicator group in non-target assessment, are important as food for S. invicta (Reagan, 2001). No major negative impacts on non-predaceous arthropods were associated with insecticide applications reported in our study. Nevertheless, our data suggest that immature crickets might have been affected by novaluron and gamma-cyhalothrin applications in Direct or residual contact with novaluron, as well as ingestion of novaluronexposed plant material, may disrupt cricket development to adulthood and kill immatures. Exposure to broad-spectrum gamma-cyhalothrin may also increase the mortality of smaller and more susceptible crickets. However, non-target assessment for immature crickets was conducted only in 2007, not allowing a generalization of the results. Tebufenozide has been shown to be exceptionally safe to nontarget arthropods in both laboratory (e.g., Smagghe and Degheele, 1995; Medina et al., 2003) and field studies (e.g., Butler et al., 1997; Gurr et al., 1999; Reagan and Posey, 2001). In our study, this ecdysone agonist had no measurable effects on the abundance of non-target arthropods. Among four previous insecticide aerial application sugarcane studies, tebufenozide was associated once with decreased ground beetles and pygmy mole crickets, but has never suppressed other non-target arthropods (Woolwine et al., 1995, 1997, 1998; McAllister et al., 2002). Gamma-cyhalothrin aerial applications were conducted for the first time in 2007 for D. saccharalis management, and possible limited non-target effects on predaceous beetles and immature 24

25 Author's personal copy 1174 J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e1176 Fig. 4. Relative abundance of immature field crickets from continuous pitfall trap sampling in sugarcane plots of insecticide aerial application experiments in Louisiana, The arrow represents the date of insecticide applications. Bars followed by the same letter are not significantly different (P > 0.05, Tukey s HSD). Fig. 3. Relative abundance of predaceous beetles from continuous pitfall trap sampling in sugarcane plots of insecticide aerial application experiments in Louisiana, 2004 and Arrows represent dates of insecticide applications. Bars within each chart followed by the same letter are not significantly different (P > 0.05, Tukey s HSD). crickets were observed. Gamma-cyhalothrin was expected to have non-target impacts similar to those of lambda-cyhalothrin. In previous large plot aerial application studies conducted on sugarcane, Woolwine et al. (1997, 1998) did not detect non-target effects associated with lambda-cyhalothrin. However, Woolwine et al. (1995) and McAllister et al. (2002) reported deleterious nontarget effects on spiders and S. invicta, respectively. Pyrethroid formulations, emulsifiable concentrate or encapsulated, varied among studies and may have impacted non-target selectivity (Pogoda et al., 2001). Negative impacts of lambda-cyhalothrin on field populations of non-target arthropods, although often temporary, have also been reported in several other agroecosystems (e.g., Pilling and Kedwards, 1996; Al-Deeb et al., 2001; Musser and Shelton, 2003). In addition to previous data on lambda-cyhalothrin, our study suggests possible non-target effects for gamma-cyhalothrin that warrant a more judicious use of this insecticide. Novaluron had no measurable negative effects on non-target arthropods observed in our study, although limited non-target effects may have occurred on immature crickets (trend for a 1.5- fold reduction over a 3-week period). Novaluron is considered relatively safe for beneficial arthropods in cotton agroecosystems (Ishaaya et al., 2001) and in greenhouses (Ishaaya et al., 2002). However, laboratory bioassays suggested that all life stages of Podisus maculiventris (Say) (Hemiptera: Pentatomidae), a beneficial predaceous non-target arthropod in potato (Solanum tuberosum L.) fields, were susceptible to novaluron, with both lethal and sublethal effects (Cutler et al., 2006). In laboratory bioassays, novaluron decreased the emergence rates of Trichogramma parasitoids (Bastos et al., 2006). Despite these non-target effects, novaluron seemed more compatible with biological control using Trichogramma wasps than organophosphate, carbamate, and pyrethroid insecticides (Bastos et al., 2006). In addition to data available in the scientific literature, our study suggests that the use of novaluron for D. saccharalis management in sugarcane is compatible with the conservation of most soil-associated non-target arthropod groups Methodological limitations Our on-farm study documented the efficacy of aerial applications of insecticides that mimicked commercial production practices, yielding results with direct practical implications compared to laboratory or small-plot experiments. Plot size (0.4 ha) minimized insecticide drift and arthropod movement from one plot to another. Our study also documented soil-associated non-target arthropod abundance using continuous pitfall trap sampling. Estimates of arthropod abundance using pitfall traps vary with absolute population size, but also with arthropod activity and habitat structure (Southwood and Henderson, 2000). Insecticides may alter arthropod activity and bias trap catches in ways not reflecting changes in the functional roles of arthropod populations. Our study did not assess the potential sublethal and long-term non-target effects of the insecticides. Pitfall trap estimates for most soil-associated arthropod groups were highly variable, making consistent patterns and differences difficult to detect. Pre-treatment sampling was included in our analyses, with the effect of insecticide applications expected to be detected with significant treatment by sampling period interactions. However, the detected interactions showed that observed treatment effects were not always consistent across sampling 25

