Wiseman: Armyworm Symposium DEDICATION OF 1994 ARMYWORM SYMPOSIUM TO DR. ROBERT L. BURTON AND MR. E. A. HARRELL: EXPERTS IN INSECT REARING

Transcription

1 Wiseman: Armyworm Symposium DEDICATION OF 1994 ARMYWORM SYMPOSIUM TO DR. ROBERT L. BURTON AND MR. E. A. HARRELL: EXPERTS IN INSECT REARING B. R. WISEMAN Insect Biology & Population Management Research Laboratory Tifton, GA ABSTRACT The 1994 Symposium on Armyworms (previously Fall Armyworm Symposium) at the Southeastern Branch of the Entomological Society of America is dedicated to Dr. Robert L. Burton and Mr. E. A. Harrell: Experts in Insect Rearing. Dr. Burton, an entomologist ( ), and Mr. E. A. Harrell, an agricultural engineer ( ), were employed by the U. S. Department of Agriculture, Agricultural Research Service at the Southern Grains Insects Research Laboratory at Tifton, GA. The systems they developed provided the means by which the fall armyworm, Spodoptera frugiperda (J. E. Smith), could be reared in mass numbers on a meridic diet that resulted in quality eggs, larvae, pupae, and adult insects. Key Words: Fall armyworm, rearing, mechanization, diet. RESUMEN El Simposio de 1994 sobre los Gusanos Trozadores (previamente Simposio sobre los Gusanos Trozadores de Otoño) en la Rama Sureste de la Sociedad Entomológica de América estuvo dedicado al tema Dr. Robert L. Burton y Sr. E. A. Harrel: Expertos en Cría de Insectos. El Dr. Burton, entomólogo ( ), y el Sr. E. A. Harrel, ingeniero agrónomo ( ), fueron empleados del Servicio de Investigación Agrícola del Departamento de Agricultura de los Estados Unidos, en el Laboratorio Sur de Investigación de Insectos de los Granos en Tifton, Georgia. Los sistemas que ellos desarrollaron posibilitaron la cría masiva del gusano trozador, Spodoptera frugiperda (J. E. Smith), en una dieta merídica que tuvo como resultado la producción de huevos, larvas, pupas e insectos adultos de calidad. The 1994 Symposium on Armyworms (previously Fall Armyworm Symposium ) presented at the Southeastern Branch Meeting of the Entomological Society of America is hereby dedicated to Dr. Robert L. Burton and Mr. E. A. Harrell: Experts in Insect Rearing. BIOGRAPHIES Dr. Robert L. Burton (Fig. 1) of Stillwater, OK, was born on August 23, 1936 in Antlers, OK, to Charles and Sally (Holton) Burton. He married Sylvia J. Gentry September 1, 1960 in Durant, OK. They have two sons, Robert L. Burton Jr. of Dana Point, CA, and Brian Gentry Burton of Stillwater. Dr. Burton died February 3, 1993 in the St. Francis Medical Center in Tulsa at the age of 56. Dr. Burton became interested in Entomology early in his education. He attended Eastern Oklahoma College, Southeastern State College, the University of Oklahoma, This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

2 398 Florida Entomologist 77(4) December, 1994 Fig. 1. Dr. Robert L. Burton, USDA, ARS Supervisory Entomologist ( ). and Oklahoma State University. Dr. Burton received his B. S. and M. S. degrees in entomology from Oklahoma State University in 1963 and 1964, respectively. He was first employed as an entomologist in 1961 with Standford-Vaddar Entomology Service of Plainview, TX, and then as a Research Assistant with Oklahoma State in Dr. Burton joined the U. S. Department of Agriculture, Agricultural Research Service in 1964 with his first assignment at the Southern Grain Insects Research Laboratory in Tifton, GA. He quickly assumed a leadership role as the entomologist in charge of insect rearing. In 1970, Dr. Burton was transferred to Stillwater where he continued both his work and studies. He completed his Ph. D. degree in Dr. Burton continued his work at Stillwater until his death, at which time he had attained the position of Supervisory Entomologist, Research Leader and Director of the USDA-ARS Plant Science and Water Conservation Laboratory as well as adjunct Professor of Entomology at Oklahoma State University. Dr. Burton authored and co-authored 126 scientific publications during his 28 years of dedicated service to agriculture. He also presented more than 114 scientific presentations during his tenure with ARS. Dr. Burton was recognized as a world authority in the areas of insect diets and the laboratory production of insects. Mr. Edsel A. Harrell (Fig.2) began his life on a little farm in South Georgia about four miles north of Whigham. He was born on October 17, 1924, the son of C. Braxton and Mabel (Moore) Harrell. Edsel believed that there was a better life somewhere

3 Wiseman: Armyworm Symposium Fig. 2. Mr. Edsel A. Harrell, USDA, ARS Agricultural Engineer (1924--). other than shaking peanuts and picking cotton by hand. Upon graduation from high school in Whigham, he left the farm for Alexandria, VA, where he began an apprenticeship as a machinist in a torpedo plant. During World War II, Edsel served his country honorably in the Army Signal Corps on isolated islands in the Pacific. His specific assignment was cryptography. He was discharged in January of Edsel then enrolled in Abraham Baldwin Agricultural College in Tifton, GA, to pursue a degree in Agricultural Engineering. He obtained his BSAE in 1950 and his MSAE in 1951 from the University of Georgia. (An interesting note: Edsel taught eighth and ninth grade math and coached the boys and girls basketball teams at Whigham high school while he was obtaining his BS degree). During this time he met and married Martha E. Elkins. They have three children: Edsel Jr. of Dallas, TX, Debbie of Nashville, TN, and Karen of Watkinsville, GA, and five grandchildren. Mr. Harrell began his research career as an Agricultural Engineer with the U. S. Department of Agriculture at the U. S. Cotton Ginning Laboratory at Stoneville, MS, in He transferred in 1961 to the Southern Grain Insects Research Laboratory (SGIRL) in Tifton, GA, where he was in charge of the Pest Control Research Project until his retirement in It was during this period of time that Edsel and Dr. Burton teamed up to solve some of the most difficult problems encountered in the mass rearing of Helicoverpa zea (Boddie) and Spodoptera frugiperda (J. E. Smith). Mr. Harrell authored or co-authored more than 60 scientific publications during his 30 years of productive service and he presented more than 40 scientific presenta-

4 400 Florida Entomologist 77(4) December, 1994 tions during his tenure with ARS. Mr. Harrell is the senior scientist on six separate patents. Mr. Harrell is recognized as an expert in the mechanization of insect rearing and insect control. RESEARCH ACHIEVEMENTS The fall armyworm has been reared in the laboratory with a variety of techniques. The first cultures were maintained on foliage of corn, millet, bean and bean pods. Because of the demands for larger numbers of insects it was inevitable that meridic diets and mass-production rearing equipment be developed. Dr. Burton (Burton 1967) developed an artificial diet for fall armyworm and was probably among the first to rear it continuously on a meridic diet. Dr. Burton also developed a detailed description of the rearing procedures that included diet preparations, egg incubation, diet dispensing and manipulation of the larvae as well as some of the first estimates of mass rearing costs. About this same time Burton and Harrell began to develop original devices (Burton et al. 1966) for speeding up the procedures of mass rearing. The first two devices developed were the diet-dispensing and larvae-isolating systems. Then came the development of an automated packaging system (Burton & Cox 1966). The machine dispensed 1-oz plastic cups, each filled with a selected amount of artificial diet, dispensed larvae onto the diet and then, in one continuous process, capped the cup. Burton & Harrell (1966) then modified the larval dispensing machine to provide more stability, thus providing a much smoother operation. In 1968, Mr. Harrell and coworkers (Harrell et al. 1968) developed equipment and a mechanized system of collecting pupae of the fall armyworm from rearing containers. This entire system of rearing the fall armyworm, and various modifications thereof (Burton & Perkins 1989), has been used at the Southern Grain Insects Research Laboratory, which is now the Insect Biology and Population Management Research (IBPMRL) since about Mr. Harrell was instrumental in the erection of a 40 x 100 ft building which now houses the rearing section of at IBPMRL. The building was originally built for the purpose of housing a mass rearing system for the boll weevil (Harrell & Griffin 1981). A side note to this is that Mr. Harrell developed equipment for use on a form-fill-seal machine that was used in the mass production of insects. Harrell et al. (1973) developed an insect diet filler to be used on the form-fill-seal machine which could fill 32 cavities with diet in less than 2 sec. Mr. Harrell and co-workers (1974a) also built equipment that could mix and sterilize economically large amounts (up to 68 gallons per hour) of insect diet. Then, Mr. Harrell developed equipment for use on the formfill-seal machine (Harrell et al. 1974b) to infest cavities (cells) with insect eggs at rates up to 544 cavities per minute. Mr. Harrell later built two environmental rooms within the building to study environmental effects on growth of insects (Harrell et al. 1979). The rooms had separate air-distribution systems designed to maintain a uniform and constant temperature within ±1.1 C. HONORARIUM The scientists that have spent time in research at the IBPMRL, and others that have used the fall armyworm reared there, are deeply indebted to Dr. Robert L. Burton and Mr. E. A. Harrell for their tireless efforts in the development of a complete rearing system that has provided adequate numbers of quality fall armyworm eggs, larvae, pupae and adults.

5 Wiseman: Armyworm Symposium Therefore, it is with high regard and great pleasure that we dedicate this Armyworm Symposium to Dr. Robert L. Burton and Mr. E. A. Harrell in honor of their contributions to insect rearing. LITERATURE CITED BURTON, R. L Mass rearing the fall armyworm in the laboratory. U. S. Department of Agriculture. Agricultural Research Service. ARS pp. BURTON, R. L., AND H. C. COX An automated packaging machine for lepidopterous larvae. J. Econ. Entomol. 59: BURTON, R. L., AND E. A. HARRELL Modification of a lepidopterous larva dispenser for a packaging machine. J. Econ. Entomol. 59: BURTON, R. L., E. A. HARRELL, H. C. COX, AND W. W. HARE Devices to facilitate rearing of lepidopterous larvae. J. Econ. Entomol. 59: BURTON, R. L., AND W. D. PERKINS WSB, a new laboratory diet for the corn earworm and the fall armyworm. J. Econ. Entomol. 65: BURTON, R. L., AND W. D. PERKINS Rearing the corn earworm and the fall armyworm for maize resistance studies, pp in Toward insect resistant maize for the third world: Proc. International Symposium on Methodologies for Developing Host Plant Resistance to Maize Insects. CIMMYT. HARRELL, E. A., AND J. G. GRIFFIN Facility for mass rearing of boll weevils: Engineering aspects. U. S. Department of Agriculture. Science and Education Administration. Advances in Agricultural Technology. AAT-S pp. HARRELL, E. A., W. W. HARE, AND R. L. BURTON Collecting pupae of the fall armyworm from rearing containers. J. Econ. Entomol. 61: HARRELL, E. A., W. D. PERKINS, AND B. G. MULLINIX, JR Environmental rooms for insect rearing. Trans, ASAE. 1979: HARRELL, E. A., A. N. SPARKS, W. W. HARE, AND W. D. PERKINS. 1974a. Processing diets for mass rearing of insects. U. S. Department of Agriculture. Agricultural Research Service. ARS-S pp. HARRELL, E. A., A. N. SPARKS, W. D. PERKINS, AND W. W. HARE An insect diet filler for an in line form-fill-seal machine. J. Econ. Entomol. 66: HARRELL, E. A., A. N. SPARKS, W. D. PERKINS, AND W. W. HARE. 1974b. Equipment to place insect eggs in cells on a form-fill-seal machine. U. S. Department of Agriculture. Agricultural Research Service. ARS-S pp.

6 402 Florida Entomologist 77(4) December, 1994 REPRODUCTIVE POTENTIAL OF ONCE-MATED MOTHS OF THE FALL ARMYWORM (LEPIDOPTERA: NOCTUIDAE) C. E. ROGERS AND O. G. MARTI, JR. Insect Biology and Population Management Research Laboratory Agricultural Research Service, U. S. Department of Agriculture Tifton, GA ABSTRACT A laboratory study of the effects of age at a single mating on the reproductive potential of the fall armyworm, Spodoptera frugiperda (J. E. Smith), revealed that of 389 pairs of moths tested, 106, 281, and 2 pairs transferred 0, 1, or 2 spermatophores, respectively. Pairs that did not transfer a spermatophore mated when males and females averaged 7.0 and 8.8 days of age, respectively. Pairs transferring a single spermatophore mated when males and females averaged 5.9 and 6.3 days of age, respectively. The pairs transferring two spermatophores during one-night s pairing averaged 8 and 9 days of age at mating for males and females, respectively. The age of females at a single mating significantly affected their fecundity (r = -0.92; P<0.01), fertility (r = -0.61; P<0.01), and longevity (r = 0.83; P <0.01). Male age at a single mating significantly influenced only the fertility of eggs laid by their respective female partner (r = -0.92; P<0.01). Two days post-emergence was the optimum age for mating by both male and female moths for maximum fecundity and fertility. Delaying mating by females significantly lengthened their survival. Key Words: Spodoptera frugiperda, fecundity, fertility, longevity RESUMEN Un estudio de laboratorio acerca de los efectos de la edad en el momento del primer apareo sobre el potencial reproductivo del gusano trozador, Spodoptera frugiperda (J. E. Smith), reveló que de 389 parejas de polillas, 106, 281, y 2 parejas transfirieron 0, 1, y 2 espermatóforos, respectivamente. Las parejas que no transfirieron ningún espermatóforo se aparearon cuando los machos y hembras tenían una edad promedio de 7.0 y 8.8 días, respectivamente. Las parejas que transfirieron un solo espermatóforo se aparearon cuando los machos y las hembras tenían un promedio de 5.9 y 6.3 días edad, respectivamente. Las parejas que transfirieron dos espermatóforos durante un apareamiento de una noche tuvieron un promedio de 8 y 9 días de edad en los machos y hembras, respectivamente. La edad de las hembras en el momento del apareamiento afectó significativamente su fecundidad (r = -0.61; P <0.01), fertilidad (r = 0.83; P <0.01), y longevidad (r = 0.83; P< 0.01). La edad del macho en el momento del apareamiento influyó solamente en la fertilidad de los huevos puestos por su respectiva pareja (r = 0.60; P = 0.01). La edad óptima para el apareo fué de dos días después de la emergencia, tanto para el macho como para la hembra, y a esta edad se obtuvieron fecundidad y fertilidad máximas. El retardo en el apareo de las hembras alargó significativamnete su sobrevivencia. The reproductive potential, behavior, fecundity, and fertility of the fall armyworm, Spodoptera frugiperda (J. E. Smith), have been studied under a variety of natural and controlled environmental conditions (Luginbill 1928, Barfield & Ashley 1987, Sim- This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

7 Rogers and Marti: Armyworm Symposium mons & Lynch 1990, Simmons & Marti 1992, Simmons & Rogers 1994). Due to its perennial pest status in the southeastern United States, numerous strategies for managing the fall armyworm have been proposed, including the combination of two or more into regional management schemes (Knipling 1980). To accurately evaluate the efficacy of these strategies, it often is necessary to standardize one or more biological parameters of the fall armyworm in a controlled environment. We recently reported that the age of female moths at their first mating significantly affected their fecundity, fertility, and longevity (Rogers & Marti 1994). Here we report the effects of moth age at a single mating on the reproductive potential of the fall armyworm. MATERIALS AND METHODS Experimental insects were from an established laboratory colony of the fall armyworm that had been collected from corn and maintained on a pinto bean-based diet (Burton 1967, Perkins 1979). As moths emerged, they were placed individually in a 0.6-liter cardboard container, provided a 10% honey solution for nourishment, and maintained at 27 C and 70-75% RH in a 14:10 (L:D) cycle. To determine the effect of age at a single mating on fecundity and fertility of the fall armyworm, pairs of different male/female age combinations were established. The age of males and females at pairing ranged from 1-11 days and 1-14 days post-emergence, respectively. Tests encompassed 18 independent trials, each of which contained from 15 to 24 mating pairs. Mating pairs were maintained overnight in the conditions mentioned above in 0.6-liter cardboard cages with their tops and bottoms covered with mesh netting. Cages were lined with waxed paper to facilitate removal and counting of eggs. Males were discarded the morning after pairing. Females were maintained in their respective cages until they died. Cages were examined daily to record data on moth mortality/ longevity and the number of eggs laid. Dead females were dissected to determine the number of spermatophores they received from males while copulating. The number of unhatched eggs and viable larvae was used to compute daily fecundity and fertility for each mating pair. All data were subjected to an analysis of variance by the General Linear Model Procedure, and significantly different means were separated by the least significant difference (LSD) test (SAS Institute 1989). RESULTS AND DISCUSSION Significant differences (F = 4.97; df = 17, 386; P <0.01) occurred across trials for all parameters studied, indicating that the age of males and females at mating interacted to affect the reproductive potential of the fall armyworm (Fig. 1). Pairs composed of both young males and young females produced eggs having the highest fertility. When either sex was aged, fertility of the eggs was reduced. However, the age of females at mating had a greater influence on egg fertility than the age of males. Of the 389 pairs, 106 females received no spermatophores, 281 females received one spermatophore and two females received two spermatophores during a single night of pairing. Spermatophore transfer during moth mating may have been affected by female age (r = 0.71; P <0.01) but not by male age (r = 0.30; P <0.01) nor by the combination of male-female ages (Pearson correlation coefficient). Pairings resulting in the transfer of a single spermatophore involved males and females averaging 5.9 and 6.3 days of age, respectively. For pairs where no spermatophore was transferred, male and female age at mating averaged 7.0 and 8.8 days, respectively. Survival of females averaged 16.3 to 16.9 days, regardless of whether 0, 1, or 2 spermatophores were re-

8 404 Florida Entomologist 77(4) December, 1994 Figure 1. Interaction of male and female age at mating on egg fertility in oncemated fall armyworm moths. Data points represent means within trials. ceived. The fertility (P <0.05) of eggs from a single mating averaged 0, 20.4, and 45.5% for females receiving 0, 1, or 2 spermatophores, respectively. Total fecundity averaged 226.5, 1,134.7, and 1,026.5 eggs for females receiving 0, 1, or 2 spermatophores, respectively. Since this study concerns the reproductive potential for females restricted to a single opportunity for mating, analyses are reported for females which received only a single spermatophore. The age of females at a single mating significantly affected the number of eggs they laid (F = 8.21; d.f. = 10, 280; P <0.01), fertility of the eggs (F = 9.14; d.f. = 10, 280; P <0.01), and longevity of mated females (F = 10.24; d.f. = 10, 281; P <0.01) (Table 1). Females which mated at the age of 1-5 days laid more eggs than females which mated at an older age. Also, females which mated at days of age laid the fewest eggs (about 50% as many as females that mated from 1 to 5 days post-emergence) (Fig. 2). The mean fertility of eggs laid by females mated at 2 days of age was significantly greater than it was for eggs laid by younger or older females (Table 1). The lowest fertility occurred in eggs laid by females whose mating was delayed until 7 days and days post-emergence. Seven-day-old females mated with 9- and 11-day-old males.

9 Rogers and Marti: Armyworm Symposium TABLE 1. EFFECTS OF FEMALE AGE AT MATING ON FECUNDITY, FERTILITY, AND LONGEV- ITY OF FALL ARMYWORM (ONCE-MATED MOTHS). Female Age Total No. Eggs Laid % Eggs Hatching No. Days Lived (Days) N a/ (x ± S.E.) b,c/ (x ± S.E.) b,d/ (x ± S.E.) b,e/ ,387 ± 79a 28 ± 4b 15 ± 1c ,569 ± 115a 67 ± 6a 12 ± 1d ,402 ± 159a 29 ± 8b 14 ± 1cd ,269 ± 91ab 14 ± 5bc 15 ± 1c ,294 ± 123a 23 ± 7b 15 ± 1c ,150 ± 88b 9 ± 5c 17 ± 1b ,002 ± 120bc 25 ± 6b 15 ± 1c ± 113cd 6 ± 6c 21 ± 1a ± 147d 5 ± 8c 17 ± 1bc ± 98d 5 ± 5c 20 ± 1a ± 174d 15 ± 9bc 19 ± 1ab a/ Number of females per respective age across trials. b/ Means within a column followed by different letters are significantly different (P < 0.05) by the SAS LSD test. c/ Least significant difference = 341; r = (P <0.01) d/ Least significant difference = 18; r = (P <0.01) e/ Least significant difference = 3; r = 0.83 (P <0.0l) Ten-day-old females mated with 1- and 10-day-old males. Eleven-day-old females mated with 8-day-old males, while 13-day-old females mated with 4- and 10-day-old males. Surprisingly, females that mated one day post-emergence laid eggs that were Figure 2. Correlation between egg production and age at mating by once-mated females of the fall armyworm. Data points represent means within trials.

10 406 Florida Entomologist 77(4) December, 1994 less fertile than eggs laid by females mated 2 days post-emergence. However, 1-dayold females mated with 1-, 5-, and 11-day-old males, while 2-day-old females mated only with 2-day-old males. Fertility of eggs from a single mating remained relatively high (>80-97%) for about 7 days for pairs composed of 2-day-old males and females (Fig. 3). Fertility of eggs from pairs mated at 1 day of age declined to about 35% by day 3 post-mating. Fertility from 3-day-old pairs was below 50% and declined to about 20% by day 7 post-mating. Mating between 1-day-old females and 10-day-old males resulted in eggs with 96.99% fertility that remained >90% fertile from 2 through 8 days. The preceding data indicate that female age at mating is more important in determining fertility than male age. This is further indicated by a significant negative correlation between mean percent fertility and age of females at mating (Fig. 4), while the effects of male age on percent fertility is relatively benign (r = -0.32; P >0.10). The sum of male and female age at mating also significantly affected fertility of eggs on the second day post-mating (Fig. 5). As the sum of ages of mating males and females advanced beyond 4 days, the fertility of eggs declined as the combined ages of the pairs increased. Both the fecundity and fertility of fall armyworm females receiving a single spermatophore were less in this study than has been reported elsewhere (Vickery 1929, Simmons & Lynch 1990, Rogers & Marti 1994); however, our females were restricted to a single mating at a specific age. In considering only females which laid fertile eggs, both fecundity (x = 1, ± no. eggs per ) and fertility (x = 73.6 ± 6.8% eggs hatching) were higher for females mated at 2 days of age than they were for all females mated at the same age. Simmons & Marti (1992) reported that fall armyworm females mate an average of 3.7 times, and that two matings in a single night are common. Simmons & Rogers (1994) reported that females of the fall armyworm mate in as few as 11.5 hours after emergence, but that such matings result in eggs with a low fertility. Figure 3. Sustained fertility of eggs from pairs composed of once-mated, two-dayold fall armyworm males and females. Data points represent means within trials.

11 Rogers and Marti: Armyworm Symposium Figure 4. Effects of age of moths at mating on egg fertility two days post-mating in once-mated fall armyworm moths. Data points represent means within trials. Fertility of eggs laid by females which had mated with 2-day-old males was significantly (F = 8.75; d.f. = 10, 280; P <0.01) greater than the fertility of eggs laid by females which had mated with either younger or older males (Table 2). Fertility of eggs laid by females which had mated with males 8 days of age or older was extremely low Figure 5. Effects of summed age of mating pairs on fertility in once-mated fall armyworm moths. Data points represent means within trials.

12 408 Florida Entomologist 77(4) December, 1994 TABLE 2. EFFECTS OF MALE AGE AT MATING ON FECUNDITY AND FERTILITY IN THE FALL ARMYWORM (ONCE-MATED MOTHS). Male Age Total No. Eggs Laid % Eggs Hatching (Days) N a/ (x ± S.E.) b,c/ (x ± S.E.) b,d/ ,251 ± 94ab 24 ± 5b ,569 ± 124a 67 ± 6a ,402 ± 172a 29 ± 8b ± 99bc 17 ± 5bc ,183 ± 145b 25 ± 7b ,002 ± 130b 25 ± 6b ,131 ± 113b 22 ± 6b ± 159c 5 ± 8c ,099 ± 154b 0 ± 8c ,066 ± 73b 9 ± 4c ,117 ± 199b 0 ± 10c a/ Number of males per respective age across trials. b/ Means within a column followed by different letters are significantly different (P <0.05) by the SAS LSD test. c/ Least significant difference = 383; r = (P >0.10) d/ Least significant difference = 19; r = (P <0.01) (<9%). The age of females mating with 8- to 11-day-old males ranged from 1-13 days post-emergence. Although the effects of male age at mating on fall armyworm reproductive potential beyond the second day post-emergence were difficult to assess, the interaction of male and female ages appeared to contribute to egg fertility (Fig. 5). For example, combined age of 4 days for males and females at mating resulted in significantly higher fecundity (F = 5.6; d.f. = 10, 280; P <0.01) in mated females and significantly higher (>2x) egg fertility (F = 7.44; d.f. = 10, 280; P <0.01). Age combinations greater than days resulted in pairs producing fewer eggs with a lower fertility than pairs of younger males and females. The longevity of fall armyworm females was significantly affected by the age at which they mated (Table 1). Females that were denied mating until 10, 13, and 14 days of age lived significantly longer (x = 20.6 ± 1.8 days) than females which mated during the first 8 days post-emergence (x = 14.7 ± 1.1 days). The association of female longevity with age at mating is expressed as a highly significant correlation. This relationship gives credence to an oogenesis-flight syndrome in the fall armyworm that commonly is found in other migratory species of insects (Rankin & Burchsted 1992), in which early flight and reproduction are physiologically antagonistic. Such a physiologically antagonistic relationship has enabled the evolution of migration in the life history of some species, e.g., S. exempta (Walker) (Gatehouse 1986). Although the effects of male age at mating on the reproductive potential of the fall armyworm were less dramatic than the effects of female age, the age of males at mating nevertheless affected both fecundity of mated females, and the fertility of their eggs (Table 2). Females of varying age that were mated with 2- to 3-day-old males laid significantly (F = 3.17: d.f. = 10, 280; P <0.01) more eggs than females mating with older males. Male age at mating beyond three days post-emergence was less important as a contributor toward mated female fecundity. How young males may have contributed to the fecundity of their mates is unclear. However, Rankin & Burchsted

13 Rogers and Marti: Armyworm Symposium (1992) reported that males of a migratory grasshopper, Melanoplus sanguinipes (Fabricius), appear to transfer key proteins to females while mating that promote oogenesis and oviposition. Snow & Carlysle (1967) reported that unmated males of the fall armyworm incorporate a brownish-black pigment in the ductus ejaculatorius simplex that is transferred to females with a spermatophore on their first mating. Perhaps the brownish-black pigmented material transferred with a male s spermatophore contributes to the reproductive potential of mated females. If this pigmented material transferred from young males contributes to female fecundity, our data indicates that its importance diminishes in males which mate after day three post-emergence. Many factors (e.g., plant resistance, irradiation, chemosterilants, biological agents, number of matings, colony age, host strain, diet, pheromones, toxic chemicals, cultural strategies, etc.) are known to adversely affect the reproductive potential of the fall armyworm (Young et al. 1968, Lynch et al. 1980, Hamm & Hare 1982, Carpenter et al. 1986, Gross & Pair 1986, Silvain 1986, Simmons & Lynch 1990, Sen- Seong et al. 1985, Chandler & Sumner 1991, Quisenberry 1991, Pashley et al. 1992, and Rogers & Marti 1994). Knowledge of the individual and collective effects of these factors on the reproductive potential of the fall armyworm is critical for the implementation of an effective regional management strategy (Knipling 1980). Of equal importance is a knowledge of the effects of age at mating on the reproductive potential of the fall armyworm if efficacious propagation, treatment, and augmentation of insects are to be realized to support regional suppression of fall armyworm populations. The age of fall armyworm moths at mating when only a single opportunity for mating exists is critical for the sustainability of reproducing populations. Female age at a single mating is more important than male age for maintenance of a high reproductive potential. However, two days post-emergence for both males and females is the optimum age for mating for enhanced reproductive potential in the fall armyworm. REFERENCES CITED BARFIELD, C. S., AND T. R. ASHLEY Effects of corn phenology and temperature on the life cycle of the fall armyworm Spodoptera frugiperda (Lepidoptera: Noctuidae). Florida Entomol. 70: BURTON, R. L Mass rearing the fall armyworm in the laboratory. USDA, ARS , Gov. Print. Office, Wash., DC. CARPENTER, J. E., J. R. YOUNG, AND A. N. SPARKS Fall armyworm (Lepidoptera: Noctuidae): Comparison of inherited deleterious effects in progeny from irradiated males and females. J. Econ. Entomol. 79: CHANDLER, L. D., AND H. R. SUMNER Effect of various chemigation methodologies on suppression of the fall armyworm (Lepidoptera: Noctuidae) in corn. Florida Entomol. 74: GATEHOUSE, A. G Migration in the African armyworm, Spodoptera exempta, pp in Danthanarayana, W. [ed.]. Insect flight: dispersal and migration. Berlin/Heidelberg: Springer-Verlag, 289 pp. GROSS, H. R., AND S. D. PAIR The fall armyworm: status and expectations of biological control with parasitoids and predators. Florida Entomol. 69: HAMM, J. J., AND W. W. HARE Application of entomopathogens in irrigation water for control of fall armyworms and corn earworms (Lepidoptera: Noctuidae) on corn. J. Econ. Entomol. 75: KNIPLING, E. F Regional management of the fall armyworm - a realistic approach. Florida Entomol. 63: LUGINBILL, P The fall armyworm. U. S. Dept. Agric. Tech. Bull. 34. Gov. Print. Office, Wash., DC.