26 Author's personal copy J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e periods. No broad-spectrum insecticides with documented consistent non-target effects were used in our study, as such chemistry is no longer recommended. For experimental purposes, future studies should include a broad-spectrum insecticide to allow for a better comparison with biorational insecticides. Using index cards soaked in peanut oil in addition to pitfall traps (Ali and Reagan, 1985) for S. invicta abundance estimation may also improve non-target assessment for this group Concluding remarks With a better understanding of ecological interactions occurring in the agroecosystem and the use of effective but narrow-range chemistry, considerable advances have been made over nearly five decades of Louisiana sugarcane IPM (Hensley, 1971; Reagan, 2001). However, in conjunction with the widespread use of D. saccharalissusceptible sugarcane cultivars, insecticides remain the primary tool for D. saccharalis management when infestations approach economically damaging levels (Reay-Jones et al., 2005). Numerous insecticides have been effective in reducing D. saccharalis infestations in sugarcane, but many of these insecticides were subsequently abandoned due to either the development of resistance or environmental issues (Vines et al., 1984; Southwick et al., 1995). For over a decade, sugarcane growers have had only pyrethroids and a diacylhydrazine available, and need a more diverse array of labeled chemicals. Novaluron appears to fit well in sugarcane IPM. This chemical provides control of economically damaging infestations when employing recommended application timing and action threshold, and also has selectivity characteristics favorable to natural enemies. Novaluron received a permanent federal label for use on sugarcane in the USA during the 2009 growing season ( 2009), providing a needed alternative to tebufenozide to which D. saccharalis populations have begun to exhibit resistance. Future research will continue to include monitoring D. saccharalis resistance to tebufenozide and potential crossresistance with novaluron, but also non-target effects that might not have been detected during the on-farm experiments of 2004 and Acknowledgements This work was supported by grants from the American Sugar Cane League, the Environmental Protection Agency Strategic Agricultural Initiative Program and various insecticide companies. We thank Grady Coburn (Pest Management Enterprises, Inc.) and Blaine Viator (Calvin Viator, Ph.D. and Associates, LLC) for technical assistance. We thank J.A. Davis, A.M. Hammond, M.J. Stout, and J.H. Temple (Louisiana State University) for their review of the manuscript. This paper is approved for publication by the Director of the Louisiana Agricultural Experiment Station as manuscript number References Akbar, W., McAllister, C.D., Reay-Jones, F.P.F., Reagan, T.E., Small plot assessment of insecticides against the sugarcane borer, Arthropod Manage. Tests 29, F84. Akbar, W., Ottea, J.A., Beuzelin, J.M., Reagan, T.E., Huang, F., Selection and life history traits of tebufenozide-resistant sugarcane borer (Lepidoptera: Crambidae). J. Econ. Entomol. 101, 1903e1910. Al-Deeb, M.A., Wilde, G.E., Zhu, K.Y., Effects of insecticides used in corn, sorghum, and alfalfa on the predator Orius insidiosus (Hemiptera: Anthocoridae). J. Econ. Entomol. 94, 1353e1360. Ali, A.D., Reagan, T.E., Vegetation manipulation impact on predator and prey populations in Louisiana sugarcane ecosystems. J. Econ. Entomol. 78, 1409e1414. Ali, A.D., Reagan, T.E., Influence of selected weed control practices on araneid faunal composition and abundance in sugarcane. Environ. Entomol. 15, 527e531. Barzani, A., Rimon, an IGR insecticide. Phytoparasitica 29, 59e60. Bastos, C.S., de Almeida, R.P., Suinaga, F.A., Selectivity of pesticides used on cotton (Gossypium hirsutum) to Trichogramma pretiosum reared on two laboratory-reared hosts. Pest Manag. Sci. 62, 91e98. Bessin, R.T., Reagan, T.E., Martin, F.A., A moth production index for evaluating sugarcane cultivars for resistance to the sugarcane borer (Lepidoptera: Pyralidae). J. Econ. Entomol. 83, 221e225. Beuzelin, J.M., Reagan, T.E., Akbar, W., Flanagan, J.W., Cormier, H.J., Blouin, D.C., Impact of Hurricane Rita storm surge on sugarcane borer (Lepidoptera: Crambidae) management in Louisiana. J. Econ. Entomol. 102, 1054e1061. Butler, L., Kondo, V., Blue, D., Effects of tebufenozide (RH-5992) for gypsy moth (Lepidoptera: Lymantriidae) suppression on nontarget canopy arthropods. Environ. Entomol. 26, 1009e1015. [CPR]Crop Protection Reference, Label & Product Listings, 24th ed., 2008 Vance Publishing Corp., Lenexa, KS. Cutler, G.C., Scott-Dupree, C.D., Tolman, J.H., Harris, C.R., Toxicity of the insect growth regulator novaluron to the non-target predatory bug Podisus maculiventris (Heteroptera: Pentatomidae). Biol. Contr. 38, 196e204. Dhadialla, T.S., Carlson, G.R., Le, D.P., New insecticides with ecdysteroidal and juvenile hormone activity. Annu. Rev. Entomol. 43, 545e569. Gurr, G., Thwaite, W., Nicol, H., Field evaluation of the effects of the insect growth regulator tebufenozide on entomophagous arthropods and pests of apples. Aust. J. Entomol. 38, 135e140. Hensley, S.D., Management of sugarcane borer populations in Louisiana, a decade of change. Entomophaga 16, 133e146. Ishaaya, I., Horowitz, A.R., Insecticides with novel modes of action: an overview. In: Ishaaya, I., Degheele, D. (Eds.), Insecticides with Novel Modes of Action, Mechanism and Application. Springer-Verlag, New York, NY, pp. 1e24. Ishaaya, I., Kontsedalov, S., Mazirov, D., Horowitz, A.R., Biorational agents: mechanisms and importance in IPM and IRM programs for controlling agricultural pests. Meded Rijksuniv Gent Fak Landbouwkd Toegep Biol. Wet. 66, 363e374. Ishaaya, I., Horowitz, A.R., Tirry, L., Barazani, A., Novaluron (Rimon) a novel IGR: mechanism, selectivity and importance in IPM programs. Meded Rijksuniv Gent Fak Landbouwkd Toegep Biol. Wet. 67, 617e626. McAllister, C.D., Posey, F.R., Bacon, T.L., Reagan, T.E., Aerial insecticidal control of the sugarcane borer, Arthropod Manage. Tests 27, F112. Medina, P., Budia, F., Del Estal, P., Vinuel, E., Effects of three modern insecticides, pyriproxyfen, spinosad and tebufenozide, on survival and reproduction of Chrysoperla carnea adults. Ann. Appl. Biol. 142, 55e61. Musser, F.R., Shelton, A.M., Bt sweet corn and selective insecticides: impacts on pests and predators. J. Econ. Entomol. 96, 71e80. Negm, A.A., Hensley, S.D., The relationship of arthropod predators to crop damage inflicted by the sugarcane borer. J. Econ. Entomol. 60, 1503e1506. Negm, A.A., Hensley, S.D., Evaluation of certain biological control agents of the sugarcane borer in Louisiana. J. Econ. Entomol. 62, 1008e1013. Oberlander, H., Silhacek, D.L., New perspectives on the mode of action of benzoylphenyl urea insecticides. In: Ishaaya, I., Degheele, D. (Eds.), Insecticides with Novel Modes of Action, Mechanism and Application. Springer-Verlag, New York, NY, pp. 92e105. Pilling, E.D., Kedwards, T.J., 18e21 November Effects of lambda-cyhalothrin on natural enemies of rice insect pests. In: Proceedings of the Brighton Crop Protection Conference: Pests & Diseases e 1996, vol. 1. British Crop Prot. Counc., Farnham, UK, Brighton, UK, pp. 361e366. Pogoda, M.K., Pree, D.J., Marshall, D.B., Effects of encapsulation on the toxicity of insecticides to the Oriental fruit moth (Lepidoptera: Tortricidae) and the predator Typhlodromus pyri (Acari: Phytoseiidae). Can. Entomol. 133, 819e826. Pollet, D.K., Insecticide applications for In: Sugarcane Research Annual Progress Report LSU AgCenter, Baton Rouge, LA, p Posey, F.R., Bacon, T.L., McAllister, C.D., Reay-Jones, F., Reagan, T.E., Small plot assessment of insecticides against the sugarcane borer, Arthropod Manage. Tests 28, F111. Posey, F.R., White, W.H., Reay-Jones, F.P.F., Gravois, K., Salassi, M.E., Leonard, B.R., Reagan, T.E., Sugarcane borer (Lepidoptera: Crambidae) Management threshold assessment on four sugarcane cultivars. J. Econ. Entomol. 99, 966e971. Posey, F.R., Reagan, T.E., Insecticidal control of the sugarcane borer e aerial application test, Arthropod Manage. Tests 25 (160F), 323. Reagan, T.E., Beneficial aspects of the imported fire ant: a field ecology approach. In: Lofgren, C.S., Vander Meer, R.K. (Eds.), Fire Ants and Leaf Cutting Ants, Biology and Management. Westview Press, Boulder, CO, pp. 58e71. Reagan, T.E., Integrated pest management in sugarcane. LA Agric. 44 (4), 16e18. Reagan, T.E., Coburn, G.E., Hensley, S.D., Effects of mirex on the arthropod fauna of a Louisiana sugarcane field. Environ. Entomol. 1, 588e591. Reagan, T.E., Posey, F.R., Development of an insecticide management program that enhances biological control. Proc. Int. Soc. Sugar Cane Technol. 24, 370e373. Reay-Jones, F.P.F., Akbar, W., McAllister, C.D., Reagan, T.E., Ottea, J.A., Reduced susceptibility to tebufenozide in populations of the sugarcane borer (Lepidoptera: Crambidae) in Louisiana. J. Econ. Entomol. 98, 955e960. Rodriguez, L.M., Ostheimer, E.A., Woolwine, A.E., Reagan, T.E., Efficacy of aerial application of selected insecticides against sugarcane borer, Arthropod Manage. Tests 20 (131F),

27 Author's personal copy 1176 J.M. Beuzelin et al. / Crop Protection 29 (2010) 1168e1176 Rodriguez, L.M., Woolwine, A.E., Ostheimer, E.A., Schexnayder Jr., H.P., Reagan, T.E., Insecticidal control of the sugarcane borer-aerial application test, Arthropod Manage. Tests 23 (140F), 287. SAS Institute, User s Manual, Version 9.2. SAS Institute, Cary, NC. Sauer, R.J., Reagan, T.E., Collins, H.L., Allen, G., Campt, D., Canerday, T.D., Larocca, G., Lofgren, C., Shankland, D.L., Trostle, M., Tschinkel, W.R., Vinson, S.B., 7e10 June Imported fire ant management strategies-panel VI. In: Proceedings of the Symposium on the Imported Fire Ant. EPA/USDA (APHIS) /70, Atlanta, GA, pp. 91e110. Schexnayder, H.P., Ostheimer, E.A., Younis, A.M., Reagan, T.E., Insecticidal control of the sugarcane borer-aerial application test, Arthropod Manage. Tests 24 (120F), 299. Schexnayder, H.P., Reagan, T.E., Ring, D.R., Sampling for the sugarcane borer (Lepidoptera: Crambidae) on sugarcane in Louisiana. J. Econ. Entomol. 94, 766e771. Smagghe,G., Degheele,D.,1995. Selectivityofnonsteroidal ecdysteroid agonists RH5849 andrh5992 to nymphs and adults of predatory soldier bugs, Podisus nigrispinus and P. maculiventris (Hemiptera: Pentatomidae). J. Econ. Entomol. 88, 40e45. Southwick, L.M., Willis, G.H., Reagan, T.E., Rodriguez, L.M., Residues in runoff and on leaves of azinphosmethyl and esfenvalerate applied to sugarcane. Environ. Entomol. 24, 1013e1017. Southwood, T.R.E., Henderson, P.A., Ecological Methods, third ed. Blackwell Science, Malden, MA. [T&OR]Turf & Ornamental Reference, Label & Product Listings, seventeenth ed., 2008 Vance Publishing Corp., Lenexa, KS. [US EPA] United States Environmental Protection Agency, EPA Pesticide Fact Sheet, Novaluron. Office of Prevention, Pesticides Environmental Protection and Toxic Substances Agency (7501C). Vines, R.C., Reagan, T.E., Sparks, T.C., Pollet, D.K., Laboratory selection of Diatraea saccharalis (F.) (Lepidoptera: Pyralidae) for resistance to fenvalerate and monocrotophos. J. Econ. Entomol. 77, 857e863. White, W.H., Viator, R.P., Dufrene, E.O., Dalley, C.D., Richard Jr., E.P., Tew, T.L., Re-evaluation of sugarcane borer (Lepidoptera: Crambidae) bioeconomics in Louisiana. Crop Prot. 27, 1256e1261. Woolwine, A.E., Rodriguez, L.M., Ostheimer, E.A., Reagan, T.E., Effects of aerially applied insecticide for SCB on non-target arthropods, Arthropod Manage. Tests 20 (134F), 257. Woolwine, A.E., Rodriguez, L.M., Ostheimer, E.A., Reagan, T.E., Effects of insecticides on non-target insects in sugarcane, Arthropod Manage. Tests 22 (135F), 322. Woolwine, A.E., Rodriguez, L.M., Ostheimer, E.A., Reagan, T.E., Impact of sugarcane borer control insecticides on non-target arthropods, Arthropod Manage. Tests 23 (142F), 288. Diamond 0.83 EC, supplemental labels greenbook.net/products.aspx?pid¼45948&sec¼supp consulted on

28 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb COMMUNITY AND ECOSYSTEM ECOLOGY Seasonal Infestations of Two Stem Borers (Lepidoptera: Crambidae) in Non-Crop Grasses of Gulf Coast Rice Agroecosystems J. M. BEUZELIN, 1 A. MÉSZÁROS, T. E. REAGAN, L. T. WILSON, 2 M. O. WAY, 2 D. C. BLOUIN, 3 AND A. T. SHOWLER 4 Department of Entomology, Louisiana Agricultural Experiment Station, Louisiana State; University Agricultural Center, Baton Rouge, LA AQ: 1 Environ. Entomol. 40(5): 000Ð000 (2011); DOI: /EN11044 ABSTRACT Infestations of two stem borers, Eoreuma loftini (Dyar) and Diatraea saccharalis (F.) (Lepidoptera: Crambidae), were compared in noncrop grasses adjacent to rice (Oryza sativa L.) Þelds. Three farms in the Texas rice Gulf Coast production area were surveyed every 6Ð8 wk between 2007 and 2009 by using quadrat sampling along transects. Although D. saccharalis densities were relatively low, E. loftini average densities ranged from 0.3 to 5.7 immatures per m 2 throughout the 2-yr period. Early annual grasses including ryegrass, Lolium spp., and brome, Bromus spp., were infested during the spring, whereas the perennial johnsongrass, Sorghum halepense (L.) Pers., and VaseyÕs grass, Paspalum urvillei Steud., were infested throughout the year. Johnsongrass was the most prevalent host (41Ð78% relative abundance), but VaseyÕs grass (13Ð40% relative abundance) harbored as much as 62% of the recovered E. loftini immatures (during the winter). Young rice in newly planted Þelds did not host stem borers before June. April sampling in fallow rice Þelds showed that any available live grass material, volunteer rice or weed, can serve as a host during the spring. Our study suggests that noncrop grasses are year-round sources of E. loftini in Texas rice agroecosystems and may increase pest populations. KEY WORDS Mexican rice borer, Eoreuma loftini (Dyar), sugarcane borer, Diatraea saccharalis (F.), alternate hosts Eoreuma loftini (Dyar) and Diatraea saccharalis (F.) (Lepidoptera: Crambidae) are stem boring pests of sugarcane (hybrids of Saccharum spp.), rice (Oryza sativa L.), corn (Zea mays L.), and sorghum [Sorghum bicolor (L.) Moench] crops in the Gulf Coast region (Long and Hensley 1972, Johnson 1984). Although D. saccharalis has been established in the southeastern United States since the 1850s (Stubbs and Morgan 1902), E. loftini has expanded its range in a northeasterly direction since its Þrst detection in south Texas in 1980 (Reay-Jones et al. 2007). E. loftini was reported in 2008 for the Þrst time in Louisiana (Hummel et al. 2010), where annual economic losses in sugarcane and rice may become as severe as $250 million within the next decades (Reay-Jones et al. 2008). In addition to crop hosts, Van Zwaluwenburg (1926) observed that E. loftini attacks practically all the grasses large enough to afford it shelter within the 1 Corresponding author, jbeuzelin@agcenter.lsu.edu. 2 Texas A&M AgriLife Research and Extension Center at Beaumont, Texas A&M University, Beaumont, TX Department of Experimental Statistics, Louisiana Agricultural Experiment Station, Louisiana State University Agricultural Center, Baton Rouge, LA USDAÐARS, Kika de la Garza Subtropical Agricultural Research Center, Weslaco, TX stalk. E. loftini has been collected from numerous grasses (Poaceae), Canna spp. (Cannaceae), and bulrush (Cyperaceae: Scirpus validus Vahl) (Osborn and Phillips 1946, Johnson 1984, Showler et al. 2011). D. saccharalis larvae also feed on a range of noncrop grasses comparable to that reported for E. loftini (Jones and Bradley 1924, Holloway et al. 1928, Box 1956, Bessin and Reagan 1990). Beuzelin et al. (2010), by using potted sentinel plants grown under natural infestations, conþrmed that a number of Gulf Coast region noncrop grasses were hosts for both E. loftini and D. saccharalis. Amazon sprangletop [Leptochloa panicoides (Presl) Hitch], a common weed in rice Þelds, was a highly suitable host, harboring the highest stem borer infestations with 75% of the plants infested with at least one larva. Johnsongrass [Sorghum halepense (L.) Pers.] and VaseyÕs grass (Paspalum urvillei Steud.), two ubiquitous perennial grasses, also supported complete larval development of both species. In contrast, broadleaf signalgrass [Urochloa platyphylla (Munro ex C. Wright) R.D. Webster], a common weed near rice Þelds, proved to be a poor stem borer host (Beuzelin et al. 2010, Showler et al. 2011). In agroecosystems, the effects of vegetation diversity on arthropod population dynamics are complex and variable (Andow 1991, Norris and Kogan 2005) X/11/0000Ð0000$04.00/ Entomological Society of America 28

29 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb 2 ENVIRONMENTAL ENTOMOLOGY Vol. 40, no. 5 Fig. 1. (A) E. loftini and (B) D. saccharalis immature densities (LS means) in noncrop habitats adjacent to rice Þelds in Texas, 2007Ð2009. Total immatures are the sum of all larvae and pupae. Error bars represent SE for total immatures LS means. AQ: 2 F1 Nearby plants may increase habitat availability for predators and offer additional shelter and food for their prey, thus increasing natural enemy density and subsequently decreasing insect pest populations (Letourneau 1987, Russell 1989). Conversely, nearby plants may increase plant host availability and release additional host-þnding stimuli for insect pests, thus enhancing pest populations (Karban 1997, Tindall et al. 2004). Previous studies have suggested that noncrop hosts could play a key role in E. loftini and D. saccharalis population dynamics in Gulf Coast agroecosystems (Beuzelin et al. 2010, Showler et al. 2011). However, the quantiþcation of noncrop host presence and use has been limited, especially when crop hosts are absent or too young to sustain stem borer development. In this study, surveys were conducted to quantify the seasonal abundance of E. loftini, D. saccharalis, and their noncrop hosts in Þeld margins and surrounding habitats of Texas rice agroecosystems. Materials and Methods Transect Sampling in Noncrop Habitats. Three farms were surveyed in the Texas Gulf Coast rice production area (Jefferson County, N, W; Chambers County, N, W; and Jackson County, N, W). These farms were sampled every 6Ð8 wk for 2 yr (April 2007-February 2008, April 2008-February 2009). For each year, two transects were located along noncultivated Þeld margins, roadsides, or ditches on each farm. Transects averaged (SE) m in length and were within 250Ð500 m of the closest rice Þelds. The minimum and maximum distances between two transects on a farm in a year averaged (SE) and (SE) m, respectively. The average distance between the centers of two transects was (SE) m. On each sampling date (Fig. 1), three representative locations per transect were sampled, with three 1-m 2 quadrats randomly selected within 10 m of the center of each location. If sections of transects were mowed by rice producers during the growing season (MarchÐ August), they were excluded from sampling for at least two consecutive sampling dates. If sections were mowed during the postseason or winter, when plant growth is the slowest, they were permanently excluded from sampling. For each quadrat, all grass-like plants, hereafter referred to as graminoids, were cut at the soil surface level and placed in 50-liter plastic bags. Bags were stored at the Texas A&M AgriLife Research and Extension Center at Beaumont, TX, in a cold room at 13Ð15 C and processed within 1 wk. Noncrop graminoids present in each quadrat were identiþed to genus or species, and their relative abundance was visually estimated per volume of sampled plant material. The number of tillers for each graminoid was recorded, except during the Þrst two sampling dates in the Þrst year of the study. During the second year of the study (April 2008-February 2009), average tiller size (from base to farthest tip) was determined for each graminoid in each quadrat from all (if tillers 4) or four randomly selected tillers. Average tiller stem diameter (as measured 1 cm below the Þrst apparent node, or 3 cm above the cut if no node present) was also determined. For tillers with ßattened stems, the average between the major and minor stem diameters was recorded. During the second year of the study, plant phenology was determined visually as the proportion of plant material that was vegetatively growing, ßowering, mature, senescent, and dead. All graminoids collected from the quadrats were visually examined for stem borer feeding injury. When a discoloration of the leaf sheath or a hole in the stem was observed, injured plants were dissected to recover E. loftini and D. saccharalis larvae and pupae, hereafter referred to as immatures. Immatures were sight-identiþed using characters reported in Browning et al. 29

30 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb October 2011 BEUZELIN ET AL.: STEM BORER INFESTATIONS IN NON-CROP GRASSES 3 (1989), Legaspi et al. (1997), and Solis (1999). For the 0Ð6% and 0Ð12% of E. loftini and D. saccharalis larvae, respectively, that were recovered in bags on each sampling date because they had crawled out of graminoid stems during sample transportation or storage, the original host plant was also determined. When a quadrat sample was comprised of a single graminoid, larvae recovered in the bag were attributed to that graminoid. When several graminoids were in a quadrat sample, larvae were attributed to a host plant based on observed injury. The size of larvae was visually determined, with small, medium-sized, and large larvae corresponding approximately to Þrst and second, third, and fourth and Þfth instars, respectively. Dependent on the number of immatures recovered on each sampling date, 10Ð60 randomly selected E. loftini and D. saccharalis immatures were reared on artiþcial diet (Southland Product Inc., Lake Village, AR) until adult eclosion to conþrm species identiþcation (Klots 1970, Legaspi et al. 1997). Transect Sampling in Rice Habitats. During the early April sampling date of each year of the study, one fallowed rice Þeld was selected and sampled on each farm to verify whether old rice stubble could host E. loftini and D. saccharalis. Fallowed rice Þelds were directly adjacent ( 35 m) to one noncrop habitat transect for at least one-third of the length of that transect, or were within 50 m of the end of one noncrop habitat transect. In addition, one rice Þeld planted between March and May was selected and sampled each year on each farm in early April, late May, and late June to verify whether newly planted rice could host stem borers. Newly planted rice Þelds were directly adjacent ( 35 m) to one noncrop habitat transect for at least one-third of the length of that transect. For each fallowed and newly planted rice Þeld, one transect was drawn and Þve (2007) or three (2008) sampling zones with three 1-m 2 quadrats in each were sampled for stem borer injury and immature presence. Adult Stem Borer Trapping. E. loftini and D. saccharalis moths were trapped on each farm near the center of each noncrop habitat transect for 7Ð14 d after transect sampling during the spring, summer, and fall. After the December and February transect sampling of noncrop habitats, moth trapping averaged 33 and 15 d, respectively, because of reduced accessibility to trapping locations. Two traps per transect, one for E. loftini and one for D. saccharalis, were positioned 10 m apart and placed 1.5 m above the soil surface on a metal pole. Bucket traps (Unitrap, Great Lakes IPM, Vestaburg, MI) were used for E. loftini moth monitoring. Each trap was baited with a synthetic female E. loftini sex pheromone lure (Luresept, Hercon Environmental, Emigsville, PA) and contained an insecticidal strip (Vaportape II, Hercon Environmental, Emigsville, PA). Sticky wing traps (Pherocon 1C Trap, Trécé Inc., Adair, OK) were used for D. saccharalis moth monitoring. Each trap was baited with two D. saccharalis female pupae nearing adult eclosion. D. saccharalis female pupae from laboratory rearing were provided by the USDA-ARS Sugarcane Research Unit, Houma, LA (Þrst year of the study) and the LSU AgCenter Rice Entomology Laboratory, Baton Rouge, LA (second year of the study). Trap catches were adjusted by the length of the sampling period to express moth abundance on a moths per trap per day basis. Data Analyses. All univariate statistical analyses were conducted using linear mixed models in Proc GLIMMIX (SAS Institute 2008). The KenwardÐRoger adjustment for denominator degrees of freedom was used in all models to correct for inexact F distributions. Unless stated otherwise, least square means standard errors from the LSMEANS statement output (Proc GLIMMIX, SAS Institute 2008) are reported. When Þxed effects were detected (P 0.05), TukeyÕs honestly signiþcant difference (HSD) ( 0.05) was used to assist in the interpretation of observed patterns and differences in least square means. E. loftini and D. saccharalis densities (number of immatures per m 2 ) were compared using univariate models with year, date, and year date as Þxed effects. Farm, farm year, transect(farm year), date transect(farm year), and location(date transect farm year) were random effects. Relative abundance was recorded simultaneously for numerous graminoids from the same observation units (i.e., quadrat). Thus, before univariate analyses, multivariate analyses including the 12 most prevalent graminoids (Table 1) were conducted using Proc GLM (SAS Institute 2008) with a MANOVA statement. Multivariate and univariate analyses included the same Þxed and random effects as for stem borer density comparisons. Graminoid tiller densities were compared using the same method as for plant relative abundance analyses. Tiller size and stem diameter, which were recorded during the second year of the study, were each compared using univariate models with date as Þxed effect and farm, transect(farm), date transect(farm), and location(date transect farm) as random effects. For each of the six graminoids consistently infested with stem borers (Table 2), percentages of recovered E. loftini as affected by year and date were compared. By transect and sampling date, the percentage of recovered E. loftini in a selected graminoid was computed as the sum of E. loftini collected from that selected plant multiplied by 100 and divided by the sum of E. loftini collected from all plants. When E. loftini were not collected from a transect on a sampling date, percentages of recovered E. loftini were not computed. In addition, when a graminoid was not recorded from a transect, the percentage of recovered E. loftini was considered zero. A multivariate analysis including the six graminoids consistently infested with stem borers was conducted before univariate analyses. Fixed effects for the multivariate model (Proc GLM with MANOVA statement, SAS Institute 2008) were year, date, and year date whereas random effects were farm, farm year, and transect(farm year). Each univariate model for each graminoid shared the same Þxed and random effects as the multivariate model. For each of the two most prevalent graminoids 30 T1 T2