14 410 Florida Entomologist 77(4) December, 1994 LYNCH, R. E., P. B. MARTIN, AND J. W. GARDNER Cultural manipulation of coastal bermudagrass to avoid losses from the fall armyworm. Florida Entomol. 63: PASHLEY, D. P., A. M. HAMMOND, AND T. N. HARDY Reproductive isolating mechanisms in fall armyworm host strains (Lepidoptera: Noctuidae). Ann. Entomol. Soc. America 85: PERKINS, W. D Laboratory rearing of the fall armyworm. Florida Entomol. 62: QUISENBERRY, S. S Fall armyworm (Lepidoptera: Noctuidae) host strain reproductive compatibility. The Florida Entomol. 74: RANKIN, M. A, AND J. C. A. BURCHSTED The cost of migration. Annu. Rev. Entomol. 37: ROGERS, C. E., AND O. G. MARTI, JR Effects of age at first mating on the reproductive potential of the fall armyworm (Lepidoptera: Noctuidae). Environ. Entomol. 23: SAS INSTITUTE SAS/STAT User s guide, version 6, 4th ed., vol. 1 and vol. 2. SAS Institute, Cary, NC. SEN-SEONG, N. G., F. M. DAVIS, AND W. P. WILLIAMS Survival, growth, and reproduction of the fall armyworm (Lepidoptera: Noctuidae) as affected by resistant genotypes. J. Econ. Entomol. 78: SILVAIN, J. F Use of pheromone traps as a warning system against attacks of Spodoptera frugiperda larvae in French Guiana. Florida Entomol. 69: SIMMONS, A. M., AND R. E. LYNCH Egg production and adult longevity of Spodoptera frugiperda, Helicoverpa zea (Lepidoptera: Noctuidae), and Elasmopalpus lignosellus (Lepidoptera: Pyralidae) on selected adult diets. Florida Entomol. 73: SIMMONS, A. M., AND O. G. MARTI, JR Mating by the fall armyworm (Lepidoptera: Noctuidae): frequency, duration, and effect of temperature. Environ. Entomol. 21: SIMMONS, A. M., AND C. E. ROGERS Fall armyworm (Lepidoptera: Noctuidae) mating: effects of age and scotophase on pre-mating time, mating incidence, and fertility. J. Entomol. Science 29: SNOW, J. W., AND T. C. CARLYSLE A characteristic indicating the mating status of male fall armyworm moths. Ann. Entomol. Soc. America 60: VICKERY, R. A Studies on the fall armyworm in the Gulf Coast of Texas. U. S. Dept. Agric. Tech. Bull Gov. Print. Office, Wash., DC. YOUNG, J. R., J. W. SNOW, AND A. N. SPARKS Mating of untreated and tepachemosterilized fall armyworm moths. J. Econ. Entomol. 61:

15 Chandler: Armyworm Symposium EVALUATION OF INSECT GROWTH REGULATOR-FEEDING STIMULANT COMBINATIONS FOR MANAGEMENT OF FALL ARMYWORM (LEPIDOPTERA: NOCTUIDAE) L. D. CHANDLER 1 Insect Biology and Population Management Research Laboratory Agricultural Research Service, U. S. Department of Agriculture Tifton, GA Current address: USDA-ARS, R.R. 3, Brookings, SD ABSTRACT Application of RH F, a neural agonist insect growth regulator, to foliage of corn (Zea mays [L.]) and southern pea (Vigna unguiculata [L.]), resulted in significant levels of fall armyworm (Spodoptera frugiperda [J. E. Smith]) larval control. The addition of a cottonseed flour-based feeding stimulant, Konsume, at 10% of total spray volume to the insect growth regulator reduced the length of time needed for larval mortality to occur compared to the insect growth regulator alone. Additionally, feeding damage and percentage of corn plants infested by fall armyworm was significantly reduced using RH F + Konsume. Residual activity of RH-5992 remained high (> 84% larval mortality at the highest evaluated rate of RH-5992) on southern pea throughout the two-week test period. Key Words: Spodoptera frugiperda, insect management, insecticide additives RESUMEN La aplicación de RH F, un agonista neural regulador del crecimiento de los insectos, al follaje del maíz (Zea mays [L.]) y del guisante sureño (Vigna unguiculata [L.]), produjo niveles significativos de control larval del gusano trozador (Spodoptera frugiperda [J.E.Smith]). La adición al regulador de crecimiento de un estimulante de la alimentación a base de harina de algodón, Kosume, al 10% del volumen total de aspersión, redujo el tiempo requerido para matar las larvas en comparación con el regulador de crecimiento solo. Adicionalmente, el daño por alimentación de las larvas y el porcentaje de plantas de maíz infestadas por el gusano trozador fueron significativamente reducidos usando RH F + Kosume. La actividad residual del RH-5992 permaneció alta (>84% de mortalidad larval en la concentración más alta de RH- 5992) en el guisante sureño durante las dos semanas de prueba. Use of phagostimulants to enhance activity of insect growth regulators against fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae), in the laboratory has been reported (Chandler 1993). The mortality rate of FAW larvae increased significantly when they fed on leaves of southern pea, Vigna unguiculata (L.), that had been treated with either of two insect growth regulators (RH-5992 or diflubenzuron), in combination with Konsume R, a cottonseed flour-based insect feeding stimulant. Larval mortality was increased 2 and 3 days after treatment when Konsume (10% of total solution) and RH-5992 were applied in combination (Chandler 1993). This increase in mortality was more notable than that observed with diflubenzuron/konsume combinations. The increased mortality during the first three days following treatment with RH-5992 is important because it can reduce FAW related This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

16 412 Florida Entomologist 77(4) December, 1994 damage sooner following treatment, which may aid acceptance of this particular type of insect growth regulator by growers. RH-5992 interferes with the normal molting process of lepidopterous larvae by acting as an agonist of insect molting hormone (Rohm and Haas Co. 1989). It has been shown to have high (> 90%) levels of activity against FAW larvae at concentrations of to 1.0% active ingredient [AI] (Chandler et al. 1992). The studies reported here were designed to evaluate the use of RH-5992/Konsume combinations under simulated and actual field conditions for management of FAW. Studies were conducted to evaluate the residual effectiveness and control potential of RH-5992, with and without the addition of the feeding stimulant, and to compare the effectiveness of RH-5992 with thiodicarb, a standard chemical insecticide used for control of FAW larvae. MATERIALS AND METHODS Spray Table Study RH F (Rohm and Haas Co., Philadelphia, Pa.) was formulated in water at the rates of 11.0 and 22.0 gm AI/ha. Both of these formulations were sprayed on southern peas, with and without the addition of Konsume (Fermone Inc., Phoenix, AZ) (10% of total volume) to the mixture. In addition, peas were treated with a Konsume (10% of total volume)-water mixture and with water alone. The peas were planted in 25 x 61 x 5 cm rectangular plastic trays filled with potting media, and these were held in a greenhouse (24-30 C and % relative humidity [RH]). In the greenhouse the plants were allowed to reach the 2 leaf stage (approximately 2 weeks old) prior to treatment. On 17 May, 1993, 24 trays of peas per treatment (6 total treatments) were sprayed with the above mixtures. The trays were placed on a conveyor-driven spray table calibrated to apply 38.2 liter per ha at 59 kg/cm 2 with a single TX-6 cone nozzle fixed 46 cm above the crop. The table was set to convey the plants at a speed of 15.2 m per min, and the spray nozzle was powered by compressed air. Following treatment, the trays of peas were returned to the greenhouse. Leaves were collected from the treated plants 2 h and 1, 3, 7, 10, and 14 days after treatment. Four trays of cowpeas were used on each collection date for each treatment. Each tray was designated as a separate replicate. Neonate FAW larvae were obtained from laboratory cultures at the Insect Biology and Population Management Research Laboratory in Tifton, GA. These larvae were reared as described by Perkins (1979). The larvae were held on pinto bean diet at 24 ± 1 C for 3 days before being exposed to treated leaf surfaces thus allowing all insects to reach a uniform age and size. On each leaf collection date, 30 leaves per replicate (4 replicates total per collection date) per treatment were placed in 30-ml plastic cups. Single 3-day-old larvae were placed on each leaf and the cup was capped. Cups containing larvae were held in environmental cabinets in the laboratory at 24 ± 1 C, a photoperiod of 12:12 (L:D), and 50 ± 5% RH. After 48 h, cups were opened and the number of living versus dead larvae was recorded. Surviving larvae were then placed in 30-ml cups containing 10 ml of bean diet and the number of living versus dead larvae recorded following an additional 3 days (120 h after application mortality). Larvae were then disposed of. Field Study Field corn (Zea mays [L.]) was planted in August 1993 at the USDA-ARS Belflower Farm in Tift Co., GA. Nine treatments, including an untreated control, were arranged

17 Chandler: Armyworm Symposium in a randomized block design with four replications. Plots were 3 rows (91 cm per row) wide by 7.6 m long. Treatments consisted of RH F applied at rates of 22.0, 45.9 and 91.8 gm AI/ha with and without the addition of Konsume at 10% of the total volume. A 10% solution of Konsume alone and thiodicarb (Larvin 3.2) at 45.9 gm AI/ha were also evaluated. Three treatments were made to whorl stage corn beginning 23 August and continued every 10 days (final treatment on 13 September). Insecticide applications were made with a CO 2 backpack sprayer calibrated to apply 30.6 liter per ha at 79 kg per cm 2 using a TX-6 nozzle and traveling at 3.2 km/hr. Prior to and every 3-5 days after initial treatment through the test period, the number of damaged plants (15 plants examined per plot) were counted. Counts were terminated on 1 October. Damaged plants were defined as those where fresh larval feeding damage and frass were found. The number of plants damaged per plot were converted to percentages for data analysis. Additionally, damage ratings were taken beginning on the second count date following treatment. Ratings consisted of examining the entire plot for FAW feeding damage on a scale of 0-5, in which 0 = no damage, 1 = 1-20% damage, 2 = 21-40% damage, 3 = 41-60% damage, 4 = 61-80% damage, and 5 = % damage. Data Analyses Means and standard deviations were calculated for percent mortality, percentage of plants damaged, and damage ratings, both by date and as combined averages (after initial treatment) over the length of the experiments. Analyses of variance (PROC GLM, SAS 1985) were conducted for all data sets. For the spray table study, set least square means were used to compare all possible treatment combinations. Contrasts were made between individual rates of RH-5992 with and without Konsume, between RH-5992 treatments combined with and without Konsume, between the untreated control and Konsume only, and between the untreated control and all other treatments for 48 and 120 h mortality data. For the field study, orthogonal contrasts were made to determine whether the tested rates of RH-5992, with or without Konsume, resulted in either a linear or quadratic response in the number of damaged plants and damage ratings per treatment. Comparisons were also made between RH-5992 with and without the addition of Konsume, Konsume alone vs the untreated control, RH vs thiodicarb, and RH-5992 vs Konsume alone. RESULTS AND DISCUSSION Spray Table Study High levels of fall armyworm larval mortality were observed following application of RH F with and without the addition of Konsume to the mix (Tables 1 and 2). Mortality ranged from 5.9 to 67.5 % and from 62.5 to 98.4 %, depending upon the date and rate of active ingredient, 48 and 120 hours after leaf collection of treated leaves, respectively. Over the entire 14-day leaf collection period, no trend in loss of RH-5992 residual activity could be discerned (Tables 1 and 2). Some mortality of fall armyworm larvae was noted following feeding on foliage treated with Konsume alone. However, the highest Konsume related mortality rates (17.5 %), compared to othertreatments, were not considered to be of great importance in inflicting death to the insect (Table 2).

18 414 Florida Entomologist 77(4) December, 1994 TABLE 1. RESIDUAL MORTALITY OF FALL ARMYWORM LARVAE FOLLOWING APPLICATION OF RH-5992 WITH OR WITHOUT KONSUME (10% OF TOTAL VOLUME) TO COWPEA FOLIAGE; 48 HOUR MORTALITY (%) AFTER FOLIAGE COLLECTED ON INDICATED DATE. 1 x ± SD Mortality (%) at Indicated Leaf Collection Time after Treatment Treatment gms AI/ha 2 Hour 1 Day 3 Days 7 Days 10 Days 14 Days Study Average RH F ± ± ± ± ± ± ± 13.9 RH F ± ± ± ± ± ± ± 13.3 RH F + Konsume ± ± ± ± ± ± ± 20.7 RH F + Konsume ± ± ± ± ± ± ± 15.5 Konsume ± ± ± ± ± ± ± 3.1 Untreated ± ± ± ± ± ± ± Application made on 17 May.

19 Chandler: Armyworm Symposium TABLE 2. RESIDUAL MORTALITY OF FALL ARMYWORM LARVAE FOLLOWING APPLICATION OF RH-5992 WITH OR WITHOUT KONSUME (10% OF TOTAL VOLUME) TO COWPEA FOLIAGE; 120 HOUR MORTALITY (%) AFTER FOLIAGE COLLECTED ON INDICATED DATE. 1 x ± SD Mortality (%) at Indicated Leaf Collection Time after Treatment Treatment gms AI/ha 2 Hour 1 Day 3 Days 7 Days 10 Days 14 Days Study Average RH F ± ± ± ± ± ± ± 5.0 RH F ± ± ± ± ± ± ± 11.7 RH F + Konsume ± ± ± ± ± ± ± 7.1 RH F + Konsume ± ± ± ± ± ± ± 13.1 Konsume ± ± ± ± ± ± ± 8.5 Untreated ± ± ± ± ± ± ± Application made on 17 May.

20 416 Florida Entomologist 77(4) December, 1994 Analysis of variance for 48 h after leaf collection mortality indicated a significant interaction between date and treatment (F = 7.74, df = 25, P = ). Similar results were obtained following analysis of variance for 120 hour after leaf collection mortality (F = 3.56, df = 25, P = ). These interactions indicated that treatments responded differently on each leaf collection date. Therefore, comparisons among treatments (least square means) were made on individual leaf collection dates. Contrasts were then conducted on seasonal average data to determine overall effects of the insect growth regulator/feeding stimulant mixture on fall armyworm mortality. Least square means resulted in significant differences in treatment comparisons based on time of mortality observations and length of time after treatment (Tables 3 and 4). In most instances mortality from the untreated control and 10% Konsume alone treatments was significantly less than all other treatments for both 48 and 120 h mortality on each leaf collection date (Tables 3 and 4). Significant increases in mortality caused by RH Konsume and RH-5992 alone treatments were most often noted within 7 days of treatment and primarily with 48 h mortality data. Data indicated that 48 h readings showed an increase in mortality within the first 7 days following treatment when Konsume was added to the insect growth regulator. Mortality observed after 120 h was similar among most insect growth regulator treatments. After 7 days, no increase in mortality could be attributed to Konsume in the formulations, but mortality rates remained high (Tables 3 and 4). RH-5992 (11.0 gms AI/ha) + Konsume resulted in significantly less mortality 14 days after application (120 h mortality) than was noted with other RH-5992 treatments. Contrasts conducted for 48 and 120 h after leaf collection summary data indicated that increasing rates of RH-5992, with and without Konsume, significantly affected fall armyworm larval mortality. RH-5992 resulted in higher mortality at the 22.0 gm AI/ha rate than at the 11.0 gm AI/ha rate on both leaf mortality observation dates (Tables 5 and 6). Also, the addition of Konsume to RH-5992 resulted in higher levels of fall armyworm larval mortality, both 48 and 120 h after leaf collection, than that achieved with RH-5992 alone, regardless of insect growth regulator active ingredient rate (Tables 5 and 6). Konsume alone resulted in significantly higher levels of mortality than the untreated control, and the untreated control had significantly less larval mortality than all of the other treatments combined 48 and 120 h after leaf collection. Field Study Percentage of corn plants damaged by fall armyworm prior to initial insect growth regulator treatment (23 Aug.) ranged from 3-22% per treatment (Table 7), but no significant differences in damaged plants were observed on this date although the data are quite variable. Damage increased in the untreated plots through 13 Sept. (Tables 7 and 8). Populations of fall armyworm then decreased as indicated by the reduced percentage of plants damaged. Damage again increased from 17 Sept. until 1 Oct. (Tables 7 and 8). Fewer plants were damaged in the plots treated with RH F at 91.8 gm AI/ha mixed with 10% Konsume; percentage of plants damaged ranged from 2 to 32%. There were no significant interactions between date and treatment for percentage of plants damaged and damage ratings (F=0.87, df=88, P = and F=0.52, df=72, P = , respectively). All dates were combined for orthogonal contrasts. Increasing rates of RH-5992, both with and without the addition of Konsume, resulted in a linear response for both reduction of percentage of plants damaged and plant damage ratings caused by fall armyworm larval feeding (Tables 9 and 10). A quadratic relationship was not indicated. The addition of Konsume to RH-5992 re-

21 Chandler: Armyworm Symposium TABLE 3. T VALUES FOR LEAST SQUARE MEANS COMPARISONS OF ALL POSSIBLE TREATMENT PAIRS 48 HOUR AFTER LEAF COLLECTED MORTALITY. t-values at Leaf Collection Time after Treatment 1 Treatment gms AI/ha Treatment Pairs 2 Hour 1 Day 3 Days 7 Days 10 Days 14 Days 1. Untreated Konsume * -3.14* -8.90* -3.30* -3.12* 3. RH * -8.84* -4.99* -8.39* -4.71* * -9.61* -3.26* -8.06* -4.34* RH * * -5.56* * -5.21* * -3.14* -8.90* -3.30* -3.12* 5. RH Konsume * -8.17* -4.99* -8.39* -4.71* * -8.94* -3.26* -8.06* -4.34* RH Konsume * * -5.56* * -5.21* * * -4.60* * -7.85* * * -4.01* * * -3.25* * *Indicates significant difference in treatment pairs, critical point for all comparisons is t > 3.01.

22 418 Florida Entomologist 77(4) December, 1994 TABLE 4. T VALUES FOR LEAST SQUARE MEANS COMPARISONS OF ALL POSSIBLE TREATMENT PAIRS 120 HOUR AFTER LEAF COLLECTION MORTALITY. t-values at Leaf Collection Time after Treatment 1 Treatment gms AI/ha Treatment Pairs 2 Hour 1 Day 3 Days 7 Days 10 Days 14 Days 1. Untreated * Konsume * * * * * * 3. RH * * * * * * 4. RH * * * * * * 5. RH Konsume * * * * * * * * -9.84* * * * 6. RH Konsume * * * * * * * * * * * * * * * * * * * * * * -3.12* * * * * 1 *Indicates significant difference in treatment pairs, critical point for all comparisons is t > 3.01.

23 Chandler: Armyworm Symposium TABLE 5. ANALYSIS OF VARIANCE TABLE FOR 48 HOUR AFTER LEAF COLLECTION FALL AR- MYWORM LARVAE MORTALITY. DF Sum of Squares Mean Square F Value Pr > F Model Error Corrected Total Type II DF Sum of Squares Mean Square F Value Pr > F Date Rep (Date) ) RH vs Gm Ai/Ha ) RH Konsume 22.0 vs Gm Ai/Ha ) RH Konsume vs. RH ) Konsume vs. Untreated ) All Treatments vs. Untreated ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date sulted in significantly less feeding damage than when RH-5992 was used alone (Tables 9 and 10). No differences in percentages of plants damaged or damage ratings were observed comparing RH-5992 with thiodicarb (Larvin 3.2) (Tables 9 and 10). Differences were observed when comparing plants treated with all rates of RH-5992 and those treated with Konsume alone or with untreated plants. RH-5992 treated plants had significantly less damage than did plants treated with Konsume alone. Plants treated with Konsume resulted in a significant reduction in damage compared to the untreated control. However, plants treated with Konsume alone did not provide needed levels of economic fall armyworm control. These findings are similar to those observed with the spray table test. In conclusion, use of RH F resulted in significant levels of fall armyworm larval control on both southern pea and field corn. Addition of Konsume (at 10% of total volume) to the insect growth regulator provided significantly greater fall armyworm mortality in a shorter period of time and significantly reduced fall armyworm feeding damage compared to use of the insect growth regulator alone. These results further confirmed the laboratory findings of Chandler (1993) which indicated that the use of a cottonseed flour-based insect feeding stimulant enhanced the activity of RH-

24 420 Florida Entomologist 77(4) December, 1994 TABLE 6. ANALYSIS OF VARIANCE TABLE FOR 120 HOUR AFTER LEAF COLLECTION FALL ARMYWORM LARVAE MORTALITY. DF Sum of Squares Mean Square F Value Pr > F Model Error Corrected Total Type II DF Sum of Squares Mean Square F Value Pr > F Date Rep (Date) ) RH vs Gm Ai/Ha ) RH Konsume 22.0 vs Gm Ai/Ha ) RH Konsume vs. RH ) Konsume vs. Untreated ) All Treatments vs. Untreated ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date and decreased time of larval mortality. Furthermore, spray table tests indicated that the residual activity of RH-5992 in a controlled environment remained relatively high throughout a two-week period. These positive results demonstrated the usefulness of insect growth regulator/feeding stimulant combinations for use in field settings. Further study is needed to refine insect growth regulator rates and to determine the economic feasibility of adding feeding stimulants to insect growth regulator mixtures. ACKNOWLEDGMENTS The author wishes to thank Lenny Atkins, Steve Hooks, Laura Abbott, Larry Walker, Wendy Tyson, and Matt Wauchope for their technical assistance in conducting these studies. Richard Layton is thanked for his help in conducting the statistical analyses of the data. Proprietary names are necessary to report factually on available

25 Chandler: Armyworm Symposium TABLE 7. PLANTS DAMAGED (%) BY FALL ARMYWORM LARVAE FOLLOWING APPLICATION OF RH F WITH OR WITHOUT KONSUME (10% OF TO- TAL VOLUME) TO WHORL STAGE CORN. 1 x ± SD Damaged Plants (%)/Treatment Treatment gm AI/ha 23 Aug Pretreatment 26 Aug 30 Aug 2 Sept 6 Sept 9 Sept 13 Sept 17 Sept 21 Sept 23 Sept 28 Sept 1 Oct Seasonal Average RH F ± ± 8 28 ± 6 53 ± ± ± ± ± ± ± 5 5 ± 6 30 ± ± 19 RH F ± ± ± ± ± 7 42 ± ± ± ± 5 15 ± 8 12 ± ± ± 18 RH F ± ± ± ± ± ± ± ± ± 8 22 ± 6 18 ± 6 47 ± ± 24 RH F + Konsume ± 4 32 ± ± 9 18 ± 6 25 ± 3 18 ± ± 4 10 ± 7 12 ± 15 7 ± 5 2 ± 3 10 ± ± 12 RH F + Konsume ± ± ± ± ± ± ± ± ± ± ± 9 27 ± ± 23 RH F + Konsume ± 9 52 ± 6 48 ± ± ± ± 9 35 ± ± ± ± ± ± ± 19 Konsume ± 7 48 ± ± ± ± ± ± ± ± ± ± ± ± 28 Larvin ± ± ± ± ± ± ± ± ± ± 9 13 ± 9 45 ± ± 24 Untreated ± ± ± ± ± ± ± ± ± ± 6 13 ± 9 35 ± ± 28 1 Applications made 23 Aug, 3 and 13 Sept.

26 422 Florida Entomologist 77(4) December, 1994 TABLE 8. DAMAGE RATINGS RESULTING FROM FEEDING BY FALL ARMYWORM LARVAE FOLLOWING APPLICATION OF RH F WITH OR WITHOUT KONSUME (10% OF TOTAL VOLUME) TO WHORL STAGE CORN. 1 x ± SD Damage Rating/Treatment Treatment gm AI/ha 30 Aug 2 Sept 6 Sept 9 Sept 13 Sept 17 Sept 21 Sept 23 Sept 28 Sept 1 Oct Seasonal Average RH F ± ± ± ± ± ± ± ± ± ± ± 0.9 RH F ± ± ± ± ± ± ± ± ± ± ± 0.8 RH F ± ± ± ± ± ± ± ± ± ± ± 1.1 RH F + Konsume ± ± ± ± ± ± ± ± ± ± ± 0.9 RH F + Konsume ± ± ± ± ± ± ± ± ± ± ± 1.1 RH F + Konsume ± ± ± ± ± ± ± ± ± ± ± 1.0 Konsume ± ± ± ± ± ± ± ± ± ± ± 1.1 Larvin ± ± ± ± ± ± ± ± ± ± ± 1.0 Untreated ± ± ± ± ± ± ± ± ± ± ± Applications made 23 Aug, 3 and 13 Sept.

27 Chandler: Armyworm Symposium TABLE 9. ANALYSIS OF VARIANCE TABLE FOR PERCENTAGE OF PLANTS DAMAGED BY FALL ARMYWORM LARVAE, ORTHOGONAL CONTRASTS. DF Sum of Squares Mean Square F Value Pr > F Model Error Corrected Total Type II DF Sum of Squares Mean Square F Value Pr > F Date Rep (Date) ) RH-5992-Linear ) RH-5992-Quad ) RH Konsume-Linear ) RH Konsume-Quad ) RH-5992 vs. RH Konsume ) Konsume vs. Untreated ) RH-5992 vs. Larvin ) RH-5992 vs. Konsume ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date TABLE 10. ANALYSIS OF VARIANCE TABLE FOR DAMAGE RATINGS RESULTING FROM FEEDING BY FALL ARMYWORM LARVAE, ORTHOGONAL CONTRASTS. DF Sum of Squares Mean Square F Value Pr > F Model Error Corrected Total

28 424 Florida Entomologist 77(4) December, 1994 TABLE 10. (CONTINUED) ANALYSIS OF VARIANCE TABLE FOR DAMAGE RATINGS RESULT- ING FROM FEEDING BY FALL ARMYWORM LARVAE, ORTHOGONAL CONTRASTS. Type II DF Sum of Squares Mean Square F Value Pr > F Date Rep (Date) ) RH-5992-Linear ) RH-5992-Quad ) RH Konsume-Linear ) RH Konsume-Quad ) RH-5992 vs. RH Konsume ) Konsume vs. Untreated ) RH-5992 vs. Larvin ) RH-5992 vs. Konsume ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date ) Interacted with Date data; however, the USDA neither guarantees nor warrants the product, and the use of the name by USDA implies no approval of the product to the exclusion of others that may be suitable. U.S. Department of Agriculture, Agricultural Research Service, Northern Plains Area, is an equal opportunity/affirmative action employer and all agency services are available without discrimination REFERENCES CITED CHANDLER, L.D Use of feeding stimulants to enhance insect growth regulatorinduced mortality of fall armyworm (Lepidoptera: Noctuidae) larvae. Florida Entomol. 76: CHANDLER, L.D., S.D. PAIR, AND W.E. HARRISON RH-5992: A new insect growth regulator active against corn earworm and fall armyworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 85: PERKINS, W.D Laboratory rearing of the fall armyworm. Florida Entomol. 62: ROHM AND HAAS CO RH-5992 insect growth regulator. Technical Information Bulletin AG pp. SAS INSTITUTE SAS/STAT User s Guide. SAS Institute, Cary, NC.

29 This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL This document was created with FrameMaker Hamm et al.: Armyworm Symposium FIELD TESTS WITH A FLUORESCENT BRIGHTENER TO ENHANCE INFECTIVITY OF FALL ARMYWORM (LEPIDOPTERA: NOCTUIDAE) NUCLEAR POLYHEDROSIS VIRUS J. J. HAMM, L. D. CHANDLER AND H. R. SUMNER Insect Biology and Population Management Research Laboratory U.S. Department of Agriculture, Agricultural Research Service Tifton, GA ABSTRACT The nuclear polyhedrosis virus (NPV) of fall armyworm, Spodoptera frugiperda (J. E. Smith), was applied in combination with Fluorescent Brightener 28 (Calcofluor White M2R, Tinopal LPW) to whorl-stage corn. Concentrations of NPV ranged from 5 larval equivalents (1 LE = 6 x 10 9 polyhedral occlusion bodies) to 1235 LE per ha. Concentrations of fluorescent brightener ranged from 0.1 to 5% by weight in water and the water volume ranged from 234 to 926 liters per ha. Two days after treatment, fall armyworm larvae were collected from the treated plants and held on bean diet to observe mortality due to NPV, parasitoids, and ascovirus. The fluorescent brightener interacted significantly with virus concentration and with water volume to increase larval mortality. There was no increase in mortality due to NPV as the percent fluorescent brightener increased beyond 1%. In the higher volumes of water, 0.25% fluorescent brightener resulted in the highest percent mortality due to NPV. Cotesia marginiventris was the most abundant parasitoid recovered from fall armyworm in these tests, and as the percent mortality due to NPV increased, the percent mortality due to parasitoids and ascovirus decreased. Thus, the total mortality was not affected as greatly as the percent mortality due to NPV by changes in water volume or fluorescent brightener concentration. The reduction in mortality due to parasitoids did not appear to be a direct effect of the fluorescent brightener on the parasitoids. However, increased infectivity of the NPV and earlier mortality from NPV associated with the fluorescent brightener resulted in more host larvae dying of NPV before the parasitoids could complete development. Key Words: Spodoptera frugiperda, nuclear polyhedrosis virus, fluorescent brightener, biocontrol, corn, Cotesia marginiventris RESUMEN El virus de la polihedrosis nuclear (VPN) del gusano trozador, Spodoptera frugiperda (J. E. Smith), fue aplicado en combinación con Fluorescent Brightener 28 (Calcofluor White M2R, Tinopal LPW) a plantas de maíz en estado vegetativo. Las concentraciones del VPN estuvieron en el rango de los 5 a 1235 equivalentes larvales (EL) por ha (1 EL = 6 x 10 9 cuerpos polihedrales de oclusión). Las concentraciones de Fluorescent Brightener estuvieron en el rango de 0.1 a 5% por peso en agua y el volumen del agua en el rango de 234 a 926 litros por hectárea. Dos días después del tratamiento las larvas del gusano trozador fueron colectadas de las plantas tratadas y mantenidas en dieta de frijoles para observar la mortalidad debida al VPN, parasitoides, y ascovirus. Fluorescent Brightener interactuó significativamente con la concentración del virus y con el volumen del agua para aumentar la mortalidad larval. No hubo aumento de la mortalidad debido al VPN cuando el porcentaje de Fluorescent Brightener aumentó a más del 1%. En los volúmenes más altos de agua, el 0.25% de Fluorescent Brightener produjo el porcentaje de mortalidad más alto. Cotesia marginiventris fue el parasitoide más abundantemente recobrado del gusano trozador en

30 426 Florida Entomologist 77(4) December, 1994 estas pruebas y en la medida en que aumentó el porcentaje de mortalidad debido al VPN, el producido por los parasitoides y ascovirus disminuyó. De esta manera, la mortalidad total no fue afectada tanto como el porcentaje de mortalidad debido al VPN por los volúmenes de agua o la concentración de Fluorescent Brightener. La reducción en la mortalidad producida por los parasitoides no pareció deberse al efecto del marcador en los mismos. Sin embargo, el incremento de la infectividad del VPN y la mortalidad temprana debidos al virus asociado con el Fluorescent Brightener provocaron que más larvas murieran por el VPN antes que los parasitoides pudieran completar su desarrollo. The nuclear polyhedrosis virus (SfNPV) is a naturally occurring pathogen of fall armyworm, Spodoptera frugiperda (J. E. Smith), (Gardner & Fuxa 1980, Fuxa 1982). However, field tests with SfNPV have resulted in rather low levels of control of fall armyworm larvae (Hamm & Young 1971, Hamm & Hare 1982). In these earlier tests, the virus was not formulated with adjuvants to protect the virus from sunlight or to otherwise enhance its infectivity. Recently, however, Shapiro (1992) demonstrated that UV protection was possible using a series of optical or fluorescent brighteners (FB). More importantly, five of the optical brighteners, including Tinopal LPW, enhanced the infectivity of an NPV that infects gypsy moth larvae, even when the virus was not exposed to UV irradiation (Shapiro & Robertson 1992). Later, Hamm & Shapiro (1992) demonstrated significant enhancement of the SfNPV by Tinopal LPW in laboratory bioassays. Because of this unique enhancement of viral infectivity for lepidopterous larvae, a patent for the use of fluorescent brighteners in biological control was awarded 23 June 1992 (Shapiro et al. 1992). The field tests reported here were conducted to determine if adding FB to the SfNPV would increase the level of control of fall armyworm larvae in whorl-stage corn. Because parasitoids (Ashley 1986) and ascovirus (Hamm et al. 1986) contribute to the natural control of fall armyworm, the effects of SfNPV and FB on parasitoids and ascovirus were studied also. Control of fall armyworm on corn is difficult, even with insecticides, because the larvae feed down into the whorl. Therefore, the insecticides are generally applied in the maximum amount of water that can be applied economically with ground equipment ( liters per ha). Consequently, the relationship between water volume, SfNPV, and FB was also evaluated. MATERIALS AND METHODS Materials The SfNPV was produced in the laboratory in fall armyworm larvae. The polyhedral occlusion bodies (POB) were partially purified by slow and high speed centrifugation and suspended in 0.1 m phosphate buffer (ph 7) containing 100 µg/ml garamycin and stored at 6 C. The virus was quantified by counting POB with a Petroff-Houser bacterial counter. Concentrations were expressed as Larval Equivalents (LE), the approximate number of POB produced per larva, based on 6 X 10 9 POB per LE. Fluorescent Brightener 28 (Calcofluor white M2R, Tinopal LPW) was obtained from Sigma Chemical Co. No other UV screens, wetting agents, or feeding stimulants were used.

31 General Procedures Hamm et al.: Armyworm Symposium Tests were conducted in whorl-stage corn in Tift Co. GA. Plots were single rows 6 to 7.6 m long separated by 6 m of untreated corn on the ends and five rows of untreated corn (4.6 m) on the sides. There were five replications of each treatment arranged in randomized complete blocks. Corn plants were artificially infested in early season tests conducted in May and June of 1989 and Newly hatched fall armyworm larvae from the Insect Biology and Population Management Research Laboratory colony in Tifton, GA, were mixed with corncob grits to the desired concentration (Wiseman & Widstrom 1980) and applied to the whorl-stage corn with a pushcart applicator (Sumner et al. 1992). Corn plots were sprayed 3 days after being artificially infested, except for May 1991 when (due to cool weather) treatments were applied 5 days after infestation. Late season tests in August and September of 1993 were naturally-infested. All viruses were applied using a Ford 4000 hi-clearance tractor equipped with a single row spray boom with a single agricultural spray nozzle and tanks pressurized with compressed air. Spray nozzles were changed, pressure adjusted, and tractor velocity set to accommodate the amounts of material applied per ha for each test. Two days after treatment, plants were cut and brought into the laboratory. Up to 30 larvae per plot were collected from these plants and placed individually in 30-ml plastic cups containing bean diet. Larvae were held for observation for 8 days. Larvae that died the first day after collection were considered to have died from injury during collection and were subtracted from the number collected. After the first day, mortality was attributed to either parasitoids, SfNPV, or ascovirus. Parasitoids were observed to emerge from the larvae and spin cocoons. Almost all parasitoids recovered were Cotesia marginiventris (Hymenoptera: Braconidae). SfNPV was indi- TABLE 1. MEAN PERCENT MORTALITY AND STANDARD DEVIATION DUE TO NPV, PARASI- TOIDS, ASCOVIRUS, AND TOTAL MORTALITY OF FALL ARMYWORM LARVAE TREATED ON WHORL-STAGE CORN WITH SFNPV AND FB IN 246 LITERS PER HA OF WATER, 15 MAY NPV Parasitoids Ascovirus Total NPV in LE per ha % FB Mean SD Mean SD Mean SD Mean SD

32 428 Florida Entomologist 77(4) December, 1994 Fig. 1. Mean percent mortality of fall armyworm larvae on whorl-stage corn caused by NPV, parasitoids (PAR), and ascovirus (AV); treated 15 May 1989 with 4 levels of NPV in 3 levels of fluorescent brightener in 246 liters per ha of water. cated when the larvae melted and/or contained POB typical of NPV. Ascovirus was indicated when the larvae remained small and contained the vesicles typical of ascovirus. Percent mortality was determined by dividing the number that died from each cause by the number of larvae collected (minus the number that died the first day after collection). Total percent mortality was computed by adding the mortality factors TABLE 2. MEAN PERCENT MORTALITY AND STANDARD DEVIATION DUE TO NPV, PARASI- TOIDS, ASCOVIRUS, AND TOTAL MORTALITY OF FALL ARMYWORM LARVAE TREATED ON WHORL-STAGE CORN WITH NPV AND FB IN 486 LITERS PER HA OF WATER, 12 JUNE NPV Parasitoids Ascovirus Total NPV in LE per ha % FB Mean SD Mean SD Mean SD Mean SD

33 Hamm et al.: Armyworm Symposium Fig. 2. Mean percent mortality of fall armyworm larvae on whorl-stage corn caused by NPV, parasitoids (PAR), and ascovirus (AV); treated 12 June 1989 with 50 LE per ha of NPV in 0, 0.1, 1, and 5% fluorescent brightener in 486 liters per ha of water plus an untreated control. together. Means and standard deviations were calculated for each morality variable. A GLM procedure (SAS 1985) was conducted for each data set using both the percent mortality and the arcsin of the square root of the percent. Percentages are shown in the tables, but the interactions are based on the analyses of the transformed percentages. Regression analyses were conducted to determine linear and quadratic effects. Rates of Virus, Fluorescent Brightener, and Water Applied The first test, in May 1989, consisted of 12 treatments: 0, 5, 50 and 500 LE per ha of SfNPV, each applied in 0, 0.1%, and 1% FB in 246 liters per ha of water. In the second test, June 1989, the amount of water was increased to 486 liters per ha and a single level of SfNPV, 50 LE per ha, was applied in 4 levels of FB, 0, 0.1, 1, and 5 percent. In 1991, two tests were conducted in which there was an untreated control; in all other treatments a constant amount of SfNPV in varying amounts of water and FB. SfNPV at 618 LE per ha was applied in 234 liters per ha water containing 0, 0.25, 0.5, 1, and 2% FB, or in 468 liters per ha water containing 0, 0.25, 0.5, and 1% FB, or in 936 liters per ha water containing 0, 0.25, or 0.5% FB. The amount of FB applied was 585, 1,169, 2,338, and 4,677 g per ha. The whorl-stage corn was infested 17 May, but due to cool, rainy weather it was not treated until 22 May, 5 days after infestation rather than 3 days as in other tests. When the test was repeated, the corn was infested 31 May and treated 3 June.

34 430 Florida Entomologist 77(4) December, 1994 In 1993, 2 tests were conducted using naturally-infested whorl-stage corn. Treatments were applied 27 August and 27 September with 936 liters per ha water containing 0 virus and 0 FB, 124 LE in 0 FB, 124 LE in 0.25% FB, 1,235 LE in 0 FB, or 1,235 LE in 0.25% FB. Thus all treatments receiving FB received 2,340 g per ha of the adjuvant. RESULTS AND DISCUSSION The 1989 treatment means and standard deviations for the first test are shown in Table 1 and the mean percent mortality due to NPV, parasitoids, and ascovirus are presented graphically in Fig. 1. GLM analysis showed an interaction between linear effects of NPV concentration and linear effects of FB concentration on percent mortality due to NPV. Percent mortality due to parasitoids, ascovirus, and percent total mortality showed only linear effects of NPV concentration. The greatest percent mortality due to NPV, 72.4%, occurred in the 500 LE per ha treatment with 1% FB. The greatest mortality due to NPV in a 50 LE per ha treatment was also in 1% FB. There was a general increase in percent mortality attributable to NPV with increasing NPV concentration and with increasing FB concentration. However, there was a general decrease in the percent mortality caused by parasitoids and ascovirus as the percent mortality due to NPV increased. Treatment means and standard deviations for test 2 in 1989 are shown in Table 2. The mean percent mortality due to NPV, parasitoids, and ascovirus is presented TABLE 3. MEAN PERCENT MORTALITY AND STANDARD DEVIATION DUE TO NPV, PARASI- TOIDS, AND TOTAL MORTALITY OF FALL ARMYWORM LARVAE TREATED ON WHORL-STAGE CORN WITH 618 LE PER HA OF NPV IN VARIOUS VOLUMES OF WATER AND CONCENTRATIONS OF FB, 22 MAY Treatments Water FB NPV Parasitoids Total Liters per ha % g per ha Mean SD Mean SD Mean SD Untreated Control

35 Hamm et al.: Armyworm Symposium TABLE 4. MEAN PERCENT MORTALITY AND STANDARD DEVIATION DUE TO NPV, PARASI- TOIDS, AND TOTAL MORTALITY OF FALL ARMYWORM LARVAE TREATED ON WHORL STAGE CORN WITH 618 LE PER HA OF NPV IN VARIOUS VOLUMES OF WATER AND CONCENTRATIONS OF FB, 3 JUNE Treatments Water FB NPV Parasitoids Total Liters per ha % g per ha Mean SD Mean SD Mean SD Untreated Control graphically in Fig. 2. Because all treatments (except the untreated control) contained the same concentration of virus, the untreated control was omitted from the GLM analysis so that virus concentration would not be a factor. Thus, the effect of FB concentration at the given level of virus could be clearly demonstrated. GLM analysis showed a quadratic effect of FB on percent mortality due to NPV and percent total mortality, but no significant effect on percent mortality due to ascovirus. There was no mortality due to parasitoids in any of the virus treatments and much less mortality due to parasitoids in the control for this test than in the first test. An increase in the FB from 1% (4,860 g per ha) to 5% (24,300 g per ha) resulted in a decrease in percent mortality due to NPV. The 1991 tests were designed to test both the effects of water volume and concentration, or amount, of FB on mortality due to NPV. Thus, the virus level remained constant except for the untreated control which was omitted from the analysis so that virus concentration would not be a factor. The treatment means and standard deviations are shown in Tables 3 and 4. No mortality due to ascovirus was detected in the 1991 tests. Mortality due to parasitoids was much higher (53.4% in the untreated control) in the 22 May test, which was treated 5 days after infestation, than in the 3 June test (19.7% in the untreated control) which was treated 3 days after infestation. In the 22 May test, with the higher rate of parasitoids, the interactions between water and FB were similar when FB was expressed as either percent or as g per ha (Fig. 3). There were interactions between quadratic effects of water and quadratic effects of FB for both percent mortality due to NPV and percent mortality due to parasitoids; again, as the mean percent mortality due to NPV increased the

36 432 Florida Entomologist 77(4) December, 1994 Figure 3. Mean percent mortality of fall armyworm larvae on whorl-stage corn caused by NPV following treatment on 22 May 1991 with 618 LE per ha of NPV in 5 levels of fluorescent brightener in 3 volumes of water. mean percent mortality due to parasitoids decreased. However, percent total mortality showed only a linear effect of FB (Fig. 4). In the 3 June test, lower rates of parasitism interfered less with the effects of water and FB on mortality due to NPV; thus, the interactions between water and FB were different when FB was expressed as percent than when it was expressed as g per ha. When FB was expressed as percent (Fig, 5), there was an interaction between the linear effects of water and the quadratic effects of FB on both percent mortality due to NPV and total mortality. However, there was no significant effect of water or FB on percent mortality caused by parasitoids. When FB was expressed in g per ha, there was a quadratic effect of FB on both percent mortality attributable to NPV (Fig. 6) and percent mortality due to parasitoids. Again, a decrease in mortality due to parasitoids was associated with the increase in mortality due to NPV. There was an inter- Fig. 4. Mean percent total mortality (NPV and parasitoids) of fall armyworm larvae on whorl-stage corn, treated 22 May 1991 with 618 LE per ha of NPV in 5 levels of fluorescent brightener in 3 volumes of water.

37 Hamm et al.: Armyworm Symposium Fig. 5. Mean percent mortality of fall armyworm larvae on whorl-stage corn caused by NPV following treatment on 3 June 1991with 618 LE per ha of NPV in 5 levels of fluorescent brightener expressed as percent of 3 volumes of water. action between the linear effects of water and the quadratic effects of FB on total mortality (Fig. 7). In the 1993 tests, a high volume of water, 936 liters per ha, was used to compare effects of 0.25% FB vs 0 FB at 3 levels of virus. The treatment means and standard deviations are shown in Tables 5 and 6. The percent mortality due to NPV, parasitoids, and ascovirus are shown graphically in Figs. 8 and 9. The 27 August test showed an interaction between the quadratic effects of virus and the linear effects of FB on percent mortality due to NPV, but only the quadratic effect of virus was noted for total mortality. There were no significant effects on percent mortality due to parasitoids and no mortality due to ascovirus. The 27 September tests showed quadratic effects of virus on percent mortality due to NPV, ascovirus, and total mortality. There was a Fig. 6. Mean percent mortality of fall armyworm larvae on whorl-stage corn caused by NPV following treatment on 3 June 1991 with 618 LE per ha of NPV in 5 levels of fluorescent brightener in 3 volumes of water.

38 434 Florida Entomologist 77(4) December, 1994 Fig. 7. Mean percent total mortality (NPV and parasitoids) of fall armyworm larvae on whorl-stage corn treated 3 June 1991 with 618 LE per ha of NPV in 5 levels of fluorescent brightener in 3 volumes of water. significant interaction between the linear effects of virus and linear effects of FB on percent mortality due to parasitoids. SUMMARY Many factors are important to the successful control of fall armyworm with the SfNPV. First the virus must be placed where the larvae are feeding and in a concentration sufficient for the larvae to ingest a lethal dose. In these tests, mortality due to SfNPV increased with increasing virus concentration up to the highest rate tested, 1235 LE per ha. It is apparent from these tests that increasing the volume of water, up to 936 liters per ha, helped to deliver the virus deep into the whorl of the corn plant TABLE 5. MEAN PERCENT MORTALITY AND STANDARD DEVIATION DUE TO NPV, PARASI- TOIDS, AND TOTAL MORTALITY OF FALL ARMYWORM LARVAE TREATED ON WHORL-STAGE CORN WITH NPV AND FB IN 936 LITERS PER HA OF WATER, 27 AUGUST Treatments NPV Parasitoids Total NPV in LE per ha % FB Mean SD Mean SD Mean SD

39 Hamm et al.: Armyworm Symposium TABLE 6. MEAN PERCENT MORTALITY AND STANDARD DEVIATION DUE TO NPV, PARASI- TOIDS, ASCOVIRUS, AND TOTAL MORTALITY OF FALL ARMYWORM LARVAE TREATED ON WHORL-STAGE CORN WITH NPV AND FB IN 936 LITERS PER HA OF WATER, 27 SEPTEMBER NPV Parasitoids Ascovirus Total NPV in LE per ha % FB Mean SD Mean SD Mean SD Mean SD where the larvae were feeding. The FB interacted significantly with virus concentration and water volume to increase mortality caused by NPV. However, there was no increase in mortality due to NPV as the percent FB increased beyond 1%. In higher volumes of water (468 and 936 liters per ha) 0.5 and 0.25% FB resulted in the highest percent mortality due to NPV. Of the FB rates, expressed as weight, 2,340 g per ha in Fig. 8. Mean percent mortality of fall armyworm larvae on whorl-stage corn due to NPV and parasitoids (PAR), following treatment on 27 August 1993 with 3 levels of NPV in 2 levels of fluorescent brightener in 936 liters per ha of water.