31 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb 4 ENVIRONMENTAL ENTOMOLOGY Vol. 40, no. 5 Table 1. Statistical comparisons for abundance and size estimates of 12 grasses commonly found in non-crop habitats adjacent to rice fields, Texas, Plant Relative abundance Tiller density Tiller size Tiller stem diam Year Date Year date Year Date Year date Date Date Johnsongrass F df 1, 2.0 6, , , 2.2 6, , , , 22.9 P VaseyÕs grass F df 1, 2.0 6, 227 6, 227 1, 2.4 6, , , , 56.7 P Ryegrass F df 1, 9.9 6, , , , , , , 3.38 P Brome F df 1, 4.0 6, , , 4.9 6, , , 4.6 3, 6.9 P Canarygrass F df 1, 235 6, 235 6, 235 1, 2.4 6, , , 8.8 1, 1.7 P Angleton bluestem F df 1, 2.0 6, , , 2.1 6, , , 3.2 6, 3.7 P Caucasian bluestem F df 1, 7.9 6, , , 8.1 6, , , 2.5 3, 3.0 P Hairy crabgrass F df 1, , , , , , , , 12.8 P Jungle rice F df 1, , , , , , , 1 1, 4.5 P Longtom F df 1, 4.0 6, 227 6, 227 1, 8.3 6, , , 12 4, 9.9 P Torpedo grass F df 1, 8.0 6, , , 8.0 6, , , 18 5, 5.3 P Non-identiÞed perennial grass a F df 1, 2 6, , , 2.1 6, , , 6.5 4, 14 P a No reproductive parts and non-distinctive vegetative material. consistently infested with E. loftini, the percentage of recovered E. loftini per percent of plant relative abundance was determined. By transect and sampling date, it was computed as the percentage of recovered E. loftini in a selected graminoid divided by the average relative abundance for that selected plant. Only univariate analyses comparing percentages of recovered E. loftini per percent of plant relative abundance as affected by year and date were conducted (same model as for percentage of recovered E. loftini analysis). The percentage of recovered D. saccharalis and recovered D. saccharalis per percent of plant relative abundance were computed using the same method as for E. loftini. Because D. saccharalis infestations were recovered almost exclusively from the two most prevalent graminoid species, only univariate analyses comparing year and date for these two plant species were conducted (same model as for proportion of recovered E. loftini analysis). E. loftini and D. saccharalis moth trap catches as affected by year and date were also compared using the same univariate models. Results Eoreuma loftini and D. saccharalis Infestations in Noncrop Habitats. E. loftini larvae and pupae were recorded in noncrop habitats during each sampling date (Fig. 1A). There was a numerical trend (F 8.78; df 1, 2.0; P 0.097) with 2.5-fold greater E. loftini 31

32 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb October 2011 BEUZELIN ET AL.: STEM BORER INFESTATIONS IN NON-CROP GRASSES 5 Table 2. Statistical comparisons for E. loftini infestations recovered from six grasses commonly found in non-crop habitats adjacent to rice fields, Texas, Plant Percentage of recovered E. loftini Year Date Year date Johnsongrass F df 1, 8.4 6, , 55.7 P VaseyÕs grass F df 1, 2.0 6, , 55.1 P Ryegrass F df 1, 2.2 6, , 61.7 P Brome F df 1, 4.2 6, , 61.4 P Canarygrass F df 1, 7.0 6, , 52.1 P Angleton bluestem F df 1, , , 63.0 P densities in these habitats during the second year of the study than during the Þrst year ( versus immatures per m 2 ). Densities changed with date (F 2.52; df 6, 60.2; P 0.030), increasing from early spring to late fall (Fig. 1A). Across both years, the lowest E. loftini densities were observed in April ( immatures per m 2 ), whereas densities were greater in October (3.1-fold) and December (3.2-fold). As shown by the nonsigniþcant year date interaction (F 1.42; df 6, 60.2, P 0.222), differences in E. loftini densities as affected by date did not change between the Þrst and the second year of the study. For D. saccharalis, differences in densities in noncrop habitats were not detected (F 1.51; df 1, 2.0; P 0.344) between the Þrst and second year ( and immatures per m 2,respectively) of the study (Fig. 1B). Although changes in D. saccharalis densities were not detected among dates (F 1.67; df 6, 66.2; P 0.143), densities were high in October 2007 ( immatures per m 2, Fig. 1B) but not in October 2008, as evidenced by the year date interaction (F 2.39; df 6, 66.2; P 0.038). Graminoid Composition in Noncrop Habitats. The 12 most prevalent graminoids surrounding rice Þelds in Texas are listed in Table 1. The multivariate analysis shows that the relative abundance of at least one of these graminoids changed with date (WilksÕ Lambda ; F 2.02, df 72, 218.0; P 0.001), but changes occurred to a different extent between the Þrst and second year of the study (WilksÕ Lambda ; F 1.53; df 48, 152.3; P for the year date interaction). In addition, multivariate analysis comparing tiller density showed that differences across dates occurred (WilksÕ Lambda ; F 2.86; df 72, 218.0; P 0.001) for at least one of the 12 graminoids. The year date interaction was not signiþcant (WilksÕ Lambda ; F 1.19; df 48, 152.3; P 0.210). For both relative abundance and tiller density, the multivariate effect of year could not be tested because of an insufþcient number of error degrees of freedom. Johnsongrass was the most often encountered and abundant graminoid (Fig. 2). However, johnsongrass relative abundance did not differ across dates despite trends (P 0.1, Table 1) for a minimum in April ( % across both years). Trends (P 0.1, Table 1) for a greater relative abundance were also observed during the second year of the study ( versus %). Tiller density (Fig. 2B) was affected by date (Table 1), with a maximum observed in August ( tillers per m 2 ). Johnsongrass size changed with date (Table 1) with the tallest tillers observed in October, and the shortest in February and April (Fig. 3A). In addition, johnsongrass stem diameter increased from the spring to the winter (Table 1; Fig. 3B). During the early spring, dead leaßess tillers remaining from the previous year as well as young green vegetative growth with an occasional emerging ßower were recorded (Fig. 4A). Flowering peaked between April and late June, and a mixture of vegetative, ßowering, and mature tillers occurred between May and August (Fig. 4A). Mature johnsongrass showed aging foliage and empty seed heads, but also green offshoots growing from nodal buds. During the fall, a majority of mature and senescing tillers were present; but vegetative and ßowering johnsongrass was observed in areas mowed in the spring or summer. During the winter, a majority of tillers were dead or senescing. In addition, young vegetative tillers had emerged in February with 0Ð14 tillers per m 2 for an average of 1.8 tillers per m 2 (Fig. 4A). VaseyÕs grass was the second most prevalent graminoid adjacent to rice Þelds (Fig. 2). Although VaseyÕs grass relative abundance was not different among dates (Table 1), trends (P 0.1) for a lower abundance in February ( % across both years) and a greater abundance in late June ( % across both years) were observed. Differences in tiller densities between years and among dates were not detected (Table 1; Fig. 2B). During the early spring, VaseyÕs grass bunches exhibited dead plant material from earlier growth, green material in a vegetative stage, and a small proportion of ßowering tillers (Fig. 4B). Flowering peaked in the spring, and during the summer, plants showed a mixture of vegetative, ßowering, mature, and senescing tillers. The proportion of senescing tillers increased in the fall. In the winter, bunches of VaseyÕs grass were composed of dead and green vegetative tillers (Fig. 4B). VaseyÕs grass tillers were the tallest in August, 1.9- and 1.5-fold taller than in April and December, respectively (Table 1; Fig. 3A). Tiller stem diameter (Table 1) was larger in May than in October (1.2-fold, Fig. 3B). Ryegrass (Lolium spp.), brome (Bromus spp.), and canarygrass (Phalaris spp.) are annual grasses 32 F2 F3 F4