40 436 Florida Entomologist 77(4) December, 1994 Fig. 9. Mean percent mortality of fall armyworm larvae on whorl-stage corn due to NPV, parasitoids (PAR), and ascovirus (AV), treated 27 September 1993 with 3 levels of NPV in 2 levels of fluorescent brightener in 936 liters per ha of water. 936 liters per ha resulted in the highest percent mortality due to NPV. In general, as the percent mortality due to NPV increased the percent mortality due to parasitoids and ascovirus decreased. This did not appear to be a direct effect of FB on the parasitoids. The increased infectivity of the NPV and earlier mortality of larvae due to NPV associated with the FB resulted in death of host larvae attributable to NPV before the parasitoids could complete development. Thus, the total mortality was not affected as greatly as the percent mortality due to NPV by changes in water volume or FB concentration. ACKNOWLEDGMENTS The authors wish to thank JoAnne Denham and Lenny Atkins for their technical assistance in conducting these studies. Richard Layton is thanked for his help in conducting the statistical analyses of the data. Proprietary names are necessary to report factually on available data; however, the USDA neither guarantees nor warrants the product, and the use of the name by USDA implies no approval of the product to the exclusion of others that may be suitable. REFERENCES CITED ASHLEY, T. R Geographical distributions and parasitization levels for parasitoids of the fall armyworm, Spodoptera frugiperda. Florida Entomol. 69: FUXA, J. R Prevalence of viral infections in populations of fall armyworm, Spodoptera frugiperda, in southeastern Louisiana. Environ. Entomol. 11:

41 Hamm et al.: Armyworm Symposium GARDNER, W. A., AND J. R. FUXA Pathogens for the suppression of the fall armyworm. Florida Entomol. 63: HAMM, J. J., AND W. W. HARE Application of entomopathogens in irrigation water for control of fall armyworms and corn earworms (Lepidoptera: Noctuidae) on corn. J. Econ. Entomol. 75: HAMM, J. J., S. D. PAIR, AND O. G. MARTI, JR Incidence and host range of a new ascovirus form fall armyworm, Spodoptera frugiperda (Lepidoptera: Noctuidae). Florida Entomol. 69: HAMM, J. J., AND M. SHAPIRO Infectivity of fall armyworm (Lepidoptera: Noctuidae) nuclear polyhedrosis virus enhanced by a fluorescent brightener. J. Econ. Entomol. 85: HAMM, J. J., AND J. R. YOUNG Value of virus presilk treatment for corn earworm and fall armyworm control in sweet corn. J. Econ. Entomol. 64: SAS INSTITUTE SAS/STAT User s Guide. SAS Institute, Cary, NC. SHAPIRO, M Use of optical brighteners as radiation protectants for gypsy moth (Lepidoptera: Lymantriidae) nuclear polyhedrosis virus. J. Econ. Entomol. 85: SHAPIRO, M., E. M. DOUGHERTY, AND J. J. HAMM Compositions and methods for biocontrol using fluorescent brighteners. U.S. Patent no. 5,124,149. SHAPIRO, M., AND J. L. ROBERTSON Enhancement of gypsy moth (Lepidoptera: Lymantriidae) baculovirus activity by optical brighteners. J. Econ. Entomol. 85: SUMNER, H. R., H. R. GROSS, AND B. R. WISEMAN Pushcart mounted rotary applicator for infesting plants with the larvae of Spodoptera frugiperda (Lepidoptera: Noctuidae). J. Econ. Entomol. 85: WISEMAN, B. R., AND N. W. WIDSTROM Comparison of methods of infesting whorl-stage corn with fall armyworm. J. Econ. Entomol. 73: YOUNG, J. R Suppression of fall armyworm populations by incorporation of insecticides into irrigation water. Florida Entomol. 63:

42 437 A GENETICALLY-MODIFIED BACILLUS THURINGIENSIS PRODUCT EFFECTIVE FOR CONTROL OF THE FALL ARMYWORM (LEPIDOPTERA: NOCTUIDAE) ON CORN J. N. ALL 1, J. D. STANCIL 1, T. B. JOHNSON 2, AND R. GOUGER 2 1 Department of Entomology University of Georgia, Athens, GA Ecogen Inc., Langhorne, PA ABSTRACT ECX9399, a variant of strain EG2348 (the active ingredient of the bioinsecticide Condor ) of Bacillus thuringiensis (Berliner) (Bt) subspecies kurstaki was developed by recombinant DNA technology by Ecogen Inc. This strain showed greater control of fall armyworm, Spodoptera frugiperda (J. E. Smith), infestations in whorl stage corn, Zea mays L., than other Bt products in field tests conducted in Georgia, Mississippi and Florida during Control by EC9399 was greatest in Mississippi, where a 4- day spray interval (total of 3 sprays) was used and least in Georgia, where a 7-day schedule (total of 3 sprays) was maintained. The new genetically-modified Bt product This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

43 438 Florida Entomologist 77(4) December, 1994 had comparable efficacy (decrease in larval number on plants and reduced defoliation) to methomyl, which was used at commercial rates as a conventional insecticide standard at the 3 locations. Key Words: Corn, Bacillus thuringiensis, Spodoptera frugiperda RESUMEN El ECX9399, una variante de la cepa EG2348 (ingrediente activo del bioinsecticida Condor*) de Bacillus thuringiensis (Berliner) (Bt) subespecie kurstaki, fue desarrollado por Ecogen Inc. mediante la tecnología de la recombinación del DNA. Esta cepa demostró mayor control de las infestaciones del gusano trozador, Spodoptera frugiperda (J. E. Smith), en el maíz, Zea mays L., en estado vegetativo que otros productos de Bt en pruebas de campo llevadas a cabo en Georgia, Mississippi y Florida durante El control con EC9399 fue mayor en Mississippi, donde fue utilizado un intervalo de aspersiones de cuatro días (3 aspersiones en total) y menor en Georgia, donde fue mantenido un programa de siete días (tres aspersiones en total). El nuevo producto de Bt genéticamente modificado tuvo una eficacia (disminución del número de larvas por planta y de la defoliación) comparable a la del methomyl, que fue empleado en concentraciones comerciales como un insecticida convencional estándar en los tres lugares. Bacillus thuringiensis (Berliner) (Bt) subspecies kurstaki products have generally exhibited moderate to low effectiveness for controlling the fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith) (Gardner & Fuxa 1980; Krieg & Langenbruch 1981; Teague 1993). Recently, products derived from new Bt strains have been commercialized, including the EG2348 strain of Bt subspecies kurstaki, the active ingredient in the bioinsecticide Condor. These products have shown improved toxicity for certain insects (Gawron-Burke & Baum 1991). EG2348 was developed by Ecogen Inc., (Langhorne, PA 19047) utilizing natural processes for transfer of plasmids with genes encoding for production of specified insecticidal crystal proteins (Gawron-Burke & Baum 1991). Recombinant DNA (rdna) technology has made it possible to improve Bt strains (Carlton & Gawron-Burke 1993), and recently Ecogen produced a rdna modified variant (ECX9399) of EG2348 that was more toxic to FAW in laboratory tests (T. Johnson, Ecogen, Inc., Langhorne PA, unpublished data). This study reports results of 1993 field trials in 3 locations with an oil flowable formulation of ECX9399 for FAW control in corn, Zea mays L. MATERIALS AND METHODS Field experiments were conducted near Athens, GA, and Oktibbeha, MS, in August and near Groveland, FL, in November. A field corn cultivar (DeKalb 689) was used in GA, Pioneer Tropical corn was planted in MS, and Silverqueen sweetcorn was employed in FL. Plots at the 3 locations varied from 1 to 4 rows 6 to 12 m long. A randomized complete block experimental design was used with 4 or 5 replications. Spray applications were made with CO 2 sprayers mounted with full-cone spray tips and calibrated at the rate of 80 to 120 liters per ha. Three applications were made in the GA (7-day intervals) and MS (4-day intervals) tests and 4 applications (4- to 6- day intervals) were made at the FL site. In each experiment, sprays were applied during the mid-whorl stage of plant development when moderate FAW infestations were

44 All et al.: Armyworm Symposium present in the fields (50% or more plants infested with small larvae). Test materials included Cutlass WP (Bt subspecies kurstaki, strain EG2371), Condor OF (Bt subspecies kurstaki, strain 2348, in an oil flowable formulation), ECX9399 OF (oil flowable formulation), and methomyl (Lannate LV) (see Table 1 for rates). Efficacy was determined at selected intervals during and following the spray applications by examining the plants in each plot for defoliation. In MS a visual estimate of defoliation was made, whereas, in GA and FL a 0 to 8 (GA) or 0 to 10 (FL) rating of defoliation and whorl injury was made, progressing from 0 to destruction of plants. Additionally, between 5 and 10 plants in each plot were examined for larvae at selected intervals during and after the spray period. They were categorized as small (<8 mm), intermediate (>8-12 mm) and large (>12 mm). To compare the data between locations, the defoliation estimates and larval counts were converted to percent control by determining the ratio of plant injury or larval counts in the treatment versus the untreated checks. Analysis of variance and Duncan s new multiple range test were conducted using a computer based statistical analysis system (SAS User s Guide: Statistics 1985). RESULTS The FAW infestations at the 3 locations were moderate to heavy. The data in Table 1 demonstrate that the formulation of the genetically-modified Bt strain ECX9399 produced control comparable to the conventional standard methomyl. Control with Bt was best in MS, where a 4-day spray interval was used, and least in GA, which had a 7-day schedule. Larval numbers also were significantly reduced on corn treated with ECX9399, but were statistically different from methomyl and Condor in GA. In the MS trial, larval populations were similar in ECX9399 and Condor plots, and both were significantly less than in the methomyl treatment. In FL, larval populations were significantly less in ECX9399 than in Cutlass, but not the methomyl treatment. TABLE 1. EFFICACY OF SELECTED BT INSECTICIDAL PRODUCTS FOR FAW ON MID-WHORL STAGE CORN IN 3 LOCATIONS DURING % Control 1 GA 3 MS 3 FL 3 Insecticide Rate 2 Damage Larvae Damage Larvae Damage Larvae ECX9399 OF a 32a 94.8a 79.3a 79.1a 82.9a Condor OF bc 0a 92.0a 77.8a Cutlass WP b 27.1b Methomyl ab 0a 89.5a 27.4b 88.4a 98.1a 1 Means followed by the same letter within a column are not significantly different in Duncan s new multiple range analysis (P < 0.05). 2 Rates of ECX9399 OF and Condor OF are presented as volume (in liters) of formulated product per ha, Cutlass WP as weight (in kg) of formulated product per ha and methomyl (Lannate LV) as weight (in kg) of active ingredient per ha. Condor was used at a rate of 1.6 liter per ha in GA and 1.07 liter per ha in MS; methomyl was used at 1.13 kg per ha rate in FL and 0.5 kg per ha at GA and MS. 3 Applications in GA were made on a 7-day schedule for 3 sprays, MS was every 4 days for 3 sprays and FL was every 4 or 6 days for 4 sprays.

45 440 Florida Entomologist 77(4) December, 1994 The FL results were similar to those reported by Teague (1993) for Cutlass and methomyl for FAW control on sweetcorn. The results show that recombinant DNA technology can be used to improve the toxicity and specificity of Bt to insects such as the FAW. ECX9399 was superior to its parent strain, EG2348 (Condor ), in controlling FAW populations and damage in three locations. The fact that ECX9399 produced similar control as methomyl (one of the most efficacious materials available for FAW (All et al. 1986)) in the experiments accentuates the potential of genetically-improved Bt strains for insect management. REFERENCES CITED ALL, J. N., A. JAVID, AND P. GUILLEBEAU Control of fall armyworm with insecticides in north Georgia sweetcorn. Florida Entomol. 69: CARLTON, B. C., AND C. GAWRON-BURKE Genetic improvement of Bacillus thuringiensis for bioinsecticide development, pp in L. Kim [ed.], Advanced engineered pesticides. Marcel Dekker, Inc., New York. GARDNER, W. A., AND J. R. FUXA Pathogens for the suppression of the fall armyworm. Florida Entomol. 63: GAWRON-BURKE, C., AND J. A. BAUM Genetic manipulation of Bacillus thuringiensis insecticidal crystal protein genes in bacteria, pp in J. K. Setlow [ed.], Genetic engineering. Plenum Press, New York. KRIEG, A., AND G. A. LANGENBRUCH Susceptibility of arthropod species to Bacillus thuringiensis, pp in H. D. Burges [ed.], Microbial control of pests and plant diseases Academic, New York. SAS USER S GUIDE: STATISTICS SAS Inst., Cary, NC. 957 pp. TEAGUE, T. G Control of fall armyworm in sweet corn with Bacillus thuringiensis, Insecticide & Acaricide Tests 18:

46 440 Florida Entomologist 77(4) December, 1994 MANAGEMENT OF THE BEET ARMYWORM (LEPIDOPTERA: NOCTUIDAE) IN COTTON: ROLE OF NATURAL ENEMIES JOHN R. RUBERSON 1,3, GARY A. HERZOG 2, WILLIAM R. LAMBERT 2, AND W. JOE LEWIS 1 1 Insect Biology and Population Management Research Laboratory United States Department of Agriculture, Agricultural Research Service Tifton, GA Dept. of Entomology, University of Georgia, Tifton, GA Current address: Dept. of Entomology, University of Georgia, P.O. Box 748, Tifton, GA ABSTRACT The beet armyworm, Spodoptera exigua (Hubner), has recently become a persistent and explosive pest of cotton in the southeastern United States. It is, however, attacked by a large and diverse complex of beneficial arthropods and pathogens that appear capable of maintaining beet armyworm populations below economically-damaging levels. Disruption of this complex contributes to outbreaks of S. exigua. It can This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

47 Ruberson et al.: Armyworm Symposium also exacerbate problems with other pests because the complex of beneficial organisms attacking the beet armyworm is comprised of generalist species that also suppress other pests in the cotton production system. Management of the beet armyworm through conservation of its natural enemies, therefore, provides multiple benefits to growers by managing other pests as well. Key Words: Beet armyworm, cotton, biological control, Spodoptera exigua, parasitoid, predator RESUMEN El gusano trozador de la remolacha, Spodoptera exigua, recientemente se ha convertido en una plaga persistente y explosiva del algodón en el sureste de los Estados Unidos; sin embargo, es atacado por un complejo grande y diverso de artrópodos útiles y patógenos que parece ser capaz de mantener las poblaciones del gusano de la remolacha por debajo de los niveles de daño económico. La alteración de este complejo favorece la aparición de brotes del gusano trozador, pero también puede aumentar los problemas con otras plagas porque el complejo de los organismos útiles que atacan el gusano de la remolacha está compuesto de especies generalistas que también pueden suprimir otras plagas en el sistema de producción del algodón. Por lo tanto, el manejo del gusano de la remolacha mediante la conservación de sus enemigos naturales también ofrece beneficios múltiples a los granjeros en el manejo de otras plagas. The beet armyworm, Spodoptera exigua (Hübner), is an introduced pest of numerous crops in the United States. It appears to be a native of southern Asia, although its origin is presently unclear. It was first reported in the United States with the collection of specimens in Oregon and California in 1876 (Harvey 1876). The insect dispersed across the country and was established in Florida by the late 1920s, where it was recorded feeding only on asparagus fern, gladiolus, and grasses (Wilson 1932). In the years since its introduction, the beet armyworm has become progressively more pestiferous in the United States on an increasingly wide range of crop plants (see Pearson 1982). Its current recorded host range in North America exceeds 90 plant species, including numerous important crop species such as corn, cotton, soybeans, peanuts, cabbage, tomatoes, and peppers (Pearson 1982). The bases for this apparent host range expansion are presently unclear; the changes suggest that this insect has considerable phenotypic plasticity in its host range [and likely genotypic, as is the case with the fall armyworm, Spodoptera frugiperda (Pashley, pers. comm.)] and thus it may become an increasingly widespread pest in the future. In addition to its broad host range, there are several facets of S. exigua s biology that may predispose it to being an explosive pest. First, S. exigua has a relatively brief developmental time under field conditions (Ali & Gaylor 1991), permitting rapid cycling of generations. Second, it has a high reproductive capacity, with average calculated fecundities ranging from to eggs per female (Wilson 1934, Hogg & Gutierrez 1980, Chu & Wu 1992). A simple calculation illustrates this point. Assuming a population sex ratio of 1 female to 1 male, a realized field fecundity of 200 eggs (approx. 2 egg masses) per female, and restricted emigration and immigration, 99% mortality within a generation would be necessary to simply maintain the population at a constant size. Thus, suppression of this pest requires high levels of mortality to counterbalance its high fecundity. Third, these insects are highly mobile and are thus capable of colonizing wide-ranging areas (French 1969, Mitchell 1979). Finally, insec-

48 442 Florida Entomologist 77(4) December, 1994 ticides typically provide less than adequate control (e.g., Cobb & Bass 1975, Meinke & Ware 1978, Brewer & Trumble 1989, Wolfenbarger & Brewer 1993). This is due, at least in part, to the insect s innate tolerance of many insecticidal materials at recommended field rates. But the beet armyworm s ovipositional and feeding biology also influences insecticide efficacy. Females oviposit eggs in masses of 46 to 230 eggs (x ± SD = 99.4 ± 40.6; n = 75 field-collected egg masses; J.R.R. unpubl.), typically on the undersurface of leaves in the lower plant canopy. Insecticide coverage is often inadequate in these areas, particularly after the canopy has expanded. Further, beet armyworm larvae feed in groups through the first and second instars, then disperse as third instars (Poe et al. 1973). This feeding behavior concentrates a large proportion of the population into a relatively small area during the period when the larvae are most susceptible to insecticides. Thus, to kill a sufficient number of larvae to attain control, the material must contact a relatively small proportion of the plant canopy in the plant region most difficult to cover a very difficult proposition when the plants are large and the canopy is closed. Despite its pestiferous potential, the beet armyworm has been historically a sporadic and minor pest of cotton in the southeastern United States (Smith 1989). In recent years, however, it has become a persistent and serious cotton pest in the southeastern and mid-southern United States, especially in regions conducting the Boll Weevil Eradication Program (e.g., Fig. 1). However, the current ubiquity and consistency of the outbreaks, both inside and outside of active eradication zones, suggest that this pest has become a more widespread and serious cotton pest for reasons independent of the Boll Weevil Eradication Program. However, this program likely provides a ready opportunity for the beet armyworm to escape natural controls. Fig. 1. Number of specific beet armyworm insecticide treatments applied per acre of cotton production in the state of Georgia from 1980 to BWEP, demarcated by the vertical dashed lines, indicates the period when the Boll Weevil Eradication Program was in its active phase in the state.

49 Ruberson et al.: Armyworm Symposium Regardless of the cause, it is critical at this juncture to devise efficacious, biorational pest management approaches. Natural enemies appear to be a key element in the management of the beet armyworm. In 1973, Eveleens et al. demonstrated that beet armyworm outbreaks could be induced by applications of organophosphate insecticides in cotton. Cotton can support a large and diverse complex of beneficial arthropods (Whitcomb & Bell 1964, van den Bosch & Hagen 1966) and in production systems receiving multiple treatments of highly toxic materials, such as organophosphates and pyrethroids, these complexes can be seriously disrupted for the remainder of the growing season. Subsequently, in the absence of the beneficial arthropods, production of an acceptable crop will require continued, repeated use of insecticides. The Boll Weevil Eradication Program relies on widespread, repetitive applications of organophosphates to suppress and eventually eliminate boll weevil populations (USDA-APHIS 1991).These treatments have a profound detrimental impact that releases beet armyworm populations from their natural biological control agents (e.g., Wilkinson et al. 1979). NATURAL ENEMIES OF THE BEET ARMYWORM The large number of predators and parasitoids that have been found associated with beet armyworm eggs and larvae are listed in Tables 1 and 2. This complex of natural enemies differs among various geographic regions; however, there are common linkages. Several parasitoid species, for example, have been found across the cotton belt, including the braconids Cotesia marginiventris, Meteorus autographae, Chelonus insularis, the ichneumonid Temelucha sp., and the tachinid Lespesia archippivora (Table 2). Their relative abundance and efficacy, however, vary among regions. Similarly, several genera of predators are shared across the cotton belt. It is noteworthy that the most commonly encountered natural enemies of the beet armyworm in all regions are generalists that attack a variety of hosts in multiple habitats. Given that the beet armyworm is an introduced pest, such a pattern is to be expected in the absence of specific imported biological control agents. In addition to predators and parasitoids, several pathogens have also been recovered from the beet armyworm. A nuclear polyhedrosis virus has been widely reported (Oatman & Platner 1972, Eveleens et al.1973, Pearson 1982, Kolodny-Hirsch et al. 1993). Fungal pathogens, however, can also be important. Wilson (1933) reported that a fungus, described at the time as Spicaria prasina (probably Nomuraea rileyi), decimated populations of beet armyworm larvae during wet weather. In our studies in Georgia, we have observed larvae infected with the fungi Erynia sp. nr. pieris (identified by Dr. Donald Steinkraus, Univ. of Arkansas) and N. rileyi. Of these two species, Erynia was the most commonly encountered. Despite the large number of natural enemies cataloged to date, there are few data to demonstrate their impact on beet armyworm populations. Eveleens et al. (1973) demonstrated in California that beet armyworm outbreaks could be induced by applications of organophosphate insecticides, which presumably disrupt the natural enemy complex. They suggested that predators were the most important mortality agents for the beet armyworm populations in their study, and that the greatest loss occurred in the egg and early larval stages. Hogg & Gutierrez (1980) also observed high rates of loss for eggs and small larvae of the beet armyworm in cotton in California and also attributed much of this loss to predators. De Clercq & Degheele (1994) recently demonstrated in the laboratory that the native predaceous pentatomid Podisus maculiventris can consume large numbers of all stages of beet armyworm. It is, however, particularly destructive to eggs (ranging from 53.5 eggs consumed during the second instar to eggs consumed per day by

50 444 Florida Entomologist 77(4) December, 1994 TABLE 1. PREDATORS OBSERVED IN ASSOCIATION WITH BEET ARMYWORM EGGS OR YOUNG LARVAE IN THE UNITED STATES. Taxon/Species State Association 1 Location References Dermaptera Labidura riparia E, L Georgia Ruberson et al Heteroptera Orius insidiosus E, L Georgia Ruberson et al Orius tristicolor E, L California Eveleens et al. 1973; Hogg and Gutierrez 1980 Geocoris pallens E, L California Eveleens et al. 1973; Hogg and Gutierrez 1980 Geocoris punctipes E, L Georgia Ruberson et al Geocoris uliginosus E, L Georgia Ruberson et al Podisus maculiventris L Florida Wilson 1933; Georgia Ruberson et al Nabis roseipennis L Georgia Ruberson et al Nabis americoferus E, L California Eveleens et al Zelus sp. E, L Georgia Ruberson et al California Eveleens et al Sinea sp. E, L California Eveleens et al Neuroptera Chrysoperla carnea E, L California Eveleens et al. 1973; Hogg and Gutierrez 1980 Chrusoperla refilabris E, L Georgia Ruberson et al Hemerobius sp. E, L Georgia Ruberson et al Coleoptera Collops E, L California Eveleens et al Notoxus calcaratus E, L California Eveleens et al Coccinella septempunctata E Georgia Ruberson et al Hymenoptera Polistes fuscatus L Florida Wilson 1933 Solenopsis invicta E, L Georgia Ruberson et al Arachnida Unidentified (3 species) L Georgia Ruberson et al Unidentified California Eveleens et al Stage association indicates with which stages of beet armyworm the predators were found; E = eggs and L = larvae. adult female predators) and small larvae. Also, most life stages of beet armyworm are reportedly suitable prey for predator development. These data provide a glimpse into the possible impact of predators on beet armyworms, although P. maculiventris appears to be only a small, and inconsistent, part of the total natural enemy complex in the field. The overall impact of natural enemies in the field, however, is poorly delin-

51 Ruberson et al.: Armyworm Symposium TABLE 2. PARASITOIDS OF BEET ARMYWORM EGG AND LARVAE RECORDED IN THE UNITED STATES. Taxon/Species Stages Attacked 1 Location References Diptera: Tachinidae Lespesia archippivora L1-L5 California van den Bosch & Hagen 1966; Henneberry et al. 1991; Eveleen et al Texas Harding 1976 Oklahoma Soteres et al Georgia Ruberson et al Eucelatoria armigera California van den Bosch & Hagen 1966; Henneberry et al Eucelatoria rubentis Florida Wilson 1933; Tingle et al Eucelatoria sp. nr. armigera California Henneberry et al Winthemia rufopicta Florida Tingle et al Archytas californiae California Eveleens et al Archytas apicifer California Henneberry et al Archytas marmoratus Georgia Ruberson et al Voria ruralis California Eveleens et al Chaetogodia monticola Hawaii Swezey 1935 Gonia crassicornis Florida Wilson 1933 Hymenoptera: Braconidae Cotesia marginiventris L1-L4 California van den Bosch & Hagen 1966; Pearson 1982; Henneberry et al Oklahoma Soteres et al Florida Wilson 1933; Tingle et al Georgia Ruberson et al Cotesia laeviceps U.S. Krombein et al Cotesia militaris No. America Krombein et al Meteorus autographae L1-L4 Florida Wilson 1933; Tingle et al Georgia Ruberson et al Texas Harding 1976 Meteorus leviventris Texas van den Bosch & Hagen 1966; Harding 1976 Meteorus rubens California Henneberry et al Meteorus laphygmae Krombein et al Stages attacked signifies larval instars (L1-L5) and eggs (E) susceptible to parasitization by the respective parasitoids. 2 Oviposits in eggs and emerges from the late larval stages.

52 446 Florida Entomologist 77(4) December, 1994 TABLE 2.(CONTINUED) PARASITOIDS OF BEET ARMYWORM EGG AND LARVAE RECORDED IN THE UNITED STATES. Taxon/Species Stages Attacked 1 Location References Chelonus insularis E-L5 2 California van den Bosch & Hagen 1966; Eveleens et al. 1973; Pearson 1982; Henneberry et al Texas Harding 1976 Oklahoma Soteres et al Florida Wilson 1933; Tingle et al Georgia Ruberson et al Aleiodes laphygmae L1-L3 Georgia Ruberson et al Cremnops haemotodes California Henneberry et al Zele melea Oklahoma Soteres et al Hymenoptera: Ichneumonidae Hyposoter exiguae L1-L3 California van den Bosch & Hagen 1966; Eveleens et al. 1973; Pearson 1982; Henneberry et al Hyposoter annulipes U.S. Krombein et al Pristomerus spinator L1-L3 California Eveleens et al. 1973; Pearson 1982; Henneberry et al Oklahoma Soteres et al Campoletis argentifrons U.S. van den Bosch & Hagen 1966 Campoletis flavicincta L1-L3 Georgia Ruberson et al Campoletis sonorensis L1-L3 U.S. Krombein et al Oklahoma Soteres et al Temelucha sp. California Pearson 1982; Henneberry et al Florida Tingle et al Nepiera fuscifemora West U.S. Krombein et al Ophion sp. Georgia Ruberson et al Therion longipes California van den Bosch & Hagen 1966; Eveleens et al Rubicundiella perpturbatrix West U.S. van den Bosch & Hagen 1966; Krombein et al Sinophorus caradrinae (?) Colorado Krombein et al Stages attacked signifies larval instars (L1-L5) and eggs (E) susceptible to parasitization by the respective parasitoids. 2 Oviposits in eggs and emerges from the late larval stages.

53 Ruberson et al.: Armyworm Symposium TABLE 2.(CONTINUED) PARASITOIDS OF BEET ARMYWORM EGG AND LARVAE RECORDED IN THE UNITED STATES. Taxon/Species Stages Attacked 1 Location References Hymenoptera: Eulophidae Euplectrus plathypenae L3-L5 Florida Wilson 1933 Hymenoptera: Trichogrammatidae Trichogramma spp. E California van den Bosch & Hagen Stages attacked signifies larval instars (L1-L5) and eggs (E) susceptible to parasitization by the respective parasitoids. 2 Oviposits in eggs and emerges from the late larval stages. eated and/or entirely unknown in the southeastern U.S. where beet armyworm problems have recently been most severe. IMPACT OF NATURAL ENEMIES ON BEET ARMYWORM POPULATIONS IN GEORGIA We have undertaken various field studies in Georgia in an effort to characterize mortality factors and levels for beet armyworm populations.these studies have focused on two areas: 1) characterization and quantification of beet armyworm parasitoids and pathogens, and 2) determination of their impact on survival of eggs, small larvae, and pupae. Larval Mortality: Impact of Parasitoids and Pathogens Beet armyworm larvae of all ages were sampled from commercial cotton fields in Georgia in 1992 and 1993 (see Ruberson et al for details). Collections were made on various dates from 15 July to 16 September in 1992 and from 24 May to 12 October in Totals of 7,545 and 7,072 larvae were collected in 1992 and 1993, respectively. The parasitoids reared from these larvae (in relation to instar collected) are presented in Table 3, with rates of parasitism by each species. The parasitism rates for the two years, pooled across larval instars and collection locales, were 46.8% and 40.2% in 1992 and 1993, respectively. The majority of parasitism, and resultant larval mortality, occurred in the early instars. In both years, C. marginiventris was the predominant species, and it accounted for more of the parasitism in 1993 than it did in 1992, particularly in the second and third instars (Table 3). This contrasts with results from California indicating that the tachinid L. archippivora and the braconid C. insularis were the most important parasitoids in cotton and alfalfa, respectively (Henneberry et al. 1991, and Pearson 1982, respectively). Soteres et al. (1984) also found C. insularis to be the most common parasitoid attacking beet armyworms in alfalfa in Oklahoma. C. marginiventris, however, is the dominant parasitoid of beet armyworm larvae from pigweed in Florida (Tingle at al. 1978). Thus, C. marginiventris appears to be the more dominant species in the eastern half of the United States, whereas C. insularis is more dominant in the west. Cotesia marginiventris is highly attracted to plants damaged by beet armyworm feeding (e.g., Turlings et al. 1991), and this response is intensified by the clumped

54 448 Florida Entomologist 77(4) December, 1994 TABLE 3. PARASITISM RATES (%) IN POPULATIONS OF BEET ARMYWORM LARVAE COL- LECTED IN GEORGIA COTTON IN 1992 AND LARVAE WERE COLLECTED FROM BARTOW, BEN HILL, DECATUR, DOOLY, LAURENS, MILLER, SEMI- NOLE, AND TIFT COUNTIES. % Parasitism of Beet Armyworm Larval Instar 1 Parasitoid Cotesia marginiventris Aleiodes laphygmae Meteorus autographae Chelonus insularis Lespesia archippivora Ichneumonidae Unknown parasites Total % parasitism No. larvae collected Cotesia marginiventris Aleiodes laphygmae Meteorus autographae Cardiochiles nigriceps Pristomerus spinator Lespesia archippivora Archytas marmoratus Unknown parasites Total % parasitism No. larvae collected Instar of larvae at time of collection. 2 Includes Campoletis sonorensis, Pristomerus spinator, and Ophion sp. feeding behavior of the beet armyworm larvae on cotton plants [A. Datema (Wageningen, The Netherlands), J.R.R., and W.J.L., unpubl.]. This parasitoid, therefore, is highly-attuned to locating clusters of beet armyworm larvae. It is, however, susceptible to several organophosphate and pyrethroid insecticides (Wilkinson et al. 1979, Ruberson et al. 1993), which could limit its efficacy in conventional, chemical-intensive cotton production. Several pathogens were also recovered from larvae collected in the field, although disease did not appear to be a substantial mortality factor. The most commonly-encountered pathogen was the fungus Erynia sp. nr. pieris (determined by Dr. Donald Steinkraus, Univ. of Arkansas), which killed 6.2% of the larvae collected in 1992, but only 0.3% of those collected in 1993 (there was exceptionally little rain that year). A few specimens collected in 1992 were infected with N. rileyi, but no N. rileyi was observed in A nuclear polyhedrosis virus was found in 1.8% of the larvae collected

55 Ruberson et al.: Armyworm Symposium in 1992 and in 0.1% of the larvae in In addition, a single larva infected with an ascovirus (determined by Dr. John J. Hamm, USDA-ARS, Tifton GA) was collected in The senior author and J.J. Hamm (USDA-ARS, Tifton, GA) found this virus to be a poor pathogen of beet armyworm larvae in laboratory tests. Overall parasitoid- and pathogen-related mortality from our collections ranged from approximately 40 to 50%. These overall rates were generally higher than those observed in the California studies noted above (Pearson 1982, Henneberry et al. 1991). However, Pearson (1982) did observe comparable parasitism levels for larvae on alfalfa in the late summer and early fall in Imperial Valley. This level of mortality comprises a relatively high level of loss in the population, but is well below the 99% needed to maintain or suppress the pest population. Impact of Predation on Eggs, Larvae, and Pupae. Two studies were undertaken to examine loss of beet armyworms in the field. The first examined the rate of loss for eggs and small larvae to assess loss prior to, and in the initial periods of susceptibility to parasitization. The second study addressed the loss of beet armyworm pupae in the soil. Egg/Larvae Predation. The study to determine egg/larval losses was conducted from 16 to 27 August, 1993, in cotton plots in Tift County, Georgia. Beet armyworm egg masses (approximately 100 eggs each), laid on wax paper, were attached to the undersides of leaves in each of four 0.5 acre plots. Two of the plots received weekly applications of conventional insecticide (the pyrethroid l-cyhalothrin) beginning the first week in July, whereas the other plots were untreated. Insecticide was applied in the treated plots immediately prior to, and twice during, the experiment. Twenty-four egg masses were placed in each plot (approximately 1 per 1000 plants; action threshold is 2-3 per 100 plants). Egg masses were observed daily for hatching and for indications of predation. After hatching, the wax paper was removed, surviving larvae were observed and counted 2, 4, 6, and 8 days post-hatch, and the presence and identity of predators on the leaves near the larval groups were recorded. High rates of loss were noted for egg masses exposed to predators in both the treated and untreated cotton, although loss was faster in the absence of insecticides (Table 5). Twice as many egg masses were entirely destroyed in the untreated cotton, however, as in the treated cotton. Most of the loss occurred in the egg and early-larval stages, as was suggested by Eveleens et al. (1973) and Hogg & Gutierrez (1980). We attribute this loss to predator activity. Thus, survival of beet armyworms was enhanced in the insecticide-treated plots. More predators were observed in association with the beet armyworm eggs and larvae in the untreated cotton than in the treated cotton (Table 4). For example, only larvae of the green lacewing, Chrysoperla rufilabris were found associated with beet armyworm eggs in the treated plots, compared with 11 different predators in the untreated plots. Thus, insecticide treatments disrupted a major portion of the beneficial arthropod complex. Two constraints limit the general applicability of these data concerning predation of beet armyworms. First, the plots were small, and widespread recolonization of treated plots by predators from adjacent untreated areas would be more rapid in these plots than would be the case for large cotton fields. Second, the density of beet armyworm egg masses placed in the plots was very low, which provided an excellent opportunity for the resident beneficial populations to eliminate them.however, this second point has some positive ramifications. Our data suggest that a conserved predator complex is capable of greatly reducing, and perhaps eliminating, low populations of the beet armyworm. Thus, the predator complex may be invaluable for eliminating

56 450 Florida Entomologist 77(4) December, 1994 TABLE 4. PREDATORS, AND THEIR FREQUENCY, FOUND IN ASSOCIATION WITH BEET AR- MYWORM EGGS AND LARVAE IN TREATED (PYRETHROID INSECTICIDE) AND UNTREATED COTTON (16-27 AUGUST 1993; TIFT. CO., GEORGIA). 1 Predator Taxon/Species Untreated Cotton Treated Cotton Heteroptera Orius insidiosus E (5, 8) L (2, 3) E (0, 0) L (3, 3) Geocoris punctipes E (2, 3) L (3, 3) E (0, 0) L (0, 0) Geocoris uliginosus E (1, 1) L (1, 1) E (0, 0) L (0, 0) Nabis roseipennis E (1, 1) L (2, 3) E (0, 0) L (0, 0) Zelus sp. E (1, 2) L (1, 1) E (0, 0) L (0, 0) Posisus maculiventris E (1, 1) L (1, 1) E (0, 0) L (0, 0) Dermaptera Labidura riparia E (1, 1) L (2, 4) E (0, 0) L (0, 0) Coleoptera Coccinella 7-punctata E (0, 0) L (1, 1) E (0, 0) L (1, 1) Neuroptera Chrysoperla rufilabris E (2, 2) L (5, 7) E (2, 2) L (0, 0) Hemerobius sp. E (1, 1) L (0, 0) E (0, 0) L (1, 2) Diptera Syrphid E (0, 0) L (1, 1) E (0, 0) L (0, 0) Hymenoptera Solenopsis invicta E (3, 24) L (3, 16) E (0, 0) L (3, 21) Araneida Spiders E (1, 1) L (1, 1) E (0, 0) L (0, 0) Totals E (19, 45) L (23, 42) E (2, 2) L (8, 27) 1 E = egg masses, L = larval clutches. The numbers in parenthesis after each letter are, respectively, 1) the number of egg masses or larval clutches on which the predator was found, and 2) the total number of the predator taxon observed in association with beet armyworm eggs or larvae. incipient beet armyworm populations, at least until sufficient egg and larval populations are present in the field to outstrip the predators capacity to consume a substantial majority of the eggs and larvae. Pupal Mortality. We examined loss in the pupal stage by placing ultimate-instar beet armyworm larvae under a styrofoam cup, with an opened, 30-ml diet cup containing artificial diet and a larva on the soil surface. Larvae were placed in two plots (100 per plot) of each of two treatments, insecticide-treated and unsprayed. A styrofoam collar (9 cm diam) into which the covering cup fit snugly was forced into the ground until its rim was level with the soil surface (about 6 cm). This prevented escape of the larvae because beet armyworms pupate in the upper 2-3 cm of soil. The opened diet cup with larva was then placed inside the collar and a styrofoam cup, which fit snugly into the collar, was placed over the cup with the insect. After the larvae had entered the soil and pupated, the covering cups were removed to expose the pupation sites to biotic and abiotic conditions in the field. Twenty additional cups and larvae were set up, with the covering cups left in place to trap the adult moths at emergence. These sentinel larvae were observed every day for adult emergence. When

57 Ruberson et al.: Armyworm Symposium TABLE 5. LOSS OF BEET ARMYWORM EGG MASSES, EGGS AND LARVAE IN TREATED AND UNTREATED COTTON (MEAN ± SD; AUGUST, 1993.) No. Egg Masses Remaining No. Larvae Remaining/Egg Mass Days of Exposure Treated Untreated Treated Untreated 0 24 ± ± ± ± 9.01 H ± ± ± ± 13.4 H ± ± ± ± 11.7 H ± ± ± ± 3.8 H ± ± ± ± H refers to hatch. Thus H+2 means 2 days after egg hatch. adult emergence was complete in the sentinel cups, all pupation sites were excavated, and the status of the pupal remains determined. Loss of pupae was surprisingly high in both treatments. Only 42.3% ± 2.12 (SD) of the pupae produced adult moths in the treated plots, compared with 21.0% ± 0.28 pupae surviving to adult emergence in the untreated plots. Thus, loss in the untreated plots was twice that observed in the insecticide-treated plots, although both treatments sustained fairly high mortality. Much of the loss observed in the experiment may be attributable to activity of imported fire ants, Solenopsis invicta. Fire ants were abundant in both fields, although they appeared to be more common in the untreated cotton than in the treated plots. Fire ants were observed removing pupal parts from pupation sites during the experiment; such sites afterward yielded no signs of pupal remains when excavated. CONCLUSIONS Although some of the results reported above are preliminary, summing up all of the mortality factors and their impacts yields a mortality rate in excess of 99% in untreated cotton. This suggests that the natural enemy complex functioning in cotton has the capacity to suppress beet armyworm populations. This conclusion, suggested by the California research reviewed above, points to the necessity of conserving the natural enemies for effective suppression of the beet armyworm. The completion of the active phase of the Boll Weevil Eradication Program in most of Georgia has provided the cotton production system an enormous opportunity to utilize natural enemies. In the absence of early-season applications of organophosphate insecticides to control the boll weevil, the natural enemy populations are able to increase in the cotton crop, and use of selective insecticides on a strictly as-needed basis will permit growers to realize the full benefits of these natural enemies. Under this system, the beet armyworm should not be a serious pest, except in cases where other pest control approaches disrupt the complex of resident beneficial organisms. Growers will reap benefits, however, beyond the natural control of beet armyworm populations. The complex of natural enemies that attacks the beet armyworm is comprised of generalists that will also provide some level of suppression of other arthropod pests in the system, as well, and benefit the overall cotton insect management program.

58 452 Florida Entomologist 77(4) December, 1994 ACKNOWLEDGMENTS We appreciate the assistance of Eddie McGriff (Decatur Co.), Mark Mitchell (Seminole and Miller Cos.), Jack Wall (Dooly Co.), and Mark Crosby (Laurens Co.) in locating fields for larval collections. Ray Wilson, Daniel West, Elizabeth Cravey, Wes Shiver, and Russ Ottens helped collect larvae and assisted with various aspects of the experiments. We greatly appreciate the determinations of the parasitoid species by Drs. R. W. Carlson (Ichneumonids), E. E. Grissell (Chalcididae), P. M. Marsh (Braconids), and N. E. Woodley (Tachinids). The comments of Drs. Robert M. McPherson (Univ. of Georgia) and Robert Lynch (USDA-ARS) on the manuscript are also appreciated. REFERENCES CITED ALI, A. M., AND M. J. GAYLOR Effects of temperature and larval diet on development of the beet armyworm (Lepidoptera: Noctuidae). Environ. Entomol. 21: BREWER, M. J., AND J. T. TRUMBLE Field monitoring for insecticide resistance in beet armyworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 86: CHU, Y., AND H. WU The studies on emergence, copulation and oviposition of adult beet armyworm (Spodoptera exigua Hübner). Chinese J. Entomol. 12: (in Chinese). COBB, P. P., AND M. H. BASS Beet armyworm: dosage-mortality studies on California and Florida strains. J. Econ. Entomol. 68: DE CLERCQ, P., AND D. DEGHEELE Laboratory measurement of predation by Podisus maculiventris and P. sagitta (Hemiptera: Pentatomidae) on beet armyworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 87: EVELEENS, K. G., R. VAN DEN BOSCH, AND L. E. EHLER Secondary outbreaks of beet armyworm by experimental insecticide applications in cotton in California. Environ. Entomol. 2: FRENCH, R. A Migration of Laphygma exigua Hübner (Lepidoptera: Noctuidae) to the British Isles in relation to large-scale weather systems. J. Anim. Ecol. 38: HARDING, J. A Heliothis spp.: parasitism and parasites plus host plants and parasites of the beet armyworm, diamondback moth and two tortricids in the Lower Rio Grande Valley of Texas. Environ. Entomol. 5: HARVEY, L. F New California and Texas moths. Canadian Entomol. 8: 54. HENNEBERRY, T. J., P. V. VAIL, A. C. PEARSON, AND V. SEVACHERIAN Biological control agents of noctuid larvae (Lepidoptera: Noctuidae) in the Imperial Valley of California. Southwest. Entomol. 16: HOGG, D. B., AND A. P. GUTIERREZ A model of the flight phenology of the beet armyworm (Lepidoptera: Noctuidae) in Central California. Hilgardia 48: KOLODNY-HIRSCH, D. M., D. L. WARKENTIN, B. ALVAREZ-RODRIGUEZ, AND R. KIRK- LAND Spodoptera exigua nuclear polyhedrosis virus as a candidate viral insecticide for the beet armyworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 86: KROMBEIN, K. V., P. D. HURD, D. R. SMITH, AND B. D. BURKS Catalog of the Hymenoptera in America north of Mexico. Smithsonian Instit. Press, Washington DC. MEINKE, L. J., AND G. W. WARE Tolerance of three beet armyworm strains in Arizona to methomyl. J. Econ. Entomol. 71: MITCHELL, E. R Migration by Spodoptera exigua and S. frugiperda North American style, pp in Movement of highly mobile insects. Concepts and methodology in research. North Carolina State University, Raleigh, NC.

59 Ruberson et al.: Armyworm Symposium OATMAN, E. R., AND G. R. PLATNER An ecological study of lepidopterous pests affecting lettuce in coastal southern California. Environ. Entomol. 1: PEARSON, A. C Biology, population dynamics, and pest status of the beet armyworm (Spodoptera exigua) in the Imperial Valley of California. Ph.D. dissertation, Univ. of California, Riverside Calif. POE, S. L., G. L. CRANE, AND D. COOPER Bionomics of Spodoptera exigua Hüb., the beet armyworm, in relation to floral crops. Proc. Trop. Reg. American Soc. Hortic. Sci. 17: RUBERSON, J. R., G. A. HERZOG, AND W. J. LEWIS Parasitism of the beet armyworm, Spodoptera exigua, in south Georgia cotton. Proc Beltwide Cotton Prod. Conf. 3: RUBERSON, J. R., G. A. HERZOG, W. R. LAMBERT, AND W. J. LEWIS Management of the beet armyworm: integration of control approaches. Proc Beltwide Cotton Prod. Conf. 2: SMITH, R. H Experiences with beet armyworm in cotton in Proc Beltwide Cotton Prod. Conf., pp SOTERES, K. M., R. C. BERBERET, AND R. W. MCNEW Parasitic insects associated with lepidopterous herbivores on alfalfa in Oklahoma. Environ. Entomol. 13: SWEZEY, O. H The winter revival of insect life in the arid region at Koko Head, Oahu. Proc. Hawaiian Entomol. Soc. 9: TINGLE, F. C., T. R. ASHLEY, AND E. R. MITCHELL Parasites of Spodoptera exigua, S. eridania (Lep.: Noctuidae) and Herpetogramma bipunctalis (Lep.: Pyralidae) collected from Amaranthus hybridus in field corn. Entomophaga 23: TURLINGS, T. C. J., J. H. TUMLINSON, F. J. ELLER, AND W. J. LEWIS Larval-damaged plants: source of volatile synomones that guide the parasitoid Cotesia marginiventris to the micro-habitat of its hosts. Entomol. exp. appl. 58: USDA-APHIS National boll weevil cooperative control program. Final environmental impact statement U.S. Govt. Printing Office, Washington DC. VAN DEN BOSCH, R., AND K. S. HAGEN Predaceous and parasitic arthropods in California cotton fields. Calif. Agric. Exp. Sta. Bull pages. WHITCOMB, W. H., AND K. BELL Predaceous insects, spiders, and mites of Arkansas cotton fields. Agric. Exp. Sta., Univ. of Arkansas, Bull pages. WILKINSON, J. D., K. D. BIEVER, AND C. M. IGNOFFO Synthetic pyrethroid and organophosphate insecticides against the parasitoid Apanteles marginiventris and the predators Geocoris punctipes, Hippodamia convergens, and Podisus maculiventris. J. Econ. Entomol. 72: WILSON, J. W Notes on the biology of Laphygma exigua Hübner. Florida Entomol. 16: WILSON, J. W The biology of parasites and predators of Laphygma exigua (Hübner) reared during the season of Florida Entomol. 17: WILSON, J. W The asparagus caterpillar: its life history and control. Florida Agric. Exp. Sta. Tech. Bull. 271: WOLFENBARGER, D. A., AND M. J. BREWER Toxicity of selected pesticides to field collected beet armyworm populations. Proc. 46th Beltwide Cotton Prod. Conf. 2:

60 454 Florida Entomologist 77(4) December, 1994 BEET ARMYWORMS (LEPIDOPTERA: NOCTUIDAE) IN NORTHEAST LOUISIANA: OBSERVATIONS ON AN UNCOMMON INSECT PEST E. BURRIS, J.B. GRAVES, B.R. LEONARD, AND C.A. WHITE Louisiana State University Agricultural Center, Baton Rouge, Louisiana ABSTRACT Outbreaks of beet armyworm, Spodoptera exigua (Hubner), in cotton in Louisiana occurred in 1983, 1988, 1992 and The outbreaks generally followed historic patterns observed in other locations, i.e., (l) local endemic populations developed rapidly for one or two generations when climatic conditions were favorable and (2) biological control organisms were suppressed by pesticides. Outbreaks of beet armyworm in Louisiana usually are less severe than in other southeastern states, because populations are usually lower and they occur in the latter part of the growing season. In 1993, beet armyworms infested more ha and caused higher levels of economic damage in Louisiana than in prior years. Insecticide screening tests conducted in 1993 indicated that Pirate (AC ) was more efficacious compared to all other insecticides. Beet armyworm larvae (2nd-3rd instar) were confined to Monsanto transgenic Bacillus thuringiensis (Bt) cotton (line 1076) and untreated Coker 312 in the laboratory. No significant (P 0.05) differences in leaf area consumed, mortality or pupal weights were detected. Key Words: Beet armyworm, Spodoptera exigua, cotton, insecticides RESUMEN Brotes del gusano trozador de la remolacha, Spodoptera exigua (Hubner), ocurrieron en el algodón de Louisiana en 1983, 1988, 1992 y Los brotes generalmente siguieron los patrones históricos observados en otras localidades, o sea, que (1) las poblaciones locales endémicas se desarrollaron rápidamente en una o dos generaciones cuando las condiciones climáticas fueron favorables y (2) los enemigos naturales fueron eliminados por los pesticidas. Los brotes del gusano trozador de la remolacha en Louisiana usualmente son menos severos que en otros estados del sureste, porque sus poblaciones son menores y aparecen al final de la estación. En 1993, los gusasnos trozadores de la remolacha infestaron más hectáreas y causaron más dano económico en Louisiana que en los anos anteriores. Las pruebas de tamizaje de insecticidas llevadas a cabo en 1993 indicaron que Pirate (AC ) fue más eficaz en comparación con otros insecticidas. En el laboratorio fueron confinadas larvas del gusano trozador de la remolacha (2 y 3 er instar) con algodón Monsanto transgénico de Bacillus thuringiensis (Bt) (línea 1076) y Cocker 312 sin tratar. No fueron detectadas diferencias significativas (P 0.05) en el area de hojas consumida, mortalidad o peso pupal. Beet armyworms, Spodoptera exigua (Hubner), were introduced into the western U.S. in the late 19th century (Chittenden 1902). They dispersed rapidly across the U.S. and, by the late 1920 s, they were recognized as a sporadic pest of cotton in the Southeastern U.S. (Wilson 1932). The earliest preserved beet armyworm specimens from Louisiana in the LSU Department of Entomology museum were collected in Baton Rouge. One specimen was collected from broadbean on 6 January 1932 and another from turnip on 29 September 1937 (Joan B. Chapin, Dept. of Entomology, LSU Agricultural Center, Baton Rouge, personal communication). Light trap collections of noctuids in Baton Rouge This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

61 Burris et al.: Armyworm Symposium from the years 1957 through 1960 revealed that beet armyworm moths were captured from 27 March through 6 December (Chapin & Callahan 1967). Beet armyworm larvae were collected from cole crops (Brassicae) in mid-december (date unknown) in Baton Rouge (Oliver & Chapin 1981). Outbreaks of beet armyworm are reported to be sporadic, occurring roughly every 2-5 years (Rabb & Kennedy 1979). These outbreaks typically occur one or two generations after favorable climatic conditions are accompanied by suppression of biological control agents by pesticides used for control of other pests. In Florida, no evidence of hibernation has been observed, and all stages of the insect are found throughout the year (Smith 1993). The ability of beet armyworms to overwinter is limited by frost kills of host plants and by temperatures below 10 C (Butler et al. 1976). Whether or not the beet armyworm hibernates and/or overwinters in Louisiana is unknown. Average daily minimum temperatures ( ) for the Northeast Research Station at St. Joseph for January, February, March, are respectively 3.0, 4.2 and 7.8 C (Thompson et al. 1983). Therefore, during most years, cold temperatures in Northeastern Louisiana may cause high mortality of beet armyworms. Other deterrents to population growth of beet armyworms in Louisiana may exist. For example, they are polyphagous feeders that damage vegetable crops, ornamentals and field crops, and commercial production of ornamentals and vegetable crops within the major cotton production regions in Louisiana is limited. Several private agricultural consultants have annually reported problems with beet armyworms in isolated fields in Northeast Louisiana. This suggests a need for studies of migration and/or overwintering biology of beet armyworm in this area. Their scouting records show armyworm egg masses (fall armyworm, Spodoptera frugiperda (J.E. Smith) or beet armyworm) were found in August of 1980, 1983 and 1985, and in July of 1988 and The 1988 beet armyworm outbreaks in Louisiana occurred within two days of severe outbreaks in Alabama (Ed Jones Consulting Service, Rayville Louisiana, personal communication). In 1993, the first beet armyworm egg masses were observed in June (Ray Young, Young Consulting Service, Wisner, Louisiana, personal communication). Beet armyworm outbreaks in Louisiana have generally followed the patterns observed in other major cotton producing states in the Southeastern U.S [Alabama (Smith 1985, 1989a, 1989b, 1993, 1994), Georgia and Mississippi], except that the percent of the total cotton ha infested is usually lower (Head ). An exception occurred in 1993 in Louisiana, when 242,820 of the 354,113 ha harvested (69%) were infested and economic injury occurred on about 80,940 ha (23% of harvested ha, Williams, 1994). Reports from Alabama indicated 73% of the ha planted to cotton were infested with beet armyworm in 1993 with 26% of harvested ha suffering economic damage (Williams 1994). Williams (1994) also reported that 82% of the cotton (546,345 ha) in Mississippi was infested with beet armyworms in 1993 and 69% received one or more insecticide applications for beet armyworm control. The outbreaks of beet armyworm in 1983 and 1992 in Louisiana provided an opportunity to evaluate the efficacy of several insecticides. In 1983, the pyrethroids cypermethrin ( kg AI/ha), flucythrinate (0.09 kg AI/ha) and tralomethrin (0.021 kg AI/ha) failed to provide satisfactory control of the pest. However, maximum labeled rates of sulprofos (1.68 kg AI/ha) and profenofos (1.12 kg AI/ha) as well as methomyl (0.5 kg AI/ha) and thiodicarb (0.67 kg AI/ha) provided satisfactory control (Burris 1983). In 1992, sulprofos (1.68 kg AI/ha), methomyl (0.51 kg AI/ha), thiodicarb (1.0 kg AI/ha) and Pirate (AC ) ( kg AI/ha) were the only insecticides that significantly reduced numbers of beet armyworm larvae (Graves 1993a, 1993b, 1993c). However, Pirate was the only insecticide that provided >90% control. Mix-

62 456 Florida Entomologist 77(4) December, 1994 tures of fenvalerate (0.17 kg AI/ha) + profenofos (0.56 kg AI/ha) and fenvalerate (0.17 kg AI/ha) + amitraz (0.28 kg AI/ha) also significantly reduced beet armyworm larval densities, but control by these treatments was only about 70%. The widespread beet armyworm infestations and numerous field control failures that occurred in 1993 in Louisiana prompted research to re-evaluate the efficacy of selected insecticides and to determine the effectiveness of transgenic cotton containing the Bt toxin on development of this pest. MATERIALS AND METHODS Insecticide Screening Tests. Northeast Research Station. Cotton (DPL 51) was planted on 8 May with plots consisting of four 19.8m rows with 102 cm centers. Treatments (see Table 1) were arranged in a randomized complete block design and replicated four times. Applications were made with a high clearance sprayer calibrated to deliver 93.5 liters total spray per ha through Teejet X-12 hollow cone nozzles (two per row) at 3.9 kg/cm 2. For Test 1, insecticide treatments were applied 29 July and 2 and 16 August. On 14 August, visual ratings were used to estimate the level of foliage feeding by the beet armyworm. TABLE 1. EVALUATION OF SELECTED INSECTICIDES AGAINST BEET ARMYWORM ON THE NORTHEAST RESEARCH STATION Treatment Rate/ha (kg AI) Visual 1 Ratings Percent 2 Control Test 1 UTC a 0 AC c 83 AC d 100 AC c 83 l-cyhalothrin a 0 Thiodicarb b 67 Thiodicarb c 83 AC l-cyhalothrin c 83 AC amitraz c 83 Test 2 UTC a 0 l-cyhalothrin ab 20 Profenofos ab 37 Profenofos + thiodicarb ab 57 Profenofos + l-cyhalothrin a 17 Profenofos + Bt (Design 100 WP) ab 20 Profenofos + methomyl b 80 1 Means followed by same letter do not significantly differ (P 0.05; Duncans MRT). For visual ratings: 0 (no feeding damage), 1 (feeding damage within the lower 1/3 of the plant), 2 (feeding damage in the lower 1/3 and middle 1/3 of the plant) or 3 (feeding damage throughout the plant). 2 Compared to UTC. 3 Formulated product.