33 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb 6 ENVIRONMENTAL ENTOMOLOGY Vol. 40, no. 5 Fig. 2. (A) Relative abundance and (B) tiller density (LS means) for seven of the most commonly sampled grasses in noncrop habitats adjacent to rice Þelds in Texas, 2007Ð2009. When a grass did not occur, markers were not included on the Þgure. that did not occur in August, October, or December. Relative abundance for ryegrass showed trends (P 0.1, Table 1) for being greater (2.5-fold) during the Þrst year (Fig. 2A). In addition, ryegrass relative abundance peaked in April (Fig. 2A). As shown by the year date interaction (Table 1), changes in relative abundance between April and May, and between May and late June, occurred to a greater extent in 2007 (2.9-fold and 58.4-fold, respectively) than in 2008 (2.3-fold and 11.5-fold, respectively) (Fig. 2A). Ryegrass tillers occurred at greater densities in the early spring (April) than during the late winter (February) (Fig. 2B). Ryegrass tiller size differed with date (Table 1). Tillers measured 70 cm during the spring (Fig. 3A), and were the smallest in February (2.9-fold smaller than in April). Differences in ryegrass tiller stem diameter (Fig. 3B) were not detected (Table 1). Brome and canarygrass relative abundances were affected by date (Table 1), peaking in April and May (Fig. 2A). Brome tillers occurred at greater densities in February and April than in May (Fig. 2B). Canarygrass 33

34 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb October 2011 BEUZELIN ET AL.: STEM BORER INFESTATIONS IN NON-CROP GRASSES 7 Fig. 3. (A) Tiller size and (B) stem diameter (LS means SE) for seven of the most commonly sampled grasses in noncrop habitats adjacent to rice Þelds in Texas, 2008Ð2009. was not collected in February, and differences in tiller density from April to late June were not detected (Table 1). Similarly to ryegrass, brome tillers were the shortest in February (Fig. 3A). In addition, brome tillers collected in February showed a trend (P 0.1, Table 1) for a smaller stem diameter (Fig. 3B). Canarygrass tillers collected in April were shorter (Table 1) than those tillers sampled in May (1.3-fold, Fig. 3A); however, stem diameter did not change (Table 1; Fig. 3B). Ryegrass, brome, and canarygrass typically were ßowering or mature in early April, senescent or dead in May, and dead in late June (Fig. 4). However, late brome growth appeared in the vegetative stage in May and June. In February, while young vegetative ryegrass and brome tillers were growing, canarygrass was not (Fig. 4). Angleton bluestem [Dichanthium aristatum (Poir.) C.E. Hubbard] and Caucasian bluestem [Bothriochloa bladhii (Retz.) S.T. Blake] are two perennial grasses that occurred sporadically on the study farms, but were sometimes abundant where present. Differences 34

35 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb 8 ENVIRONMENTAL ENTOMOLOGY Vol. 40, no. 5 Fig. 4. Stem borer noncrop host phenology in habitats adjacent to rice Þelds in Texas, 2008Ð2009. in Angleton bluestem relative abundance were detected (Table 1), with relative abundance greater in the fall and winter than during the spring and summer (Fig. 2A). However, differences in tiller density (Fig. 2B), size (Fig. 3A), and stem diameter (Fig. 3B) were not detected (Table 1). For Caucasian bluestem, differences in relative abundance (Fig. 2A), tiller density (Fig. 2B), size (Fig. 3A), and stem diameter (Fig. 3B) were not detected (Table 1). Angleton bluestemõs phenology was similar to that of johnsongrass. Caucasian bluestem exhibited vegetative growth from the spring to the fall, senescent tillers with dry foliage in December, and both dead tillers and vegetative growth in February. Hairy crabgrass [Digitaria sanguinalis (L.) Scop.] and jungle rice [Echinochloa colona (L.) Link] are two summer annual grasses that were found in noncrop habitats directly adjacent to rice Þelds during the summer and the fall. Hairy crabgrass relative abundance changed with date (Table 1), peaking between August and October, with a maximum of % recorded in October However, only limited ev- 35

36 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb October 2011 BEUZELIN ET AL.: STEM BORER INFESTATIONS IN NON-CROP GRASSES 9 idence for differences in tiller density was detected (Table 1), even with a maximum of tillers per m 2 (October 2007). When hairy crabgrass tillers were present, both size ( Ð cm) and stem diameter ( Ð mm) were not different among dates (Table 1). Similarly to hairy crabgrass, jungle rice does not grow in the spring, and plants were not collected in April and May. However, differences among dates in relative abundance and tiller density (with respective maxima of % and tillers per m 2 in August 2007 were not detected (Table 1). When jungle rice tillers were present, differences in size ( Ð cm) were not detected, but there were trends (P 0.1, Table 1) for a larger stem diameter in October compared with December ( and mm, respectively). Hairy crabgrass and jungle rice were vegetative early in the summer, ßowering in August, and senescing in October. Only decaying tillers were observed in December. A nonidentiþed perennial grass with no reproductive parts and nondistinctive vegetative material was collected in wet areas of noncrop habitats surrounding rice Þelds. The relative abundance and tiller density for this grass did not differ throughout the seasons (Table 1), with a maximum of % (August 2007) and tillers per m 2 (June 2007), respectively. Tiller size and stem diameter changed with date (Table 1), with size increasing from spring to fall ( cm in April to cm in October) and stem diameter being larger in the spring ( mm in April) than during the summer and fall ( mm in June). In poorly drained areas, torpedo grass (Panicum repens L.) was also collected. Relative abundance and tiller density for torpedo grass were not different throughout the seasons (Table 1), with a maximum of % (February 2009) and tillers per m 2 (December 2008), respectively. Whereas differences in tiller stem diameter ( Ð mm) were not detected (Table 1), there were trends (P 0.1, Table 1) for shorter tillers in the spring than in the fall ( cm in April versus cm in October). Longtom (Paspalum denticulatum Trin.) was collected sporadically with relative abundance and tiller density reaching % and tillers per m 2, respectively, in June 2007 (Table 1). When longtom tillers were present, both their size ( Ð cm) and stem diameter ( Ð mm) did not differ among dates (Table 1). Other graminoids collected during this study include fall panicgrass (Panicum dichotomiflorum Michx.); longspike beardgrass [Bothriochloa longipaniculata (Gould) Allred & Gould]; browntop signalgrass [Urochloa fusca (Sw.) B.F. Hansen & Wunderlin]; bushy bluestem [Andropogon glomeratus (Walter) Britton, Sterns & Poggenb]; Bermuda grass [Cynodon dactylon (L.) Pers.]; dallisgrass (Paspalum dilatatum Poir.); ßatsedge (Cyperaceae: Cyperus spp.); bristlegrass (Setaria spp.); and NealleyÕs sprangletop (Leptochloa nealleyi Vasey). E. loftini Infestations in Noncrop Plants. Multivariate analyses showed that for at least one of the six graminoids consistently infested with stem borers (Table 2) the percentage of recovered E. loftini differed with date (WilksÕ Lambda ; F 4.12, df 36, 222.3, P 0.001). The year date interaction was signiþcant (WilksÕ Lambda ; F 2.28; df 36, 222.3; P 0.001). The multivariate effect of year could not be tested because of an insufþcient number of error degrees of freedom. The percentage of E. loftini recovered from johnsongrass differed among dates (Fig. 5A, Table 2), increasing from April to August (2.2-fold) and decreasing during the fall and winter (2.3-fold). In addition, the univariate analysis (Table 2) suggested that the percentage of E. loftini recovered from johnsongrass was greater (1.5-fold) during the second year of the study than during the Þrst. During the winter, E. loftini infesting johnsongrass were observed near nodes or within 5 cm of the soil surface, where visibly live plant tissue was found inside stems. In addition, dead desiccated E. loftini larvae were observed in February and early April. The percentage of E. loftini recovered per percent of johnsongrass relative abundance (Fig. 5B) changed with date (F 4.59; df 6, 56.3; P 0.001), following a pattern comparable to that of the percentage of recovered E. loftini. Throughout the seasons, the percentage of E. loftini recovered from VaseyÕs grass changed (Table 2), with an increase (3.3-fold) from April to late June, followed by a decrease (2.2- fold) in August and an increase (3.2-fold) during the fall and winter (Fig. 5A). The percentage of recovered E. loftini per percent of VaseyÕs grass relative abundance changed with date (F 7.70; df 6, 60; P 0.001), peaking during the winter (Fig. 5B). At this time of the year, pupae were observed in dry sections of the plants while larvae fed within green vegetative tillers close to soil level. Ryegrass and brome harbored E. loftini during the spring in 2007 and 2008 (Fig. 5A), and one E. loftini larva was recovered from brome in February The percentage of E. loftini recovered from ryegrass in April was greater (6.1-fold) during the Þrst year of the study than during the second (Table 2). A comparable trend (P 0.1, Table 2) was observed for E. loftini recovered from brome (4.0- fold). E. loftini infestations in canarygrass were found only during the spring 2007 (Fig. 5A), but differences in percentages of recovered E. loftini were not detected among dates (Table 2). Angleton bluestem was infested with E. loftini all year (Fig. 5A). However, differences in percentages of E. loftini recovered from this perennial grass were not detected among dates (Table 2). In total, 617 and 1,515 E. loftini immatures were recovered during the Þrst and second year of the study, respectively. Ninety-six point one and 98.0% of these immatures, for the Þrst and second year of the study, respectively, infested the six graminoids addressed in the previous paragraph. The remaining E. loftini immatures were recovered from 12 of the less abundant grasses and sedges (Table 3). E. loftini was 36 F5 T3

37 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb 10 ENVIRONMENTAL ENTOMOLOGY Vol. 40, no. 5 Fig. 5. Relative stem borer infestations (LS means SE) in grasses growing in noncrop habitats adjacent to rice Þelds in Texas, 2007Ð2009. (A) Percentage of recovered E. loftini in six grasses. (B) Percentage of recovered E. loftini per percent johnsongrass and VaseyÕs grass abundance. (C) Percentage of recovered D. saccharalis in johnsongrass and VaseyÕs grass. (D) Percentage of recovered D. saccharalis per percent johnsongrass and VaseyÕs grass abundance. Markers were not included on the Þgure when stem borers were not recovered. not collected from torpedo grass, Bermuda grass, or bristlegrass. D. saccharalis Infestations in Noncrop Plants. In total, 94 and 42 D. saccharalis immatures were recovered during the Þrst and second year of the study, respectively. These immatures were collected almost exclusively from johnsongrass and VaseyÕs grass, which harbored together 94 and 100% of the infestations for the Þrst and second year of the study, respectively. The remaining D. saccharalis larvae were collected from Angleton bluestem (four larvae), jungle rice (one larva), and browntop signalgrass (one larva). Differences in percentages of D. saccharalis recovered from johnsongrass and percentages of D. saccharalis recovered per percent of johnsongrass relative abundance (Fig. 5) were not detected between the 2 yr of the study (F 0.77; df 1, 9.5; P and F 0.26; df 1, 16; P 0.618, respectively) and among dates (F 1.01; df 6, 10.3; P and F 1.08; df 6, 16; P 0.417, respectively). In VaseyÕs grass, differences in percentages of recovered D. saccharalis and percentages of recovered D. saccharalis per percent plant relative abundance (Fig. 5) were not detected between years (F 0.93; df 1, 8.5; P and F 0.48; df 1, 8.0; P 0.508, respectively) and among dates (F 1.02; df 6, 11.1; P and F 0.67; df 6, 6.4; P 0.681, respectively). In addition, for johnsongrass and VaseyÕs grass, year date interactions were not detected for the percentages of recovered D. saccharalis (F 0.30; df 3, 10.3; P and F 0.27; df 3, 11.3; P 0.843, respectively) and recovered D. saccharalis per percent plant relative abundance (F 0.01; df 3, 16; P and F 1.13; df 3, 6.5; P 0.404, respectively). Spring Stem Borer Infestations in Rice Fields. In early April, old rice stubble was present in all sampled fallow Þelds but one, which had been grazed by cattle. When present, rice stubble showed evidence of stem borer injury from the previous year, but did not host E. loftini immatures. However, one D. saccharalis pupa was recovered in April 2008 [i.e., immatures per m 2 (mean SE)]. Although dead rice stubble was the only rice material available in fallow Þelds during the Þrst year of the study (April 2007), young rice plants grew in April Young rice tillers, present at a density of tillers per m 2, measured cm (mean SE) and harbored E. loftini immatures per m 2 (mean SE). Among the 17 recovered E. loftini immatures, 64, 18, and 18% were small, medium, and large larvae, respectively. Weedy 37

38 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb October 2011 BEUZELIN ET AL.: STEM BORER INFESTATIONS IN NON-CROP GRASSES 11 Table 3. Eoreuma loftini larval infestations recovered from 12 grasses and sedges found sporadically in non-crop habitats adjacent to rice fields, Texas, Plant No. quadrats infested 2007Ð2008 No. E. loftini recovered No. quadrats infested 2008Ð2009 No. E. loftini recovered Caucasian bluestem 1 on 19 Dec on 17 Feb Hairy crabgrass 2 on 15 Aug on 11 Oct on 19 Dec on 17 Feb a Jungle rice 1 on 15 Aug Longtom on 13 Dec Non-identiÞed perennial 1 on 12 Oct on 19 Dec Fall panicgrass 2 on 30 June on 19 Dec on 17 Feb Longspike beardgrass on 24 May on 28 June Browntop signalgrass 2 on 15 Aug Bushy bluestem 1 on 17 Feb on 13 Dec Dallisgrass 1 on 30 June Flatsedge on 14 Feb NealleyÕs sprangletop 1 on 15 Aug a pupa was collected. F6 grasses were also collected in fallow rice Þelds. Canarygrass was present at densities of and tillers per m 2 (mean SE) in April 2007 and 2008, respectively, with one recovered E. loftini larva in April 2007 (100% of the recovered immatures in fallow rice). Bristlegrass was present at densities of and tillers per m 2 (mean SE) in April 2007 and 2008, respectively, with Þve recovered E. loftini larvae in April 2008 (23% of the recovered immatures in fallow rice Þelds). During both years of the study, stem borer injury or infestations in young rice plants were not observed in early April and late May. By late June 2007, newly planted rice Þelds on each of the three farms of the study were at panicle differentiation or boot stages. Stem borer injury, comprised of one bored tiller and one tiller with feeding signs in the leaf sheath [i.e., injured tillers per m 2 (mean SE)], was recorded in the older rice Þeld (boot stage) in June By late June 2008, young rice Þelds were at panicle differentiation, 70% boot and 30% heading, or 100% heading stages. Stem borer injury and infestations were observed in one Þeld (70% boot and 30% heading), with an average of injured tillers per m 2 (mean SE) and a total of three D. saccharalis larvae recovered from one quadrat [i.e., immatures per m 2 (mean SE)]. Adult Stem Borer Trapping. E. loftini moth trap catches (Fig. 6) were two-fold greater during the second year than during the Þrst year of the study (F 7.68; df 1, 7.9; P 0.025). Differences in trap catches among dates were also detected (F 5.60; df 6, 56.9; P 0.001), with moth catches across both years lowest during the winter and greatest in October (Fig. 6). However, there was a trend (P 0.1) for a year date interaction (F 1.97; df 6, 56.9; P 0.086). For both years of the study, trap catches were comparable for fall and winter trapping. However, the greatest trap catches during the second year of the study were associated with greater catches between April and August with a peak in May, which was not observed during the Þrst year of the study (Fig. 6). D. saccharalis traps did not function during December and February samplings of both years because the eclosion of virgin females used as lures did not occur. Thus, data on D. saccharalis ßight activity during the winter were not collected. D. saccharalis moth trap catches were variable but showed differences among dates (F 4.30; df 4, 38.1; P 0.006), with an increase (8.4 -fold) from April to October (Fig. 6). Differences in D. saccharalis moth trap catches between the 2 yr of the study were not detected (F 1.80; df 1, 4.3; P 0.247), and the year date interaction was not signiþcant (F 1.26; df 4, 38.1; P 0.303). Discussion E. loftini Infestations in Noncrop Hosts. As early as in the 1920s (Van Zwaluwenburg 1926), it was recognized that many large-stemmed grasses could host E. loftini. However, E. loftini noncrop hosts have only recently received consideration for pest management (Beuzelin et al. 2010, Showler et al. 2011). Our study provides the Þrst quantiþcation of seasonal E. loftini infestations in plants other than Þeld crops. Under on-farm conditions of Texas Gulf Coast rice agroecosystems, infestations in noncrop grasses occurred early during the spring when young rice does not harbor E. loftini. E. loftini infestations in noncrop grasses subsequently built up during the rice growing season, and were as high as 4.8 immatures per m 2 in December, suggesting that weedy habitats surrounding rice Þelds are major overwintering areas. April sampling in fallow rice Þelds that had not been plowed showed that 38

39 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb 12 ENVIRONMENTAL ENTOMOLOGY Vol. 40, no. 5 Fig. 6. E. loftini and D. saccharalis adult trap catches (LS means SE) in noncrop habitats adjacent to rice Þelds in Texas, 2008Ð2009. Markers were not included on the Þgure when traps did not function. overwintering E. loftini larvae are not found in rice stubble. However, grassy weeds and volunteer rice growing in fallowed Þelds can serve as host during the spring. Pheromone trap data showed that, despite reduced numbers during the cold season, E. loftini moths ßy year-round in or near noncrop habitats. This is consistent with adult seasonal patterns reported by Beuzelin et al. (2010) and with observations of all developmental stages present at any time of the year in sugarcane Þelds of the Texas Lower Rio Grande Valley (Van Leerdam et al. 1986, Meagher et al. 1994). Rodriguez-del-Bosque et al. (1995) also showed that E. loftini adults continuously emerged during the winter and spring in northern Tamaulipas, Mexico. Thus, the relative role of various host plants in E. loftini population dynamics is a function of plant availability, attractiveness, and suitability throughout the year. Assessment of the seasonal abundance and phenology of noncrop graminoids of Texas Gulf Coast rice agroecosystems, as well as associated E. loftini infestations, assisted in identifying primary noncrop hosts and their potential role in the pestõs population dynamics. Johnsongrass, VaseyÕs grass, ryegrass, brome, Angleton bluestem, and hairy crabgrass were effective E. loftini hosts that allowed larval feeding and life cycle completion. Other grasses and sedges might also be suitable hosts. Graminoids observed in our study presented a wide range of plant height and stem diameter. Physical constraints associated with these plant size characteristics likely affect host suitability for stem borer larval development, with host suitability increasing with plant height and stem diameter (Beuzelin 2011, Showler et al. 2011). However, stem hardness and nutritional quality are other key factors impacting host plant suitability (Beuzelin 2011, Showler et al. 2011). Our study suggests that johnsongrass, which is abundant throughout the year, plays a substantial role in E. loftini population build-up during the rice growing season. The observed lack of live johnsongrass tissue during the winter, however, probably decreased host suitability and subsequently E. loftini survival during this season. In addition to low temperatures, desiccation is a primary abiotic stem borer mortality factor during the winter (Rodriguez-del-Bosque et al. 1995). Therefore, we contend that E. loftini larvae establishing in johnsongrass during the fall will complete their life cycle during the winter despite increased mortality. However, it is unlikely that dead johnsongrass supports the development of young larvae from E. loftini moths emerging during the winter. For VaseyÕs grass, the high percentage of recovered E. loftini and percentage of recovered E. loftini per percent plant relative abundance in February indicate that this host becomes increasingly infested during the winter. VaseyÕs grass is less infested than johnsongrass at comparable phenological stages (Beuzelin et al. 2010, Showler et al. 2011) but maintains numerous green vegetative tillers throughout the year. Thus, the substantial perennial availability of live plant tissue suitable for E. loftini development likely allows VaseyÕs grass to be a primary overwintering host. In areas with relatively less johnsongrass or VaseyÕs grass (e.g., transition between farm roads and Þeld margins), a more diverse mixture of graminoids was observed. Ryegrass and brome are E. loftini hosts in the spring, also playing a role in population build-up early during the rice growing season, even if only for a short window of time. Our study also indicated that canarygrass may play a comparable role in E. loftini 39