63 Burris et al.: Armyworm Symposium For Test 2, treatments were applied on 19, 23, 27 July and 2, 6 and 17 August. Visual ratings of beet armyworm damage to foliage were made on 19 August. A visual defoliation rating for each plot was scored as follows: 0 (no feeding damage), 1 (feeding damage within the lower 1/3 of the plant), 2 (feeding damage in the lower 1/3 and middle 1/3 of the plant), or 3 (feeding damage throughout the plant). Macon Ridge Branch. Cotton (Stoneville 887) was planted on 2 June with plots consisting of four 15.2m rows with 102 cm centers. Treatments (see Table 2) were arranged in a randomized complete block design and replicated four times. Applications were made with a high clearance sprayer through Teejet X-8 hollow cone nozzles (2 per row) at 3.2 kg/cm 2. For Test 1, insecticide treatments were made on 4, 9, 20 and 31 August with 56.1 liters total spray per ha. Visual ratings of beet armyworm damage to all plots were made on 8 September using the rating system previously described. For Test 2, insecticide treatments were made on 30 August with 93.5 liters total spray per ha. The plots were sampled 7 days after treatment using a shake cloth. Two samples were taken between the two center rows in each plot (total of 1.8 meters per plot). Plants were vigorously shaken on both rows to dislodge all larvae, which were then counted. Effects of Transgenic Bt Cotton on Beet Armyworm. A randomized block experimental design with four replications was used to compare the development of beet armyworms on cotton plants expressing the Bacillus thuringiensis (Bt) toxin (Monsanto line 1076) or the nontransgenic parent (Coker TABLE 2. EVALUATION OF SELECTED INSECTICIDES AGAINST BEET ARMYWORM ON THE MACON RIDGE BRANCH OF THE NORTHEAST RESEARCH STATION Treatment Rate/ha (kg AI) Efficiency Rating Percent 2 Control Test 1 Visual Rating 1 UTC a 0 AC c 97 AC c 90 AC c 100 AC methomyl c 100 l-cyhalothrin b 23 Profenofos ab 13 Test 2 Larvae/1.8m UTC a 0 AC b 87 Bt (Javelin 100WG) ab 47 Thiodicarb a 20 Thiodicarb ab 18 Methomyl a 18 Chlorpyrifos ab 38 1 Means followed by same letter do not significantly differ (P 0.05; Duncans MRT). For visual ratings: 0 (no feeding damage), 1 (feeding damage within the lower 1/3 of the plant), 2 (feeding damage in the lower 1/3 and middle 1/3 of the plant) or 3 (feeding damage throughout the plant). 2 Compared to UTC. 3 Formulated product.

64 458 Florida Entomologist 77(4) December, ). Seed for both genotypes was supplied by Monsanto Company (Agricultural Products, 800 North Lindbergh Boulevard, St. Louis, MO 63167) and planted 17 May in plots four 9.2m rows with 102 cm centers. All plots received an in-furrow treatment of PCNB plus etridiazole (1.40 kg AI/ha) plus acephate (0.84 kg AI/ha) at planting. Visual rating of defoliation were made on 2 August using the system previously described. A laboratory experiment was conducted to further examine the effects of transgenic Bt cotton on beet armyworm. Newly hatched aggregates of beet armyworm larvae were collected from several fields at the Northeast Research Station on 13 August and transported to the laboratory. Ten leaves per plot were randomly collected from the second and third position below the terminal from transgenic Bt cotton and Coker 312 cotton plants. Each leaf was placed in a petri dish and five beet armyworm larvae (50 per plot) were placed in each dish. The petri dishes were covered and larvae were allowed to feed for 72 h. Leaf area was determined for each leaf at the beginning and end of the experiment using a LiCor, (Li-3100) Area Meter (Lincoln, Nebraska). Surviving larvae were transferred to a petri dish containing fresh leaves collected from the same plots as previously described. The experiment was terminated when larvae pupated. Percent pupation and pupal weights were determined. RESULTS AND DISCUSSION Insecticide Screening Tests. Pirate (AC303630) was the only insecticide among those evaluated that consistently provided satisfactory control of beet armyworm larvae. Applications of Pirate at rates of kg AI/ha resulted in % control (Tables 1 and 2) at both locations of the Northeast Research Station. Similar control was observed at the same application rate (0.28 kg AI/ha) using two different efficacy ratings (83% control using visual ratings, Test 1, Table 1 versus 87% control using shake cloth, Test 2, Table 2). In all tests and at every rate tested Pirate, either alone or in combination with other insecticides, significantly (P 0.05) decreased defoliation. Also, significantly (P 0.05) fewer live larvae were observed than in the untreated plots or the plots treated with l-cyhalothrin, profenofos, methomyl or thiodicarb at low rates (0.31 and 0.45 kg AI/ ha). Thiodicarb at high rates (0.9 and 1.01 kg AI/ha) resulted in 41 and 83% control, respectively (Tables 1 and 2). Chlorpyrifos at 1.12 kg AI/ha only provided 38% control (Table 2). Transgenic Cotton Evaluations. Natural infestations of beet armyworm were present in all field plots. The visual observations of damaged leaves on 2 August indicated no significant (P 0.05) differ- TABLE 3. EVALUATION OF TRANSGENIC BT COTTON FOR BEET ARMYWORM CONTROL ON THE NORTHEAST RESEARCH STATION Treatment Percent Pupating Pupal Weight (g) Leaf area (% Consumed) Visual Ratings 1 Bt Line a a 49.35a 2.2a Coker a a 49.35a 2.4a 1 Means followed by same letter do not significantly differ (P 05; Duncan s MRT). For visual ratings: 0 (no feeding damage), 1 (feeding damage within the lower 1/3 of the plant), 2 (feeding damage in the lower 1/3 and middle 1/3 of the plant) or 3 (feeding damage throughout the plant).

65 Burris et al.: Armyworm Symposium ences for beet armyworm damage between nontransgenic cotton and transgenic cotton plants (Table 3). However, leaf area measurements were significantly (P 0.05) higher for Bt line 1076 than for Coker 312 parent line on 16 August (data not presented). When beet armyworm larvae were confined to Bt line 1076 and Coker 312 parent line leaves in the laboratory, there was no significant (P 0.05) difference in leaf consumption, mortality (% pupating) and pupal weights (Table 3). The Bt endotoxin present in line 1076 appeared to have little or no effect on beet armyworm development. REFERENCES CITED BUTLER, G. D., JR., S. L. POE, G. L. CRANE, C. NELLINGER, AND D. CLARK A computer model to evaluate chemical control of beet armyworms in chrysanthemum ranges in Florida. Florida Entomol. 59: BURRIS, E Evaluation of cotton insecticides in Louisiana. Proceedings Mississippi Entomological association. Vol. 3:14 CHAPIN, J. B., AND P. S. CALLAHAN A List of the Noctuidae (Lepidoptera, Insecta) collected in the vicinity of Baton Rouge, Louisiana. The Proceedings of the Louisiana Academy of Sciences, Vol. 30: CHITTENDEN, F. H Some insects injurious to vegetable crops. USDA Div. of Entomol. Bull. 33 N.S. pp GRAVES, J. B., B. R. LEONARD, AND P. A. CLAY. 1993a. Evaluation of selected insecticide mixtures against late season populations of bollworm, tobacco budworm and beet armyworm. Insecticide and Acaricide Tests 18:223. GRAVES, J. B., B. R. LEONARD, AND P. A. CLAY. 1993b. Evaluation of selected insecticide mixtures against late season populations of bollworm, tobacco budworm and beet armyworm. Insecticide and Acaricide Tests 18:223. GRAVES, J. B., B. R. LEONARD, AND P. A. CLAY. 1993c. Evaluation of Pirate against late season populations of bollworm, tobacco budworm and beet armyworm. Insecticide and Acaricide Tests 18:224. HEAD, R. B Cotton insect losses, in Proceedings, Beltwide Cotton Prod. Res. Conf., National Cotton Council, Memphis, Tenn. OLIVER, A. D., AND J. B. CHAPIN Biology and illustrated key for the identification of twenty species of economically important noctuid pests. Louisiana Agric. Expt. Stn. Bull. No RABB, R. L., AND G. G. KENNEDY Movement of highly mobile insects: concepts and methodology in research, pp in Proceedings of a Conference, Movement of Selected Species of Lepidoptera in the Southeastern United States, Raleigh, North Carolina. SMITH, R.H Fall and beet armyworm control in Proc. Beltwide Cotton Prod. Res. Conf., National Cotton Council, Memphis, TN. SMITH, R. H. 1989a. Experiences with beet armyworm control in cotton in 1988, pp in Proceedings Beltwide Cotton Prod. Res. Conf., National Cotton Council, Memphis, Tenn. SMITH, R. H. 1989b. Beet armyworms on cotton. Alabama Cooperative Extension Service, Auburn University, Alabama, Circular Anr-538. SMITH, R. H Managing cotton to avoid beet armyworms. Auburn University, Alabama, Newsletter 5, pp. May 6, SMITH, R. H Beet Armyworm: A costly caterpillar. Proceedings Beltwide Cotton Prod. Res. Conf., National Cotton Council, Memphis, Tenn. (In Press) THOMPSON, R. C., R. A. MULLER, AND S. H. CRAWFORD Climate at the Northeast Research Station, St. Joseph, Louisiana, Louisiana Agric. Expt. Bull. No.755. WILLIAMS, M.R Cotton insect losses estimates, Proc. Beltwide Cotton Conference, National Cotton Council, Memphis, TN. WILSON, J. W Note on the biology of Laphygma exigua (Huebner). Florida Entomol. 16:33-39.

66 This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL This document was created with FrameMaker Florida Entomologist 77(4) December, 1994 KEYS TO SOLDIER AND WINGED ADULT TERMITES (ISOPTERA) OF FLORIDA RUDOLF H. SCHEFFRAHN AND NAN-YAO SU Ft. Lauderdale Research and Education Center University of Florida, Institute of Food & Agric. Sciences 3205 College Avenue, Ft. Lauderdale, FL ABSTRACT Illustrated identification keys are presented for soldiers and winged adults of the following 17 termite species known from Florida: Calcaritermes nearcticus Snyder, Neotermes castaneus (Burmeister), N. jouteli (Banks), N. luykxi Nickle and Collins, Kalotermes approximatus Snyder, Incisitermes milleri (Emerson), I. minor (Hagen), I. schwarzi (Banks), I. snyderi (Light), Cryptotermes brevis (Walker), and C. cavifrons Banks, Family Kalotermitidae; Coptotermes formosanus Shiraki, Reticulitermes flavipes (Kollar), R. hageni Banks, R. virginicus (Banks), and Prorhinotermes simplex (Hagen), Family Rhinotermitidae; and Amitermes floridensis Scheffrahn, Mangold, & Su, Family Termitidae. Key Words: Identification, Kalotermitidae, Rhinotermitidae, Termitidae RESUMEN Se presentan claves ilustradas de identificacion para los soldados y los adultos con alas de las 17 especies de termes conocidas de la Florida, U.S.A.: Calcaritermes nearcticus (Snyder), Neotermes castaneus (Burmeister), N. jouteli (Banks), N. luykxi Nickle y Collins, Kalotermes approximatus Snyder, Incisitermes milleri (Emerson), I. minor (Hagen), I. schwarzi (Banks), I. snyderi (Light), Cryptotermes brevis (Walker), y C. cavifrons Banks, Familia Kalotermitidae; Coptotermes formosanus Shiraki, Reticulitermes flavipes (Kollar), R. hageni Banks, R. virginicus (Banks), y Prorhinotermes simplex (Hagen), Familia Rhinotermitidae; y Amitermes floridensis Scheffrahn, Mangold, y Su, Familia Termitidae. A number of identification keys have been published for the termites of the United States (Banks & Snyder 1920; Light 1934a,b; Snyder 1954; Weesner 1965), including four for the termites of Florida and the southeastern states (Emerson & Miller 1943, Miller 1949, Weesner 1965, Gleason & Koehler 1980). Additionally, Nickle & Collins (1989) have published a key to drywood termites (Kalotermitidae) of the eastern U.S., all of which occur in Florida. In fact, the only eastern Nearctic termite species not found in Florida are Reticulitermes arenincola Goellner known from Indiana and the Boston, Massachusetts, area (Dobson 1918, Snyder 1949) and R. tibialis Banks which extends its eastern range into Illinois and Indiana (Snyder 1954). Unclear or sparse illustrations, illustrations not drawn to scale or lacking a scale, heavy reliance on morphometrics, use of obsolete names, and typographical errors have, in some earlier termite keys, led to confusion and incorrect identifications, especially by nonspecialists. The recent addition of three species to Florida s termite fauna, Incisitermes minor (Hagen) (Scheffrahn et al. 1988), Amitermes floridensis Scheffrahn et al. (1989), and Neotermes luykxi Nickle & Collins (1989), has further rendered previous keys obsolete.

67 Scheffrahn and Su: Florida Termite Keys 461 The condition of a specimen will greatly affect the probability of a correct identification. Desiccated specimens are the most difficult to identify because of the resultant shrinkage, color change (usually darkening), and fragile nature resulting in lost or broken appendages. Usually only the wings and mandibles of dry specimens tend to remain relatively unaltered. Termites are best kept alive after collection and then killed by freezing just before being keyed. If specimens cannot be kept alive, they should be immersed in aqueous ethanol or isopropanol of at least 40% (i.e., 80-proof liquor or rubbing alcohol). For long-term museum storage, 85% ethanol has proven to be the best preservative (M.S. Collins, pers. comm.). The mandibles of dead soldiers are usually crossed and the labrum may be retracted or folded. Alate wings, especially in the critical costal region, may curl ventrally. Wings alone are often collected following a dispersal flight and can yield at least a generic identification using the adult key. Wings must be completely flat in order to see the costal venation in proper perspective. This can be accomplished by immersing the wing in a reservoir of water or alcohol, sliding it onto a microscope slide or other clear flat surface, covering it with another slide, and allowing it to dry. Alternatively, a dried wing can be flattened by laying the dorsal surface on a drop of water. Wing membrane texture can best be observed when dry. Because the winged adult key uses the forewing, several wings should be examined and keyed if detached from the body. Alates collected before dispersal flights may be incompletely sclerotized causing cuticle, wing membrane, and veins to be lighter in color than when fully mature for flight. From a practical standpoint, correct identification is especially critical for pest taxa, such as termites, which may require very different control methods depending upon the target species. Although the morphological diversity among the termites of Florida is moderately broad, some species are not easily distinguished. Fortunately, the tentative identification of the soldier caste can be confirmed or refuted by independent identification of the winged adult (alate), and vice versa. Alates, however, are seasonal, may be difficult to find, and occur only in a mature colony. In the following keys, we attempt to separate species by parsimonious use of the most recognizable and consistent characters even if resultant groupings are not taxonomically related (e.g., grouping by presence or absence of wing membrane pigmentation). Simple measurements are used to supplement couplets or when dimension provides a clear separation of a group or species. This reduces the confusion resulting from the presence of long- and short-headed soldier forms in some kalotermitid species (Nickle & Collins 1989) or size variations among conspecific soldiers due to colony size, age, or nutritional status. Adult measurements are less variable than those of soldiers. To further help in identification, we have incorporated the known Florida distribution, pest status, and dispersal flight data based on a previous survey (Scheffrahn et al. 1988) and unpublished records. These are given only as general guidelines and exceptions may occur (i.e. autumn flights by Incisitermes snyderi (Light) and Reticulitermes spp.). When available, generally accepted common names (Snyder 1954, R.H.S. unpublished) or accepted common names (ESA 1989) are also provided. Several taxonomic issues must be addressed with respect to this work. The first and most troublesome, is the character overlap between Neotermes jouteli (Banks) and Neotermes luykxi. All measurements and characters provided in the description of N. luykxi soldiers and adults (Nickle & Collins 1989) fall within the range of those designated as N. jouteli in our reference collection. Apparently, the only diagnostic characters separating the two are their respective chromosome numbers and allozyme patterns (Luykx et al. 1990) neither of which can be obtained from preserved specimens. Chromosome number has been shown to vary within single insect species (Emmel et al. 1973) although it appears to be stable within species of Kalotermitidae

68 462 Florida Entomologist 77(4) December, 1994 (Luykx 1990). Therefore, we cannot differentiate between these two Neotermes species in either key. Secondly, Prorhinotermes simplex (Hagen), the Florida dampwood termite, shares its habit of nesting in moist, decaying wood with the true dampwood termites (Neotermes spp.), although the former is actually a subterranean termite (Family Rhinotermitidae) which we have observed tunneling in soil. Thirdly, separating soldiers of Reticulitermes species is difficult. Although head and pronotum measurements and mandibular characters are useful, precise micromeasurements are required and some interspecific overlap may occasionally surface. In subsequent studys (Hostettler et al. 1995), labrum shape, although also subtle, has been found to be an effective additional character for separating soldiers of all three Reticulitermes species. Finally, an ongoing revision of Nearctic Reticulitermes suggests that an undescribed or erroneously synonymized species may occur in Florida s panhandle (T. Myles pers. comm.). MATERIALS AND METHODS Line drawings of specimens were prepared at 20-80x magnification with the aid of a camera lucida attached to an Olympus SZH light microscope. Measurements were made with an ocular micrometer. Scanning electron micrographs were made with a Hitachi S-4000 field emission microscope (6-8 kv). Specimens were dehydrated in absolute ethanol and 1,1,1,3,3,3- hexamethyldisilazane (Nation 1983) prior to sputter coating with gold. Material examined for this key is from the authors reference collection containing about 1,200 colony samples taken in Florida between 1985 and 1994 including 785 samples collected from structures in peninsular Florida (Scheffrahn et al. 1988), the Florida State Collection of Arthropods (Fla. Dept. Agric. Cons. Serv., Division of Plant Industry, Gainesville, Florida) and the E.M. Miller collection from the University of Miami on loan from P. Luykx containing 111 samples taken in Florida between 1930 and KEY TO TERMITE SOLDIERS OF FLORIDA 1 Pronotum (Fig. 8a) as wide or wider than width of head viewed from above (Figs dorsal views); for species with prominent mandibles, inner margin of left mandible (Fig. 8e) with two or more marginal teeth (Figs dorsal views). (Drywood and dampwood termites) Family Kalotermitidae Pronotum width narrower than width of head viewed from above; each mandible with no exposed teeth or only one tooth visible on inner margin (Figs dorsal views). (Subterranean termites) Families Rhinotermitidae and Termitidae Frons nearly vertical with deep furrow or rimmed above by a ridge; head plug-like; mandibles not prominent; head color deep reddish brown to black; (Figs. 1,2,4-7). (Powderpost drywood termites) Frons (Fig. 8c) slopes more or less gradually from plane of vertex (Fig. 8b), surface smooth; head flattened, quadrate or elongate; mandibles project prominently; head color orange to reddish brown; (Figs. 8-21). (Drywood and dampwood termites) Frons with deep furrow (Figs. 1,2); foretibia with one prominent spur at right angle to tibial axis and two small apical spurs (Fig. 3). (Rare in structures, known from Clay Co. to Sebring.) Calcaritermes nearcticus a

69 Scheffrahn and Su: Florida Termite Keys 463 Figs Dorsal and lateral views of heads and pronota of termite soldiers from Florida: Calcaritermes nearcticus, Figs. 1-2 (foreleg, Fig. 3; bar = 3.2 mm); Cryptotermes cavifrons, Figs. 4-5; Cryptotermes brevis, Figs. 6-7; Neotermes jouteli, Figs. 8-9 (pronotum 8a, vertex 8b, frons 8c, antenna 8d, mandible 8e, labrum 8f, eyespot Fig. 9a). Bar = 2 mm.

70 464 Florida Entomologist 77(4) December, Ridge surrounding frons forming bowl (Figs. 5,7); foretibia lacking a prominent apical spur Vertex smooth (Figs. 4,5); smaller species. (Uncommon structural pest (moderate moisture requirement), known from St. Johns Co. south) Cryptotermes cavifrons a - Vertex rough, wrinkled, often concave (Figs. 6,7); larger species. (Introduced species; common pest of structures and furniture statewide; never found in non-structural wood - West Indian powderpost termite.)......cryptotermes brevis 5 Eyespot (Fig. 9a) black; antennae with up to 19 segments; third antennal segment about as long as fourth and fifth combined; soldiers usually large (Figs. 8,9). (Occasional pest in moisture-exposed wood, known from Ft. Pierce south)... Neotermes jouteli a,b - Eyespot hyaline or indistinct; number and size of antennal segments variable, soldier size variable Third antennal segment greatly enlarged and club-like, as long or longer than fourth through sixth combined, and about twice as wide as fourth (Figs. 10,11,S1); larger species. (Regularly introduced into Florida as a structural pest, may be permanently established in some areas - western drywood termite)... Incisitermes minor - Third antennal segment shorter than fourth through sixth combined and less than twice as wide as fourth Anterior margin of pronotum weakly and evenly concave; length of third antennal segment less than fourth and fifth combined (Figs ) Anterior margin of pronotum incised (Figs. 16,18,20); length of third antennal segment about equal to or greater than fourth and fifth combined (Figs. S3,S5) Pronotum more than twice as wide as long, collar-like, posterior margin with rounded corners; third antennal segment equal to or slightly longer than second or fourth; lateral margins of mandibles widen near bases but do not constitute humps ; large species (Figs. 12,13). (Occasional pest in moisture-exposed wood and living trees, known from Lake Co. south.)... Neotermes castaneus a - Pronotum more square, less than twice as wide as long, posterior margin nearly straight with square corners; third antennal segment longer than second but shorter than fourth and fifth combined; lateral margins of mandibles with distinct humps near bases; medium-small species (Figs. 14,15). (Uncommon structural pest, known from Sarasota north.)...kalotermes approximatus 9 Small species; antennae with segments; pronotum about 1 mm wide, posterior margin rather evenly convex (Figs. 16,17). (Known only from Florida Keys, pest status unknown)... Incisitermes milleri a - Medium species; antennae with segments; pronotum mm wide, posterior margin more straight or slightly concave in middle, corners rounded (Figs. 18,20) Tip of labrum bluntly pointed (Fig. S2); third antennal segment as long as fourth and fifth combined (Fig. S3); antennae with segments (Figs. 18,19). (Common in structural wood statewide - southeastern drywood termite)... Incisitermes snyderi - Tip of labrum truncate (Fig. S4); third antennal segment longer than fourth and fifth combined (Fig. S5); antennae with up to 16 segments (Figs. 20,21). (Rare structural pest, known mostly from coastal south)...

71 Scheffrahn and Su: Florida Termite Keys 465 Figs Dorsal and lateral views of heads and pronota of termite soldiers from Florida: Incisitermes minor, Figs ; Neotermes castaneus, Figs ; Kalotermes approximatus, Figs ; Incisitermes milleri, Figs Bar = 2 mm.... Incisitermes schwarzi a 11 Teeth on inner margin of mandibles reduced to serrations at base and so usually hidden from view by labrum; head capsule not elliptical when viewed laterally (Figs ). Family Rhinotermitidae...12

72 466 Florida Entomologist 77(4) December, 1994 Figs. S1-S7. Scanning electron micrographs of Incisitermes minor soldier antenna, Fig. S1; I. snyderi soldier labrum, Fig. S2, and antenna, Fig. S3; I. schwarzi soldier labrum, Fig. S4 (mandible tips broken), and antenna, Fig. S5; Reticulitermes virginicus soldier mandibles, Fig. S6; R. hageni soldier mandibles, Fig. S7.

73 Scheffrahn and Su: Florida Termite Keys One prominent triangular tooth on inner margin of each sickle-shaped mandible, head capsule elliptical when viewed laterally (Figs. 22,23); smallest soldier caste in Florida. Family Termitidae. (Occasionally associated with structural lumber, known from west central Florida - Florida dark-winged subterranean termite)... Amitermes floridensis a 12 Head outline rectangular from above, sides of head parallel (Figs. 24,26,28). (Reticulitermes spp) Head outline oval or egg-shaped from above, narrowing in front, sides of head convex (Figs. 30,32) Pronotum width usually greater than 0.90 mm; head length with mandibles equal to or greater than 2.8 mm; points of mandibles, especially left, curved inward about (Figs. 24,25). (Widespread pest throughout state - eastern subterranean termite.)...reticulitermes flavipes c - Pronotum width usually less than 0.85 mm; head length with mandibles less than or equal to 2.7 mm; curvature of mandible points Larger species (Figs. 26,27), pronotum width greater than 0.70 mm; points of mandibles, especially left, curved inward about 70-90, points of mandibles broader (Fig. S6) than following species; basal serrations of left mandible, when exposed for viewing, more prominent (Fig. S6); distinct and gradual inward curvature of blade of right mandible (Fig. S6). (Widespread pest throughout state - dark southern subterranean termite)...reticulitermes virginicus c - Smaller species (Figs. 28,29), pronotum width less than or equal to 0.70 mm; points of mandibles, especially left, curved inward about 45, points more slender (Fig. S7) than above species; basal serrations of left mandible, when exposed for viewing, less prominent (Fig. S7); blade of right mandible more straight before point (Fig. S7). (Less common in structures statewide - light southern subterranean termite) Reticulitermes hageni c 15 Fontanelle consisting of a prominent, oval, anterior-facing opening arising from a mound on vertex and frons (Figs. 30,31a). (Introduced species common in or near structures in certain areas of Broward, Dade, Hillsborough, and Orange Cos., and coastal panhandle - Formosan subterranean termite.)... Coptotermes formosanus - Fontanelle consisting of a minute, circular, dorsal-facing opening on surface of vertex (Figs. 32a, 33). (Occasionally in structures in Broward and Dade Cos. - Florida dampwood termite)......prorhinotermes simplex a KEY TO WINGED ADULT TERMITES OF FLORIDA 1 With wing unfolded and flattened between glass plates, three or more sclerotized veins in costal field (costal margin, subcosta, radius, radial sector, and, in some, median, e.g., Fig. 37a,c-e and Fig. 45a-c) at about one-third wing length from wing suture; in most species, numerous diagonal cross veins connecting two or more remaining veins in costal field along remaining length of wing (Figs. 34,35-37, 39,41-45). (Drywood and true dampwood termites) Two sclerotized veins in costal field (costal margin and radial sector, e.g., Fig. 46a-b) in foremargin of wings visible along entire length of wing

74 468 Florida Entomologist 77(4) December, 1994 and, in most species, connected by short vertical cross veins in distal third of wing (Figs ,50,52,54). (Subterranean termites) When viewed over white background or with several wings overlapping as when folded over the abdomen, entire wing membrane translucently pigmented blackish; veins in costal field darker than membrane... 3 Figs Dorsal and lateral views of heads and pronota of termite soldiers from Florida: Incisitermes snyderi, Figs ; I. schwarzi, Figs ; Amitermes floridensis, Figs ; Reticulitermes flavipes, Figs ; R. virginicus, Figs ; R. hageni, Figs ; Coptotermes formosanus, Figs (fontanelle 31a); Prorhinotermes simplex, Figs (fontanelle 32a). Bar = 2 mm.