40 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb October 2011 BEUZELIN ET AL.: STEM BORER INFESTATIONS IN NON-CROP GRASSES 13 AQ: 3 population dynamics. Other annual and perennial grasses (i.e., crabgrass, Angleton bluestem) probably play a minimal role in E. loftini population dynamics although they may have more substantial roles if abundant in localized areas. The current study is the Þrst to our knowledge to quantitatively describe graminoids in noncrop habitats (i.e., Þeld margins, roadsides, ditches) surrounding rice Þelds in the Texas Upper Gulf Coast area. These habitats were more variable than adjacent rice Þelds because they were not under intensive management, and plant species composition was not intentionally controlled by the producers. However, the three study farms exhibited comparable noncrop habitat compositions, regardless of management (mowing, burning, herbicide applications, absence of management) or localized soil and weather variations. Based on our observations, noncrop habitats sampled in our study appear to be representative of those habitats encountered throughout rice areas of the Texas Gulf Coast. The generalization of our results to other Gulf Coast agroecosystems, however, will require additional sampling in Texas and Louisiana. D. saccharalis Infestations in Noncrop Hosts. Complementing earlier studies (Jones and Bradley 1924, Bynum et al. 1938, Bessin and Reagan 1990), we provided the Þrst year-round quantiþcation of D. saccharalis infestations in noncrop habitats. D. saccharalis was found mostly in johnsongrass and VaseyÕs grass, and infestations were low relative to E. loftini infestations. Low area-wide D. saccharalis populations in the study areas might explain the predominance of E. loftini. Diatraea saccharalis might also rely less on noncrop hosts than E. loftini. Adult D. saccharalis trapping data from our study provide evidence of moth activity in the vicinity of noncrop sampling areas. In addition, D. saccharalis infestations in experimental rice plots located within 1.25 km of noncrop sampling transects in Jackson County represented 99% of stem borer infestations in JulyÐAugust 2007 (Beuzelin 2011). In the Louisiana sugarcane agroecosystem, Bynum et al. (1938) and Ali et al. (1986) concluded that johnsongrass only played a minor role in D. saccharalis population build-up and overwintering. These observations suggest that noncrop hosts might contribute less to D. saccharalis populations than to E. loftini populations. Nevertheless, oviposition preference and immature performance studies would assist in quantifying the relative role of noncrop hosts in D. saccharalis population dynamics. Pest Management Implications. Although weeds in rice Þelds such as Amazon sprangletop can increase stem borer infestations (Tindall 2004, Beuzelin et al. 2010), cultural management typically keeps weed populations low (Kendig et al. 2003), which is why exclusively noncrop habitats surrounding rice Þelds were the focus of our study. Research in several agroecosystems showed that alternate hosts in noncrop habitats could contribute to increased pest populations. Examples of this relationship include increased consperse stink bug, Euschistus conspersus Uhler, infestations in California tomato, Solanum lycopersicum L., Þelds (Pease and Zalom 2010), and the build-up of the pyralid Mussidia nigrivenella Ragoon in Benin (Sétamou et al. 2000). Populations of the tarnished plant bug, Lygus lineolaris (Palisot de Beauvois), and twospotted spider mite, Tetranychus urticae Koch, feed on weedy hosts before moving into nearby cotton, Gossypium hirsutum L., Þelds (Fleischer and Gaylor 1987, Wilson 1995). Our study showed that noncrop grasses are sources of E. loftini populations. Thus, noncrop habitat management tactics including mowing, applications of herbicides or insecticides, or the modiþcation of weed species composition (Landis et al. 2000) could help improve rice integrated pest management (IPM). However, the value of this approach remains to be demonstrated. Relationships between noncrop host abundance, stem borer population levels, and associated crop yield losses have not been quantiþed. In addition, noncrop habitats can be a source of biodiversity enhancing natural enemy abundance (Altieri and Letourneau 1982, Norris and Kogan 2005). Although the red imported Þre ant (Solenopsis invicta Buren), spiders, and predaceous beetles suppress D. saccharalis injury to weedy Louisiana sugarcane (Ali and Reagan 1985, Showler and Reagan 1991), their interactions with stem borer populations in noncrop habitats have not been determined. E. loftini noncrop hosts might also represent refuges for parasitic wasps (Meagher et al. 1998) observed during sampling. Therefore, designing noncrop habitat management tactics for rice IPM will have to integrate the weed contribution to both pest and natural enemy populations (Landis et al. 2000, Norris and Kogan 2005). Concluding Remarks. Assuming that host-speciþc sympatric stem borer strains do not occur (Pashley and Martin 1987, Martel et al. 2003, Vialatte et al. 2005), our study showed that noncrop grasses have the potential to increase E. loftini pest populations. Thus, the manipulation of E. loftini noncrop sources may help decrease infestations in crop Þelds and slow the spread of this invasive species through Louisiana. Further research needs to be conducted to quantify the relative contribution of E. loftini oviposition preference, immature performance, movement, and natural enemy suppression to pest source-sink interactions in the agroecosystem. Subsequently, the efþcacy and economic beneþts of noncrop habitat management tactics, implemented at both Þeld and regional scales, will have to be assessed. Because E. loftini noncrop hosts can sustain D. saccharalis populations, management tactics targeting noncrop habitats could also decrease D. saccharalis pest populations. Together with previous research (Reay-Jones et al. 2008, Beuzelin et al. 2010), our study provides a foundation for a more comprehensive stem borer management strategy including crop and noncrop components of the agroecosystem. Acknowledgments We thank rice growers Bill Dishman, Jr., John and Jay Jenkins, and Gary and Michael Skalicky for permitting us use 40

41 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb 14 ENVIRONMENTAL ENTOMOLOGY Vol. 40, no. 5 of their farmland and for technical assistance. We thank Lowell Urbatsch (Herbarium of Louisiana State University) and Eric Webster (LSU AgCenter School of Plant, Environmental and Soil Sciences) for identifying numerous grass samples. We thank Mike Hiller (Texas A&M AgriLife Extension), Waseem Akbar, Blake Wilson, Kyle Baker (LSU AgCenter), and Jannie Castillo (Texas A&M AgriLife Research and Extension Center at Beaumont) for their technical assistance. We thank Jeff Davis, Mike Stout (LSU Ag- Center), and two anonymous referees for participating in the review of the manuscript. This work was supported by USDA-CSREES Crops-At-Risk IPM program grant This paper is approved for publication by the Director of the Louisiana Agricultural Experiment Station as manuscript number 2011Ð References Cited Ali, A. D., and T. E. Reagan Vegetation manipulation impact on predator and prey populations in Louisiana sugarcane ecosystems. J. Econ. Entomol. 78: 1407Ð1414. Ali, A. D., T. E. Reagan, L. M. Kitchen, and J. L. Flynn Effects of johnsongrass (Sorghum halepense) density on sugarcane (Saccharum officinarum) yield. Weed Sci. 34: 381Ð383. Altieri, M. A., and D. K. Letourneau Vegetation management and biological control in agroecosystems. Crop Prot. 1: 405Ð430. Andow, D. A Vegetational diversity and arthropod population response. Annu. Rev. Entomol. 36: 561Ð586. Bessin, R. T., and T. E. Reagan Fecundity of the sugarcane borer (Lepidoptera: Pyralidae) as affected by larval development on gramineous host plants. Environ. Entomol. 19: 635Ð639. Beuzelin, J. M Agroecological factors impacting stem borer (Lepidoptera: Crambidae) dynamics in Gulf Coast sugarcane and rice. Ph.D. dissertation, Louisiana State University, Baton Rouge, LA. Beuzelin, J. M., T. E. Reagan, M. O. Way, A. Meszaros, W. Akbar, and L. T. Wilson Potential impact of Mexican rice borer non-crop hosts on sugarcane IPM, pp. 806Ð814 (Paper BE10). In Proceedings, XXVII Congress of the International Society of Sugar Cane Technologists, March 7Ð11,Veracruz, Mexico. Box, H. E New species and records of Diatraea Guilding and Zeadiatraea Box from Mexico, Central and South America (Lepid., Pyral.). Bull. Entomol. Res. 47: 755Ð776. Browning, H. W., M. O. Way, and B. M. Drees Managing the Mexican rice borer in Texas. Texas Agricultural Experiment Station Publ. B-1620, College Station, TX. Bynum, E. K., W. E. Halley, and L. J. Charpentier Sources of infestation by the sugarcane borer and trash treatment for the destruction of overwintering borers. Proc. Int. Soc. Sugar Cane Technol. 6: 597Ð611. Fleischer, S. J., and M. J. Gaylor Seasonal abundance of Lygus lineolaris (Heteroptera: Miridae) and selected predators in early season uncultivated hosts: implications for managing movement into cotton. Environ. Entomol. 16: 379Ð389. Holloway, T. E., W. E. Halley, U. C. Loftin, and C. Heinrich The sugar-cane borer in the United States. U.S. Department of Agriculture Technical Bulletin 41: 1Ð77. Hummel, N. A., T. Hardy, T. E. Reagan, D. K. Pollet, C. E. Carlton, M. J. Stout, J. M. Beuzelin, W. Akbar, and W. H. White Monitoring and Þrst discovery of the Mexican rice borer Eoreuma loftini (Lepidoptera: Crambidae) in Louisiana. Fla. Entomol. 93: 123Ð124. Johnson, K.J.R IdentiÞcation of Eoreuma loftini (Dyar) (Lepidoptera: Pyralidae) in Texas, 1980: Forerunner for other sugarcane boring pest immigrants from Mexico? Bull. Entomol. Soc. Am. 30: 47Ð52. Johnson, K.J.R., and M. B. Van Leerdam Range extension of Acigona loftini into the Lower Rio Grande Valley of Texas. Sugar Y Azucar 76: 34. Jones, T. H., and W. G. Bradley Certain wild grasses in relation to injury to corn by the borer (Diatraea saccharalis Fab.) in Louisiana. J. Econ. Entomol. 17: 393Ð 395. Karban, R Neighbourhood affects a plantõs risk of herbivory and subsequent success. Ecol. Entomol. 22: 433Ð439. Kendig, A., B. Williams, and C. W. Smith Rice weed control, pp. 457Ð472. In C. W. Smith and R. H. Dilday (eds.), Rice: Origin, History, Technology, and Production. Wiley, Inc. Hoboken, NJ. Klots, A. B North American Crambinae: Notes on the tribe Chiloini and a revision on the genera Eoreuma Ely and Xubida Schaus (Lepidoptera: Pyralidae). J. N.Y. Entomol. Soc. 78: 100Ð120. Landis, D. A., S. D. Wratten, and G. M. Gurr Habitat management to conserve natural enemies of arthropod pests in agriculture. Annu. Rev. Entomol. 45: 175Ð201. Legaspi, J. C., R. R. Saldaña, and N. Roseff Identifying and managing stalkborers on Texas sugarcane. Texas Agricultural Experiment Station Publication MP-1777, College Station, TX. Letourneau, D. K The enemies hypothesis: tritrophic interaction and vegetational diversity in tropical agroecosystems. Ecology 68: 1616Ð1622. Long, W. H., and S. D. Hensley Insect pests of sugar cane. Annu. Rev. Entomol. 17: 149Ð176. Martel, C., A. Réjasse, F. Rousset, M.-T. Bethenod, and D. Bourguet Host-plant-associated genetic differentiation in Northern French populations of the European corn borer. Heredity 90: 141Ð149. Meagher, R. L., Jr., J. W. Smith, Jr., H. W. Browning, and R. R. Saldaña Sugarcane stemborers and their parasites in southern Texas. Environ. Entomol. 27: 759Ð 766. Meagher, R. L., Jr., J. W. Smith, and K.J.R. Johnson Insecticidal management of Eoreuma loftini (Lepidoptera: Pyralidae) on Texas sugarcane: a critical review. J. Econ. Entomol. 87: 1332Ð1344. Norris, R. F., and M. Kogan Ecology of interactions between weeds and arthropods. Annu. Rev. Entomol. 50: 479Ð503. Osborn, H. T., and G. R. Phillips Chilo loftini in California, Arizona, and Mexico. J. Econ. Entomol. 39: 755Ð 759. Pashley, D. P., and J. A. Martin Reproductive incompatibility between host strains of the fall armyworm (Lepidoptera: Noctuidae). Ann. Entomol. Soc. Am. 80: 731Ð 733. Pease, C. G., and F. G. Zalom Inßuence of non-crop plants on stink bug (Hemiptera: Pentatomidae) and natural enemy abundance in tomatoes. J. Appl. Entomol. 134: 626Ð636. Reay-Jones, F.P.F., L. T. Wilson, T. E. Reagan, B. L. Legendre, and M. O. Way Predicting economic losses from the continued spread of the Mexican rice borer (Lepidoptera: Crambidae). J. Econ. Entomol. 101: 237Ð 250. Reay-Jones, F.P.F., L. T. Wilson, M. O. Way, T. E. Reagan, and C. E. Carlton Movement of the Mexican rice 41 AQ: 4

42 balt2/zen-env-ent/zen-env-ent/zen00511/zen7486d11z xppws S 1 8/12/11 7:48 Art: EN st disk, 2nd kmb October 2011 BEUZELIN ET AL.: STEM BORER INFESTATIONS IN NON-CROP GRASSES 15 borer (Lepidoptera: Crambidae) through the Texas rice belt. J. Econ. Entomol. 100: 54Ð60. Rodriguez-del-Bosque, L. A., J. W. Smith, Jr., and J. Martinez Winter mortality and spring emergence of corn stalkborers (Lepidoptera: Pyralidae) in subtropical Mexico. J. Econ. Entomol. 88: 628Ð634. Russell, E. P Enemies hypothesis: a review of the effect of vegetational diversity on predatory insects and parasitoids. Environ. Entomol. 18: 590Ð599. SAS Institute UserÕs Manual, Version 9.2. SAS Institute, Cary, NC. Sétamou, M., F. Schulthess, S. Gounou, H.-M. Poehling, and C. Borgemeister Host plants and population dynamics of the ear borer Mussidia nigrivenella (Lepidoptera: Pyralidae) in Benin. Environ. Entomol. 29: 516Ð524. Showler, A. T., and T. E. Reagan Effects of sugarcane borer, weed, and nematode control strategies in Louisiana sugarcane. Environ. Entomol. 20: 358Ð370. Showler, A. T., J. M. Beuzelin, and T. E. Reagan Alternate crop and weed host plant oviposition preferences by the Mexican rice borer (Lepidoptera: Crambidae). Crop Prot. 30: 895Ð901. Solis, M. A Key to selected Pyraloidea (Lepidoptera) larvae intercepted at U.S. ports of entry: revision of Pyraloidea in Keys to some frequently intercepted Lepidopterous larvae by Weisman Proc. Entomol. Soc. Wash. 101: 645Ð686. Stubbs, W. C., and H. A. Morgan Sugarcane borer moth. La. Agric. Exp. Stn. Bull. 70: 885Ð927. Tindall, K. V Investigation of insect-weed interactions in the rice agroecosystem. Ph.D. dissertation. Louisiana State University, Baton Rouge, LA. Tindall, K. V., M. J. Stout, and B. J. Williams Effects of the presence of barnyardgrass on rice water weevil (Coleoptera: Curculionidae) and rice stink bug (Hemiptera: Pentatomidae) populations on rice. Environ. Entomol. 33: 720Ð726. Van Leerdam, M. B., K.J.R. Johnson, and J. W. Smith, Jr Ovipositional sites of Eoreuma loftini (Lepidoptera: Pyralidae) in sugarcane. Environ. Entomol. 15: 75Ð78. Van Zwaluwenburg, R. H Insect enemies of sugarcane in western Mexico. J. Econ. Entomol. 19: 664Ð669. Vialatte, A., C.-A. Dedryver, J.-C. Simon, M. Galman, and M. Plantegenest Limited genetic exchanges between populations of an insect pest living on uncultivated and related cultivated host plants. Proc. R. Soc. B. 272: 1075Ð Wilson, L. J Habitats of twospotted spider mites (Acari: Tetranychidae) during winter and spring in a cottonproducing region of Australia. Environ. Entomol. 24: 332Ð 340. Received 17 February 2011; accepted 5 July

43 HARVEST CUTTING HEIGHT AND RATOON CROP EFFECTS ON STEM BORER INFESTATIONS IN RICE J. M. Beuzelin 1, A. Mészáros 1, M. O. Way 2, and T. E. Reagan 1 1 Department of Entomology, LSU AgCenter 2 Texas A&M AgriLife Research and Extension Center at Beaumont A two-year field study near Ganado, TX compared infestations of the Mexican rice borer (MRB) and sugarcane borer (SCB) in rice as affected by main crop harvest cutting height and the production of a ratoon crop. Substantial infestations (> 0.52 stem borers/ft 2 ) remained in rice culms regardless of main crop cutting height (8 vs. 16 in.). However, the 8-in. cutting height reduced MRB infestations 70 to 81% whereas SCB infestations were not affected (Fig. 1). Plant dissections prior to main crop harvest showed that compared to SCB, relatively more MRB are located above 8 in. from the base of the culm (Fig. 2). In October, the ratoon crop was more infested with stem borers than the unmanaged main crop stubble during the first year of the study. The opposite was observed during the second year. Differences in unmanaged main crop stubble phenology between the two years likely caused these differences in infestation levels. During the post-growing season, infestations in main and ratoon crop stubble decreased over the winter. After favorable winter conditions, infestations in main and ratoon crop stubble were not different, attaining 0.31 MRB/ft 2 and 0.04 SCB/ft 2 by March 2008 (Fig. 3). In March 2009, rice stubble harbored 0.03 MRB/ft 2 and 0.02 SCB/ft 2 regardless of whether only a main crop or a main and ratoon crop had been produced (Fig. 3). This study showed that a lower rice harvest cutting height can suppress late season MRB populations. Furthermore, rice stubble under favorable conditions represents a substantial overwintering habitat, thus warranting evaluation of pest management tactics targeting overwintering populations. August 2007 August 2008 No. stem borers / sq. ft in 16 in * No. stem borers / sq. ft * 8 in 16 in 0 MRB SCB Fig. 1. Stem borer infestations in rice main crop stubble as affected by harvest cutting height in 2007 and 2008, Ganado, Texas. For a stem borer species in a year, * indicates infestations differed (P < 0.05). 0 MRB SCB 43

44 No. stem borers / sq. ft Below 8 in Above 8 in 0 MRB SCB Fig. 2. Stem borer infestations by location in rice culms prior to main crop harvest in 2008, Ganado, Texas Non-ratoon Ratoon Non-ratoon Ratoon No. MRB / sq. ft No. MRB / sq. ft October January March Fig. 3. Late and post-growing season MRB infestations in rice, Ganado, Texas, and October January March 44