75 Scheffrahn and Su: Florida Termite Keys Wing membrane unpigmented or very faintly yellow-brown, veins in costal field white to medium brown when viewed as above In forewing, median vein is sclerotized and runs near veins in costal field (Fig. 34a), no diagonal cross veins connecting veins in costal field; wing membrane with distinct papillae or bumps; length with wings 7 mm. (Rare in structures, known from Clay Co. to Sebring, midday flights March to May.)...Calcaritermes nearcticus a - In forewing, median vein is unsclerotized and runs midway between veins in costal field above, and cubitus below; diagonal cross veins between sclerotized veins in costal field (Figs. 35,36) Head and pronotum orange-brown, abdomen dark brown; stout-bodied, medium-large species, length with wings mm; hairs on head shorter than diameter of eye; arolium absent between tarsal claws; in forewing, few diagonal cross veins branching forward from radial sector (Fig. 35). (Regularly introduced into Florida as a structural pest, may be permanently established in some areas, midday flights September to November - western drywood termite.)...incisitermes minor - Head, thorax, and abdominal tergites (plates) reddish brown; medium-small species, length with wings mm; hairs on head longer than diameter of eye; arolium present between tarsal claws; in forewing, few cross veins branching forward from median vein (Fig. 36a). (Uncommon in structures, from Sarasota north, daytime flights September to November)... Kalotermes approximatus 5 In forewing, four sclerotized veins in costal field (costal margin, radius, radial sector, and median, e.g., Fig. 37a, c-e, respectively) at about one-third wing length from body; sclerotized media running close to radial sector (Figs. 37,39); large, stout-bodied species.... (Dampwood termites) In forewing, three sclerotized veins in costal field (costal margin, radius, and radial sector, e.g., Fig. 45a-c, respectively) at about one-third wing length from body, media (Fig. 45d) not sclerotized and running midway between radial sector and cubitus (Figs ); size variable.... (Powderpost and drywood termites) Head, body, and veins in costal field chestnut brown; long erect hairs on head and pronotum (Fig. 38); largest alate caste in Florida, length with wings about mm (forewing, Fig. 37). (Occasional pest in moisture-exposed wood and living trees, known from Lake Co. south, evening flights peak in October and November.)......Neotermes castaneus a - Head, body, and veins in costal field light yellowish-brown to reddish-brown; very short hairs on pronotum (Fig. 40); wing membrane very faintly yellow-brown; length with wings mm (forewing, Fig. 39). (Occasional pest in moisture-exposed wood, known from Vero Beach south, evening flights Spring or Fall.)...Neotermes jouteli a,b 7 In forewing, unsclerotized media curving near mid-wing to join veins in costal field (Figs. 41a,42a; note variations in C. brevis forewing veination in Scheffrahn et al. (1988, Fig. 2.); head and body brown In forewing, unsclerotized media running to near tip of wing even if branched along its course (Figs. 43a,44a,45d) Small dull-brown species, length with wings 8-9 mm, wing membrane weakly tuberculate (pimply); head width (through eyes) mm;

76 470 Florida Entomologist 77(4) December, 1994 Figs Right forewing of termite adults from Florida: Calcaritermes nearcticus, Fig. 34 (median vein 34a); Incisitermes minor, Fig. 35 (median vein 35a); Kalotermes approximatus, Fig. 36 (median vein 36a); Neotermes castaneus, Fig. 37 (costal margin 37a, subcostal vein 37b, radius 37c, radial sector 37d, and media 37e); and dorsal view of pronotum, Fig. 38; N. jouteli, Fig. 39, and dorsal view of pronotum, Fig. 40; Cryptotermes cavifrons, Fig. 41 (median vein 41a); C. brevis, Fig. 42 (median vein 42a); Incisitermes milleri, Fig. 43 (median vein 43a); I. snyderi, Fig. 44 (median vein 44a). Bar = 4 mm for forewings, 2.4 mm for pronota.

77 Scheffrahn and Su: Florida Termite Keys 471 antennae with segments (forewing, Fig. 41). (Uncommon structural pest (moderate moisture requirement), known from St. Johns Co. south, evening flights year-round.)...cryptotermes cavifrons a - Medium dull-brown species, length with wings mm; head width mm; antennae with segments (forewing, Fig. 42). (Introduced species, common pest of structures and furniture statewide, never found in non-structural wood, evening and night flights April to July - West Indian powderpost termite)... Cryptotermes brevis 9 Head, thorax, and body dark brown; veins in costal field brown, wing membrane tuberculate; head width (through eyes) about 0.9 mm; ocellus more elliptical; small species, length with wings 7-8 mm; (forewing, Fig. 43). (Known only from Florida Keys, pest status unknown, daytime flights April to July)...Incisitermes milleri a - Head and body color pale yellow-brown to pale reddish brown; veins in costal field pale yellow-brown in distal half of wing; head width mm; ocellus more round; medium species, length with wings mm (forewing, Fig. 44). (Common in structures statewide, evening flights May to August - southeastern drywood termite.)......incisitermes snyderi - Head and body color medium brown; veins in costal field brown along entire length of wing; head width mm; ocellus more elliptical; medium-large species, length with wings mm (forewing, Fig. 45). (Rare structural pest, known mostly from coastal south, small evening or night flights except during winter, peaking in April and May) Incisitermes schwarzi a 10 Wing membrane smooth between veins; wing surface and margin adorned with fine hairs (Figs. 46,47) Wing membrane net-like (reticulate) between veins, no hairs on wing surface or margin (Figs. 48,50,52,54) Head and pronotum yellow-brown; wing membrane unpigmented; veins in costal field (Fig. 46a,b) yellowish brown at base to almost white at tip; large species, length with wings about 14 mm (forewing, Fig. 46). (Introduced species common in or near structures in certain areas of Broward, Dade, Hillsborough, and Orange Cos., and coastal panhandle, late afternoon and evening flights April to July- Formosan subterranean termite)... Coptotermes formosanus - Head and pronotum dark brown; wing membrane dark with black interior veins (Fig. 47); small species with wings long in proportion to body length; length with wings about 9 mm. (Occasional structural and outdoor nuisance (large swarms), known from west central Florida, daytime flights following rain June to September - Florida dark-winged subterranean termite (Family Termitidae)) Amitermes floridensis a 12 Body color pale brown to light reddish brown Body color dark brown to black Forewing not broad in middle, costal margin not convex, median vein runs uninterrupted below veins in costal field (Fig. 48a); thorax and abdomen narrow (Fig. 49); small species, length with wings 7-8 mm. (Less common in structures statewide, midday flights in sunshine December to April - light southern subterranean termite.) Reticulitermes hageni

78 472 Florida Entomologist 77(4) December, 1994 Figs Right forewing of termite adults from Florida: I. schwarzi, Fig. 45 (costal margin 45a, radius 45b, radial sector 45c, and media 45d); Coptotermes formosanus, Fig. 46 (costal margin 46a and radial sector 46b); Amitermes floridensis, Fig. 47; Reticulitermes hageni, Fig. 48 (median vein 48a), and dorsal view of body, Fig. 49; Prorhinotermes simplex, Fig. 50 (median vein 50a), and dorsal view of body, Fig. 51; R. flavipes, Fig. 52, lateral view of head, Fig. 53; R. virginicus, Fig. 54, lateral view of head, Fig. 55. Bar = 4 mm for forewings, 2.4 mm for heads and bodies. - Forewing broad in middle, costal margin covex, median vein disjointed, indistinct (Fig. 50a); thorax and abdomen broader than above (Fig. 51); medium-small species, length with wings 9-10 mm. (Occasionally in structures in Broward and Dade Cos., evening and night flights October to January - Florida dampwood termite.) Prorhinotermes simplex a

79 Scheffrahn and Su: Florida Termite Keys Medium-small species, length with wings mm; ocellus about one time its diameter or more from compound eye (Fig. 53); veins in costal field of wing light brown, membrane faintly yellow-brown (forewing, Fig. 52). (Widespread pest throughout state, midday flights in sunshine January to April - eastern subterranean termite)......reticulitermes flavipes - Small species, length with wings mm, usually mm; ocellus less than its diameter from compound eye (Fig. 55); veins in costal field of wing whitish or hyaline except near base, membrane hyaline (forewing, Fig. 54). (Widespread pest throughout state, midday flights in sunshine March to May - dark southern subterranean termite.)......reticulitermes virginicus FOOTNOTES FOR KEYS a In the United States, known only from Florida. b Indistinguishable from Neotermes luykxi. See introduction. c For additional characters see Hostettler et al. (1995). ACKNOWLEDGMENTS We thank D.S. Williams of the ICBR Electron Microscope Core Facility at the University of Florida, Gainesville, for technical assistance with electron microscopy; and M. S. Collins, F. W. Howard, S. Jones, J. Krecek, P. Luykx, T. Myles, J. Peters, and J. Tsai for reviewing and improving various stages of this contribution no. R of the Florida Agricultural Experiment Stations Journal Series. REFERENCES CITED BANKS, N. AND T.E. SNYDER A revision of the Nearctic termites with notes on biology and geographic distribution. U.S. Natl. Mus. Bull. 108, 228 pp. DOBSON, R.J A European termite Reticulotermes [sic!] lucifugus Rossi in the vicinity of Boston. Psyche 25: EMERSON, A.E., AND E.M. MILLER A key to the termites of Florida Entomol. News 54: EMMEL, T.C., H.R. TREW, AND O. SHIELDS Chromosomal variability in a Nearctic lycaenid butterfly, Philotes sonorensis (Lepidoptera: Lycaenidae). Pan-Pacific Entomol. 49: ENTOMOLOGICAL SOCIETY OF AMERICA Common names of insects and related organisms. Lanham, MD. 199 pp. GLEASON, R.W., AND P.G. KOEHLER Termites of the eastern and southeastern United States: pictorial keys to soldiers and winged reproductives. Flor. Coop. Ext. Serv., Inst. Food Agric Sci., Univ. Florida Bull HOSTETTLER, N.C., D.W. HALL, AND R.H. SCHEFFRAHN Morphometric variation and labral shape in Florida Reticulitermes (Isoptera: Rhinotermitidae): significance for identification. Florida Entomol. (in press). LIGHT, S.F. 1934a. The desert termites of the Genus Amitermes, pp in C. A. Kofoid [ed.]. Termites and termite control. University of California Press, Berkeley, Calif. 795 pp. LIGHT, S.F. 1934b. Dry-wood termites, their classification and distribution, ibid. pp LUYKX, P A cytogenetic survey of 25 species of lower termites from Australia. Genome 33:

80 474 Florida Entomologist 77(4) December, 1994 LUYKX, P., D.A. NICKLE, AND B.I. CROTHER A morphological, allozymic, and karyotypic assessment of the phylogeny of some lower termites (Isoptera: Kalotermitidae). Proc. Entomol. Soc. Washington 92: MILLER, E.M A handbook on Florida termites. Tech. Ser., Univ. Miami Press, Coral Gables, FL, 30 pp. NATION, J.A A new method using hexamethyldisilazane for the preparation of soft insect tissue for scanning electron microscopy. Stain Technol. 55: NICKLE, D.A., AND M.S. COLLINS Key to the Kalotermitidae of eastern United States with a new Neotermes from Florida (Isoptera). Proc. Entomol. Soc. Washington 91: SCHEFFRAHN, R.H., J.R. MANGOLD, AND N.-Y. SU A survey of structure-infesting termites of peninsular Florida. Florida Entomol. 71: SCHEFFRAHN, R.H., N.-Y. SU, AND J.R. MANGOLD Amitermes floridensis, a new species and first record of a higher termite in the eastern United States (Isoptera: Termitidae: Termitinae). Florida Entomol. 72: SNYDER, T.E Catalog of the termites (Isoptera) of the world. Smithsonian Misc. Coll. No. 3953, 112: SNYDER, T.E Order Isoptera. The termites of the United States and Canada. Natl. Pest Contr. Assn., New York, NY, 64 pp. WEESNER, F.M Termites of the United States, A handbook. Natl. Pest Contr. Assn., Elizabeth, New Jersey, 70 pp.

81 474 Florida Entomologist 77(4) December, 1994 ODONTOTAENIUS FLORIDANUS NEW SPECIES (COLEOPTERA: PASSALIDAE): A SECOND U.S. PASSALID BEETLE JACK C. SCHUSTER Systematic Entomology Laboratory Universidad del Valle de Guatemala Aptdo. 82 Guatemala City, GUATEMALA ABSTRACT Larvae and adults of Odontotaenius floridanus New Species are described from the southern end of the Lake Wales Ridge in Highland Co., FL. This species may have evolved as a population isolated during times of higher sea level from the mainland species O. disjunctus (Illiger) or a close common ancestor. It differs notably from O. disjunctus in having much wider front tibiae and a less pedunculate horn. A key is given to the species of the genus. Key Words: Florida, endemism, Lake Wales RESUMEN Son descritas las larvas y adultos de Odontotaenius floridanus Nueva Especie del extremo sur de Lake Wales Ridge, en Highland Co., Florida. Esta especie pudo haber evolucionado, como una población aislada en épocas en que el nivel del mar era This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

82 Schuster: Odontotaenius floridanus, A New U.S. Passalid 475 más alto, a partir de O. disjunctus (Illiger) o de otro ancestro común cercano. Difiere notablemente de O. disjunctus en tener las tibias delanteras más anchas y el cuerno menos pedunculado. Se ofrece una clave para las especies del género. Only one species of Passalidae, Odontotaenius disjunctus (Illiger), is known to occur in the U.S. at present, though two other species may have occurred in Arizona at the turn of the century (Schuster 1983). Recently, while examining O. disjunctus in the Florida State Collection of Arthropods in Gainesville, I noted two specimens with a somewhat different morphology. On examining the labels, I noticed that the collector, L. L. Lampert, had also remarked on their uniqueness. Both came from the same area of south-central Florida in Highlands Co. Later, I had an opportunity to examine the collection of the Archbold Biological Station and encountered three more similar specimens. Subsequently, with the aid of four other Florida entomologists, I conducted a search for this elusive insect at the Archbold site. One of the collaborators, Paul Skelley, managed to find an oak log with a pair of adults and four larvae. All seven adult specimens fit the description below. The genus Odontotaenius was revalidated by Reyes-Castillo (1970) and includes eight species (Reyes-Castillo 1970, Castillo et al. 1988). Three of these species, known only from the types, are probably synonyms of either O. striatopunctatus (Percheron) or O. zodiacus (Truqui). The types need to be examined to make further synonymies. Odontotaenius is characterized by a marked fronto-clypeal suture, thick anterior clypeal border, median frontal structure of striatopunctatus type, posterior half of supraorbital ridges bifurcate, and short antennal lamellae with the distal margins rounded (Reyes-Castillo 1970). Odontotaenius floridanus Schuster New Species DESCRIPTION. Head: Anterior border of labrum deeply concave. Clypeus greatly swollen in middle. Median frontal structure ( horn ) of striatopunctatus type (Reyes-Castillo 1970) (Fig. 1). Supraorbital ridge bifurcated posteriorly. Canthus extends laterally same distance as eye margin, anterior corner with right angle. Eyes small, ratio of dorsal eye width to head width 1:11.4. Lateral circular scars of mentum indistinct or absent. Thorax: Lateral fossa of pronotum with 1-4 light punctations. Mesosternum bare, without lateral depressions, with wide (0.7 mm) matt bands along anterior borders. Mesepisternum pubescent. Metasternum with punctations delimiting each latero-posterior side of disk, anterior angles pubescent, lateral fossae narrow and pubescent. Elytra: Anterior profile convex, anterior face of elytra sloping. Striae 7 and 8 united anteriorly. Wing: Normal size, not reduced. Legs: Tibia I very wide (Fig. 2), tibia ratio (see Fig. 3) , x = 0.312, n = 7. Dimensions (mm): Total length, mandible tip to elytral tip 36-42, x = 40.2; elytral length , x = 20.2, elytral width 12-15, x = 13.6, pronotal width , x = LARVA. The larva has only AR setae and 7-8 internal coxal setae. Otherwise, the basic setal pattern is the same as in O. disjunctus. Larval head widths: instar III = 6.0, instar II =

83 476 Florida Entomologist 77(4) December, 1994 Fig. 1. Lateral view of heads of O. disjunctus (left) and O. floridanus (right) showing horn (median frontal structure) shapes. DIAGNOSIS. The adult is most similar to O. disjunctus. The latter species differs from O. floridanus in having a rounded canthus which projects beyond the eye margin, larger eyes (dorsal eye width to head width ratio 1:9.5-1:10), mentum with dis- Fig. 2. Dorsal view of front tibiae of O. disjunctus (left) and O. floridanus (right) showing size difference.

84 Schuster: Odontotaenius floridanus, A New U.S. Passalid 477 Fig. 3. Passalid tibia, ventral view, showing parameters of tibia ratio. The ratio is the width at the widest point divided by the length shown. tinct lateral circular scars, usually no punctations in lateral prothoracic fossae, mesosternum with narrow (0.3 mm) matt bands along anterior borders, metasternum without punctations delimiting latero-posterior sides of disk, flatter anterior profile of elytra, anterior face of elytra vertical, elytral striae 7 and 8 not united anteriorly, and narrow tibia I (ratio , x = 0.216, n = 8 Florida specimens). The aedeagi of the two species are very similar. The horn on the head of O. floridanus is not as pedunculate as in O. disjunctus or O. striatopunctatus (Fig. 1). The larva is similar to that of O. disjunctus (Schuster & Reyes-Castillo 1981). It has fewer AR setae and usually more internal coxal setae than larvae of the latter species. MATERIAL EXAMINED. Seven adults and four larvae. TYPE MATERIAL. Holotype male, Florida, Highlands Co., Archbold Biological Station near Lake Placid, 18 I 93, P. Skelley, in a hardwood log. Allotype female, same data as holotype, together with four larvae. Paratypes: Four specimens collected at Archbold Biological Station: 28 III 1973, L.L. Lampert, in pitfall trap; VIII 1970, J. Douglas; 20 IX 1992, U.G. Mueller; 23 IV 1983, M. Deyrup, ground. Another specimen was collected at Sebring, Highlands Co., Flamingo Villas Scrub by P. Martin, 11 IX Types will be deposited in the Florida State Collection of Arthropods and the collection of the Universidad del Valle de Guatemala. ETYMOLOGY. The trivial name floridanus refers to the state where this species is apparently endemic.

85 478 Florida Entomologist 77(4) December, 1994 DISCUSSION This species is obviously derived from Odontotaenius disjunctus or a close common ancestor. It differs most notably in the horn, or median frontal structure, and the enlarged front tibiae. This latter characteristic is found in passalids which burrow in the ground under logs, as in Taeniocerus spp. (Kon & Araya 1992, Kon & Johki 1987), in leaf-cutter ant detritus as in Ptichopus angulatus (Percheron) or in passalids in other detritus-like habitats (Johki & Kon 1987). Almost nothing is known concerning the habits of O. floridanus. O. floridanus is known only from the southern terminus of the Lake Wales Ridge of Highlands Co., Florida. A high concentration of endemic Florida scrub biota is recognized from this area (Deyrup 1989, 1990). The history of Odontotaenius in North America may be proposed as follows: at some point during the Cenozoic when mesic forest (Quercus, Liquidambar, Acer, etc.), similar to that of the southeastern U.S. at present, extended relatively unbroken as far as Honduras, the ancestor of the U.S. species migrated into the U.S. from eastern Mexico. Subsequently, a dry barrier formed in southern Texas and Tamaulipas, isolating the U.S. populations. At various times since the Miocene, the ridges of Florida, especially the Lake Wales ridge, may have been isolated from the rest of the mainland by marine transgressions (Deyrup 1989). The ridges, then islands in the Florida Archipelago, may have given rise to various endemic species, including O. floridanus; however, if one considers the fact that many endemic species of the Florida ridges appear to be relict species which had wider distributions at one time (Deyrup 1990), O. floridanus may have originated elsewhere and migrated into Florida, only now being restricted in its range. Of particular interest now is whether O. floridanus is restricted to the Lake Wales ridge (as are various other taxa (Deyrup 1989)), or whether it or other endemic passalid species occur on other Florida ridges. Although its wings appear normal, O. floridanus has not been found in other than Highlands Co., despite the fact that I have examined hundreds of passalid specimens from Florida. O. disjunctus is found throughout much of Florida north of Lake Okeechobee (Schuster 1983), including the Archbold Biological Station, where it is sympatric with O. floridanus. Other interesting questions concern the degree of ecological overlap between these two Florida species. Does habitat sympatry occur, i.e., do they occur in the same forest types? I suspect this may be the case, considering that O. disjunctus inhabits a wide variety of broadleaf forests in North America, including the relatively dry turkey-oak sandhills of north central Florida (Schuster 1978). Does microhabitat sympatry occur, i.e., do they occur in the same kind (species, degree of decomposition) of logs, or even the same log? This would not be surprising, considering that Luederwaldt (1931) found 10 species in a single log in Brazil, and frequently three or four species are found in the same log in the tropics. Further collecting in Florida should answer some of these questions. The following key is based on that of Castillo et al. (1988): KEY TO THE SPECIES OF ODONTOTAENIUS 1. Frontal fossae glabrous, clypeus swollen or with triangular projection in middle Frontal fossae pubescent, clypeus uniform width throughout or narrower in middle... 2

86 Schuster: Odontotaenius floridanus, A New U.S. Passalid Metasternal disc delimited by punctations, eyes reduced; Mexico, Sierra Madre Oriental...O. zodiacus 2 Metasternal disc not delimited by punctations, eyes normal; Mexico, Jalisco, Sierra de Manantlán... O. cerastes 3. Clypeus with triangular projection in middle, body length <35mm; Mexico to Colombia...O. striatopunctatus 3 Clypeus gently swollen in middle Body length >35mm; U.S.A., Canada Body length 25-26mm; Ecuador...O. striatulus 5. Prothoracic tibiae narrow, horn pedunculate (Figs. 1,2); eastern U.S.A., southeastern Canada...O. disjunctus 5 Prothoracic tibiae wide, horn extends forward without marked peduncle (Figs. 1,2); south-central Florida... O. floridanus ACKNOWLEDGMENTS Special thanks to Michael Thomas and Mark Deyrup for facilitating the expedition to Highlands Co., Paul Skelley for finding the beasts with larvae in the field, and Mark Deyrup and the F.S.C.A. for providing other specimens. Mark Deyrup, Pedro Reyes-Castillo, Gary Steck and an anonymous reviewer provided cogent criticism of the manuscript. The Universidad del Valle de Guatemala provided support. REFERENCES CITED CASTILLO, C., L.E. RIVERA-CERVANTES, AND P. REYES-CASTILLO Estudio sobre los Passalidae (Coleoptera: Lamellicornia) de la Sierra de Manantlán, Jalisco. Acta Zool. Mexicana (n.s.) 30: DEYRUP, M Arthropods endemic to Florida scrub. Florida Scientist 52(4): DEYRUP, M Arthropod footprints in the sands of time. Florida Entomol. 73: JOHKI, Y. AND, M. KON Morpho-ecological analysis on the relationship between habitat and body shape in adult passalid beetles (Coleoptera: Passalidae). Mem. Fac. Sci., Kyoto Univ., (Ser. Biol.), 2: KON, M., AND K. ARAYA On the microhabitat of the Bornean passalid beetle, Taeniocerus platypus (Coleoptera, Passalidae). Elytra, Tokyo, 20(1): KON, M., AND Y. JOHKI A new type of microhabitat, the interface between the log and the ground, observed in the passalid beetle of Borneo Taeniocerus bicanthatus (Coleoptera: Passalidae). J. Ethology 5(2): LUEDERWALDT, H Monographia dos passalideos do Brasil (Col.). Rev. Mus. Paul., 17 (1st parte). REYES-CASTILLO, P Coleoptera, Passalidae: Morfología y división en grandes grupos; géneros americanos. Folia Entomol. Mexicana 20-22: SCHUSTER, J Biogeographical and ecological limits of New World Passalidae (Coleoptera). Coleopterists Bulletin 32(1): SCHUSTER, J The Passalidae of the United States. Coleopterists Bulletin 37(4): SCHUSTER, J., AND P. REYES-CASTILLO New World genera of Passalidae (Coleoptera): a revision of larvae. An. Esc. nac. Cienc. Bio., Mexico. 25:

87 480 Florida Entomologist 77(4) December, 1994 FEEDING BY BAGOUS AFFINIS (COLEOPTERA: CURCULIONIDAE) INHIBITS GERMINATION OF HYDRILLA TUBERS K.E. GODFREY AND L.W.J. ANDERSON USDA, ARS, Aquatic Weed Control Research Laboratory University of California Davis, CA ABSTRACT Bagous affinis Hustache (Coleoptera: Curculionidae) larvae feed inside subterranean turions or tubers of hydrilla (Hydrilla verticillata (L.f.) Royle, Hydrocharitaceae) during low water conditions. This results in reduced germination of the tubers. To determine the number of B. affinis required to reduce tuber germination, dioecious hydrilla tubers were exposed to various B. affinis egg to tuber ratios. The tubers were then held for germination. The number of adults produced and the number of tubers germinating for each treatment and damage category were recorded. In all treatments, tuber germination was significantly reduced compared with the controls. The proportion of tubers germinating tended to decrease with an increase in the number of eggs initially placed in the treatment. This reduction in germination resulted from an increase in feeding damage. The results of this study suggest that B. affinis should be released in the field with an egg to tuber ratio of 2:1 or greater. Key Words: Biological control, aquatic weed control, hydrilla tuber weevil, insect feeding damage RESUMEN Las larvas de Bagous affinis Hustache (Coleoptera: Curculionidae) se alimentan de los tallos subterráneos (tubérculos) de la elodea de la Florida (Hydrilla verticillata [L.f.] Royle, Hydrocharitaceae) cuando el agua es poco profunda, lo que reduce su germinación. Para determinar el número de B. affinis requerido para reducir la germinación de los tallos subterráneos de la elodea, fueron expuestos tubérculos dióicos a varias densidades de huevos del insecto y se esperó a que germinaran. El número de adultos producido, el número de tubérculos que germinaron y la categoría de los daños fueron registrados en cada tratamiento. En todas las variantes la germinación de los tubérculos fue significativamente reducida con respecto a los testigos. La proporción de los tubérculos germinados tendió a disminuir con el aumento del número de huevos inicialmente colocados en cada tratamiento. Esta reducción de la germinación fue el resultado del aumento del daño producido por los insectos al alimentarse de los tallos. Los resultados de este estudio sugieren que B. affinis debe liberarse en el campo a una proporción de huevos por tubérculo de 2:1 o mayor. Bagous affinis Hustache (Coleoptera: Curculionidae), the hydrilla tuber weevil, is a biological control agent for hydrilla (Hydrilla verticillata (L.f.) Royle; Hydrocharitaceae), a submersed aquatic weed. The life cycle of this weevil is geared to a wet-dry seasonal climate. In the dry season, the weevils feed upon the above-ground portions of the hydrilla plant that are exposed as water recedes from an aquatic site (Baloch et al. 1980, Buckingham 1988). Female weevils oviposit in moist organic matter found This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

88 Godfrey & Anderson: B. affinis & Hydrilla Tubers 481 in and among the stranded hydrilla plants (Baloch et al. 1980, Buckingham 1988). Upon egg hatch, the larvae burrow through the soil seeking subterranean turions (tubers) of hydrilla. The larvae complete three instars while feeding inside the tubers and then pupate either within the tuber or in the soil (Bennett & Buckingham 1991). The feeding activity of the larvae destroys the tubers by consuming the meristems, or by providing an entryway for other organisms such as fungi or bacteria. Destruction of populations of hydrilla tubers, known as tuber banks, is important in controlling hydrilla because the tubers are a source of new infestations for up to 4 years after formation of the tubers (Van & Steward 1990, L. W. J. A., unpublished data). Hydrilla is classified as a Category A pest in California and, as such, must be managed with eradication as the objective. In a cooperative program with the California Department of Food and Agriculture, we investigated the use of the hydrilla tuber weevil in an inundative release program to reduce and possibly eliminate tuber banks at selected sites in California. The hydrilla tuber weevil was selected for use in this program because in its native range it infested almost 100% of hydrilla tubers at a site during the dry season. In the following wet season, there was little or no regrowth of the hydrilla at this site (Baloch et al. 1980). In California, some of the water systems infested with hydrilla undergo a seasonal drawdown, thereby potentially exposing hydrilla tubers to attack by B. affinis. To estimate the number of weevils to be released at an infested site, the number of weevils and the amount of feeding damage required to cause a reduction in germination of a population of tubers must be determined. In this study, the relationship between B. affinis density and reduction in tuber germination was investigated by measuring the amount of germination by dioecious hydrilla tubers after exposure to different numbers of B. affinis larvae. This study was conducted in the laboratory because the Category A pest designation of hydrilla would not allow the establishment of field plots. MATERIALS AND METHODS The ratio of B. affinis to hydrilla tubers required to reduce tuber germination was investigated in experiments that were conducted at the USDA Aquatic Weed Control Research Laboratory, Davis, California, from 16 December 1992 to 20 May The insects used in these experiments had been in laboratory culture for 8 to 10 generations. The weevils used to originate this colony were collected outside Bangalore, India in April They were cultured in quarantine at the Florida Biological Control Laboratory, Gainesville, Florida, for 1 generation before shipment to California in the summer of The dioecious hydrilla tubers were obtained from the USDA Aquatic Plant Management Laboratory, Ft. Lauderdale, Florida. Known numbers of hydrilla tubers were exposed to different numbers of B. affinis larvae. Eggs were used to initiate the experiments because placing eggs on the soil surface more closely reflects actual field conditions in which adults are released and allowed to oviposit. Eggs are also more amenable to transfer to experimental containers than neonate larvae. Of the eggs used in these experiments, approximately 90% hatched (K. E. G., unpublished data). Fifty replicates of each of the following egg to tuber ratio treatments were established: 1:5 (2 eggs: 10 tubers), 1:4 (2 eggs: 8 tubers), 1:2 (2 eggs: 4 tubers), 1:1 (2 eggs: 2 tubers), 2:1 (4 eggs: 2 tubers), and 5:1 (10 eggs: 2 tubers). These treatments represent the following tuber densities: 10 tubers, 3,306 per m 2 ; 8 tubers, 2,645 per m 2 ; 4 tubers, 1,323 per m 2 ; and 2 tubers, 662 per m 2. The treatments describe the initial experimental conditions. Each replicate consisted of a plastic rearing container (5.5 x 5.5 x 6 cm) filled with a sandy loam soil that had been moistened with a 1% benomyl

89 482 Florida Entomologist 77(4) December, 1994 solution until damp, but friable. The benomyl solution was used to prevent the growth of fungi (Bennett & Buckingham 1991). Hydrilla tubers were weighed individually, and the required number buried approximately 3 cm below the soil surface. B. affinis eggs were dissected from water-soaked wood (an oviposition media) that had been placed in a colony cage for h. The appropriate number of eggs was placed on moist filter paper on the soil surface, and the container was covered with foil to maintain soil moisture. Controls were set up exactly like the experimental containers, except that no eggs were included. Thirty-five replicates were set up as controls for each of the four tuber densities in the six treatment ratios (i.e., tuber density of 10 for the 1:5 ratio; 8 for the 1:4 ratio; 4 for the 1:2 ratio; and 2 for the 1:1, 2:1, and 5:1 ratios). All containers, both treatments and controls, were held at 27 C for 25 days. The containers were misted 3 times per week with tap water to maintain soil moisture. To determine germination of the tubers, all B. affinis and tubers were recovered and counted. The tubers were then broken in half medially. The interior of each tuber was examined and scored according to the following feeding damage scheme: 0, 1-25, 25-50, 50-75, and % of the interior damaged. The tubers were then grouped according to treatment, replicate, and feeding damage category, and placed in petri dishes (9 cm diam). The tubers were covered with tap water and placed at 27 C with a photoperiod of 16:8 (L:D) for 7 days. Under these conditions, any non-dormant tubers capable of germinating should have germinated (Spencer & Anderson 1986). The effect of breaking the tubers in half medially on germination was investigated by examining the germination of 100 tubers, 50 broken, and 50 left entire. The tubers were placed in petri dishes (9 cm diam), covered with tap water, and held for 7 days at 27 C with a photoperiod of 16:8 (L:D). The number of tubers germinating was recorded. Comparisons of the proportion of tubers in each feeding damage category among ratio treatments were done using Χ 2 analysis (Steel & Torrie 1960). The effect of tuber size on the amount of feeding damage was investigated by assigning tubers to one of five size classes ( gm, gm, gm, gm, or gm) and comparing the proportion of tubers in each feeding damage category among size classes. This comparison was done using Χ 2 analysis (Steel & Torrie 1960). The proportion of tubers germinating among ratio treatments, between ratio treatments and controls, among feeding damage categories, and between broken and entire tubers were compared using Χ 2 analysis (Steel & Torrie 1960). RESULTS The number of B. affinis adults produced increased with an increase in the egg to tuber ratio treatment (Table 1). The treatments were set up with differing numbers of eggs, so the proportion of adults produced were compared among treatments. Significantly lower proportions of adults were produced at the 5:1, 2:1, and 1:1 treatment ratios than at the 1:5, 1:4, and 1:2 treatments (Table 1; Χ 2 =39.02, df=5, P<0.05). This lower production of adults may be due to greater intraspecific competition among the larvae. Such competition could result in greater mortality of the larvae in the higher treatments as compared with the lower treatment ratios, even though the larvae are not cannibalistic (Bennett & Buckingham 1991). The proportion of tubers fed upon increased with an increase in the egg to tuber treatment ratio (Fig. 1A; Χ 2 = 91.1, df=5, P<0.01). The proportion of tubers damaged was found to be independent of the weight of the tuber (Table 1; Χ 2 =6.1, df=4, P>0.10), suggesting that the increase in damage was the result of an increase in the number of larvae present. The proportions of tubers withinthe feeding damage categories dif-

90 Godfrey & Anderson: B. affinis & Hydrilla Tubers 483 TABLE 1. THE MEAN WEIGHT ± STD. ERR. OF TUBERS, THE MEAN NUMBER ± STD. ERR. OF B. AFFINIS ADULTS PRODUCED AND THE PROPORTION OF EGGS SURVIVING TO THE ADULT STAGE IN EACH RATIO TREATMENT. Ratio Treatments Mean Wt. of Tubers (gm) Mean No. of B. affinis Produced Proportion of Eggs Surviving To Adult 1: ± ± : ± ± : ± ± : ± ± : ± ± : ± ± fered significantly among the egg to tuber treatment ratios (Fig. 1B; Χ 2 =304.6, df=20, P<0.01). At the low treatments (1:5), more of the tubers were in the no or low (0%, 1-25%) feeding damage categories, whereas, at the higher treatments, more tubers were found in the higher feeding damage categories (50-75%, %; Fig. 1B). Germination of the tubers was not influenced by breaking the tubers in half medially (Χ 2 =2.38, df=1, P>0.10). Of the tubers that were broken in half, 64% (n = 50) germinated. Of the tubers left entire, 78% (n = 50) germinated. Comparisons of the proportion of tubers germinating in the treatments with those in the controls summed over all feeding damage categories revealed significant differences (Fig. 2A; 1:5: Χ 2 =45.39, df=1, P<0.01; 1:4: Χ 2 =15.76, df=1, P<0.01; 1:2: Χ 2 =9.91,df=1, P<0.01; 1:1: Χ 2 =14.69, df=1, P<0.01; 2:1: Χ 2 =14.69, df=1, P<0.01; 5:1: Χ 2 =14.96, df=1, P<0.01). In all egg to tuber treatment ratios, except the 1:2 treatment, the proportion of tubers germinating was less in the treatments than in the controls (Fig. 2A). This demonstrated the ability of B. affinis to reduce tuber germination. In the 1:2 egg to tuber treatment ratio, a greater proportion of tubers germinated in the treatment than in the control (Fig. 2A). The reason for this difference is unclear. However, in this treatment, the proportion of tubers germinating in all feeding damage categories was greater than in other treatments (Fig. 2B). In general, there was a reduction in tuber germinationwith an increase in the density of B. affinis and the amount of feeding damage (Figs. 2A, 2B, and 3). Comparison of the proportion of tubers germinating among egg to tuber treatment ratios without regard to feeding damage category, revealed a significant decrease in germination as the treatment ratio increased (Fig. 2A; Χ 2 =71.0, df=5, P<0.05). The proportion of tubers germinating in the controls and in each feeding damage category, regardless of ratio treatment, decreased significantly with an increase in damage category (Fig. 3; Χ 2 =101.58, df=5, P<0.01). The proportion of tubers germinating decreased substantially for those tubers in the and 50-75% feeding damage categories. No tubers germinated in the % feeding damage category (Fig. 3). DISCUSSION The results suggest that for B. affinis to decimate hydrilla tuber banks, they should be released with an egg to tuber ratio of 2:1 or greater. The objectives of the release should dictate the ratio used. For example, if B. affinis was used in an inoculative release program where establishment of the weevil was the objective, the egg to tuber ratio for release should be 1:1 or 2:1. These lower ratios should be used because

91 484 Florida Entomologist 77(4) December, 1994 Fig. 1. A.) The proportion of tubers that had been fed upon or not fed upon for the 1:5 (n = 500 tubers), 1:4 (n = 399 tubers), 1:2 (n = 200 tubers), 1:1 (n = 99 tubers), 2:1 (n = 99 tubers), and 5:1 (n = 100 tubers) egg to tuber ratio treatments. Please note 1 tuber was unaccounted for in the 1:4, 1:1, and 2:1 treatments. The proportion of tubers fed upon increased significantly (P < 0.01) with an increase in the ratio treatment. B.) The proportion of tubers in each feeding damage category in which feeding damage occurred for each treatment. The proportion of tubers within feeding damage categories differed significantly (P < 0.01) among the ratio treatments.

92 Godfrey & Anderson: B. affinis & Hydrilla Tubers 485 Fig. 2. A.) The proportion of tubers germinating in the controls and in each ratio treatment. There was a significant decrease (P < 0.05) in germination with an increase in ratio treatment. Within each ratio treatment, the proportion of tubers germinating differed from that in the controls (P < 0.01). (See text for Χ 2 values). B.) The proportion of tubers germinating in each feeding damage category for each treatment.

93 486 Florida Entomologist 77(4) December, 1994 Fig. 3. The proportion of tubers germinating in each feeding damage category summed over the controls and all egg to tuber ratio treatments. There was a significant (P < 0.01) decrease in germination with an increase in feeding damage category. they resulted in proportionally more adults being produced from the eggs than the 5:1 egg to tuber ratio. However, if B. affinis was used in an inundative release program where the objective was maximum tuber destruction, then the egg to tuber ratio for release should be 5:1 or greater. The higher egg to tuber ratio should be used because production of adult B. affinis would not be a priority. The egg stage of B. affinis may not be the most convenient life stage for release in the field. Conversion of the number of eggs to the number of adults requires knowledge of the mean fecundity, the sex ratio of a population of weevils, and the percent egg eclosion. For B. affinis in the laboratory, the mean fecundity is eggs per female (Bennett & Buckingham 1991), the sex ratio is approximately 1:1 (Bennett & Buckingham 1991), and approximately 90% of all eggs hatch (K. E. G., unpublished data). To achieve a 2:1 egg to tuber ratio at a site would require 1 weevil for every 52 tubers, assuming that the life history attributes for B. affinis given above are representative of those in the field. For the 5:1 egg to tuber ratio, 1 weevil would be required for every 21 tubers. In hydrilla-infested aquatic sites in Florida and California, tuber densities ranged from and 20-1,000 tubers per m 2, respectively (Bowes et al. 1979, Anderson & Dechoretz 1982, Sutton & Portier 1985). Reduction of the tuber banks in infested sites in Florida using B. affinis would have required the release of between weevils per m 2 to achieve the 2:1 egg to tuber ratio, and between weevils per m 2 for the 5:1 ratio. In California, between 1-20 weevils per m 2 would have to be released for the 2:1 ratio, and between 1-48 weevils per m 2 for the 5:1 ratio. In practice, the number of weevils released should probably be greater than those given above because the weevils may not be as successful in the field as they are in the

94 Godfrey & Anderson: B. affinis & Hydrilla Tubers 487 laboratory. In two other studies where B. affinis was released in the field, the percent of tubers attacked was not as great as that in the laboratory. In Florida, B. affinis was released at an egg to tuber ratio of about 1:5. In the tubers recovered from these sites, % had been fed upon (Buckingham et al. 1994). In California, B. affinis was released at an egg to tuber ratio of approximately 1.2:1, and 11.2% of the sentinel tubers (tubers that were placed in the field to monitor the success of a release) were fed upon (Godfrey et al. 1994). In this laboratory study, 37.2 and 52.5% of the tubers had been fed upon in the 1:5 and 1:1 ratios, respectively. The lower rate of larval attack in the field may have been due to a variety of factors such as soil temperature, soil texture, or movement by the adults before oviposition (Buckingham et al. 1994). The ratios of weevils to tubers required for maximum tuber destruction determined in this study should be viewed as guidelines for release numbers. Many factors influence the ability of B. affinis to destroy tubers. However, the results of this study suggest that under favorable conditions, B. affinis has the ability to impact hydrilla tuber banks. ACKNOWLEDGMENTS We acknowledge K. Steward for supplying many of the hydrilla tubers used in this study and D. Davis for her technical assistance. We thank S. Sheldon, F. Ryan, and L. Godfrey for reviewing an earlier draft of this manuscript. This research was supported with a grant from California Department of Food and Agriculture. Mention of a proprietary product does not constitute an endorsement or a recommendation for its use by USDA. REFERENCES CITED ANDERSON, L.W.J., AND N. DECHORETZ Growth, reproduction and control of Hydrilla verticillata (L.f.) Royle in an irrigation system in the southwestern U.S. Proc. EWRS 6th Symposium on Aquat. Weeds. pp BALOCH, G.M., SANA-ULLAH, AND M.A. GHANI Some promising insects for the biological control of Hydrilla verticillata in Pakistan. Trop. Pest Manage. 26: BENNETT, C.A., AND G.R. BUCKINGHAM Laboratory biologies of Bagous affinis and B. laevigatus (Coleoptera: Curculionidae) attacking tubers of Hydrilla verticillata (Hydrocharitaceae). Ann. Entomol. Soc. America 84: BOWES, G., A.S. HOLADAY, AND W.T. HALLER Seasonal variation in the biomass, tuber density, and photosynthetic metabolism of hydrilla in three Florida lakes. J. Aquat. Plant Manage. 17: BUCKINGHAM, G.R Reunion in Florida - hydrilla, a weevil, and a fly. Aquatics 10: BUCKINGHAM, G.R., C.A. BENNETT, AND E.A. OKRAH Temporary establishment of the hydrilla tuber weevil (Bagous affinis) during a drawdown in north-central Florida. J. Aquat. Plant Manage. 31: (in press). GODFREY, K.E., L.W.J. ANDERSON, S.D. PERRY, AND N. DECHORETZ Overwintering and establishment potential of Bagous affinis (Coleoptera: Curculionidae) on Hydrilla verticillata (Hydrocharitaceae) in northern California. Florida Entomol. 77: SPENCER, D.F., AND L.W.J. ANDERSON Photoperiod responses in monoecious and dioecious Hydrilla verticillata. Weed Sci. 34: STEEL, R.G.D., AND J.H. TORRIE Principles and procedures of statistics. McGraw-Hill Book Company, Inc. New York. 481pp.

95 488 Florida Entomologist 77(4) December, 1994 SUTTON, D.L., AND K.M. PORTIER Density of tubers and turions of Hydrilla in South Florida. J. Aquat. Plant Manage. 23: VAN, T.K., AND K.K. STEWARD Longevity of monoecious hydrilla propagules. J. Aquat. Plant Manage. 28:

96 488 Florida Entomologist 77(4) December, 1994 THREE NEW SPECIES OF RHYPAROCHROMINE LYGAEIDAE (HEMIPTERA: HETEROPTERA) FROM HISPANIOLA JAMES A. SLATER 1 AND RICHARD M. BARANOWSKI 2 1 Dept. Ecology & Evolutionary Biology University of Connecticut Storrs, CT University of Florida Institute of Food and Agricultural Sciences Tropical Research and Education Center Homestead, FL ABSTRACT Three new species of rhyparochromine Lygaeidae, of the tribe Myodochini, Heraeus caliginosus New Species, Heraeus concolor New Species, and Catenes spiculus New Species from Hispaniola are described. H. caliginosus and C. spiculus are figured. Catenes Distant has previously been known only from a single Central American species. Key Words: West Indies, Myodochini, Heraeus, Catenes RESUMEN Se describen tres nuevas especies de Lygaeidae rhyparochromine, de la tribu Myodochini, Heraeus caliginosus Nueva Especie, Heraeus concolor Nueva Especie y Catenes spiculus Nueva Especie colectados en Española. Se proveen figuras de H. caliginosus y de C. spiculus. El género Catenes Distant se conocía previamente en base a una sola especie Centroamericana. During the course of our ongoing study of the lygaeid fauna of the West Indies, we have had occasion to study several interesting specimens taken in light traps in the Dominican Republic. Several of these specimens represent undescribed species of the tribe Myodochini that are treated below. The genus Heraeus Stal is a complex taxon with four species previously known from the West Indies (Slater 1964). Catenes Distant, however, has been known previously only from a single Central American species, Catenes porrectus Distant (Distant 1893). This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

97 Slater & Baranowski: New Species of Lygaeidae (Hemiptera) 489 One of the striking features of these new species is that all of them show a contrastingly dark and light color pattern. This is also true for a species of Ozophora (tribe Ozophorini) which will be discussed in a later paper. This convergence of color pattern in what are otherwise unrelated taxa must be an adaptation to a substrate which will only be clarified when material can be taken in situ in the field. All measurements are in millimeters. Catenes spiculus Slater and Baranowski, New Species (Fig. 1) DESCRIPTION: Elongate, nearly parallel-sided. Head black, shining. Anterior pronotal lobe dark chestnut brown with a black median stripe running through entire pronotum and scutellum. Anterior pronotal collar reddish brown, concolorous with posterior pronotal lobe. Posterior pronotal lobe reddish brown on either side of black midline, becoming darker reddish brown laterally. Scutellum dark chocolate brown with a median black stripe, apex white. Hemelytra nearly uniformly testaceous with punctures strongly constrastingly dark brown; lateral corial margins pale yellow; corium posteriorly with an apical dark brown macula and a second macula laterally not reaching lateral margin of corium, located at level of middle of apical corial margin. Membrane fumose with veins contrastingly white. Thoracic pleura and sterna dark chocolate brown, almost black. Abdomen bright reddish brown. Legs white, or pale yellow, with distal third of meso- and metafemora and distal half of forefemora strongly contrastingly dark brown. Labium pale yellow. Antennae with first segment chocolate brown, second segment pale yellow with extreme distal end infuscated, third segment pale yellow on proximal two thirds, but with an extensively developed dark brown distal third, fourth segment with base and apical third dark chocolate brown, remainder of segment white. Dorsal surface clothed with upstanding hairs. Pronotum and scutellum dull, strongly contrasting with shining surface of head. Head extremely elongate and tapering, apex exceeding distal end of first antennal segment. Eyes sessile, set midway along lateral margins of head, remote from anterior pronotal margin. Length head 1.36, width 0.92, interocular space Pronotum with a distinct anterior collar, anterior lobe much narrower than posterior, transverse impression complete. Length pronotum 1.30, width Scutellum lacking a prominent median carina, although mesal area somewhat elevated. Length scutellum 1.02, width Hemelytra with corium nearly parallel sided, slightly concave at level of apex of scutellum. Length claval commissure Midline distance apex clavus to apex corium Midline distance apex corium to apex abdomen Metathoracic scent gland auricle directed slightly postero-laterad, short, subelliptical. Evaporative area large, occupying most of anterior lobe of metapleuron, narrowing posteriorly and extending anteriorly onto posterior area of mesopleuron. Mesepimeron emergent. Forefemora relatively slender, slightly incrassate, armed below distally with a simple series of spines as follows: two major spines with three minor spines between them, a proximally placed hair spine and a single small distal spine. Labium elongate, extending onto second abdominal sternum. First labial segment extending caudad beyond posterior margin of eye but not reaching base of head; second segment reaching between forecoxae; third segment reaching or slightly exceeding metacoxae. Length labial segments I 1.04, II 1.24, III 1.24, IV Antennae slender, fourth segment narrowly fusiform. Length antennal segments I 0.56, II 1.40, III 1.20, IV Total body length HOLOTYPE. Female. Dominican Republic: Guanumo, Finca Goya, 30.V.1989 (Gustavo Anzerro) (blacklight trap). In National Museum of Natural History (NMNH).

98 490 Florida Entomologist 77(4) December, 1994 Figure 1. Catenes spiculus Slater and Baranowski New Species. ETYMOLOGY. Referring to the sharp needle-like anterior end of the head. Distant s (1893) original description of Catenes porrectus, the only previously known species in the genus, is very short. It was described from Guatemala and Panama and is known only from these locations. Distant s (1893) color plate shows C. porrectus differing from C. spiculus n. sp. in having a pale yellow first antennal segment, the forefemora yellow with dark dots over the entire surface, the darkened distal third of the meso- and metafemora interrupted by a pale yellow annulus, dark distal ends

99 Slater & Baranowski: New Species of Lygaeidae (Hemiptera) 491 to all tibiae, a reddish brown head and scutellum and lacking a subapical dark macula along the costal margin of the corium. Catanes porrectus is said to be 8 mm long. We have examined two males from Venezuela (Miranda EST. EXT. Rio Negro cr. CAPAYA, 100 m VI.80) (in Universidad Central de Venezuela) which agree with the figure and description of C. porrectus (Distant 1893) in all respects, except that the meso- and metafemoral annuli are obscure, all femora have numerous black spots, and the hemelytra are completely and uniformly yellowish. Heraeus caliginosus Slater and Baranowski, New Species (Fig. 2) DESCRIPTION. Coloration chiefly black to extremely dark brown. Head black, strongly shining. Pronotum and scutellum dull black, with two small yellow macula on either side of midline on posterior lobe immediately behind transverse impression. Scutellum gray on anterior half, with a narrow black median stripe and an oblique black macula midway between meson and lateral margin on each basal half. Hemelytra in large part dark chocolate brown. Clavus yellow on anterior two-thirds with contrasting dark brown punctures, extreme base and distal third dark brown. Corium yellowish basally, area at level of claval commissure interspersed with yellowish and dark brown. A large rectangular white macula distally on corium at level of middle of apical corial margin, extending from costal margin nearly to apical corial margin, but not actually attaining latter. Membrane dark brown, veins in part pale yellow, a very conspicuous rectangular white bar mesally on apical third of membrane. Thoracic pleura and sterna uniformly dull black. Abdomen shining reddish brown, a quadrate yellow macula present along dorsal margin of sternum five. Entire forefemora and distal halves of middle and hind femora dark chocolate brown, strongly contrasting with white proximal halves of middle and hind femora. Tibiae and tarsi pale yellow, tibiae infuscated with brown at extreme proximal and distal ends. Labium pale yellowish brown. First antennal segment dark red-brown, second and third segments almost uniformly yellow, segment three becoming ochraceous distally, fourth segment dark brown with a short, narrow, inconspicuous, poorly differentiated, pale annulus on proximal third (figure exaggerates pale annulus). Dorsal surface with a few scattered upright hairs present, these more numerous and elongate on head and scutellum. Thoracic punctures weak, indistinct, those on clavus and corium larger and conspicuous. Head elongate, porrect. First antennal segment slightly exceeding apex of tylus. Eyes placed near middle of head, area behind eyes characteristically constricted. Length head 1.34, width 0.88, interocular space Pronotum with anterior collar narrow dorsally becoming strongly widened ventrally (typical for genus). A row of deep conspicuous punctures on depression behind posterior margin of anterior collar. Anterior pronotal lobe moderately convex but not elevated to level of posterior lobe, transverse impression complete and punctate. Length pronotum 1.0, width Length scutellum 0.92, width Lateral corial margins nearly parallel-sided, slightly concave at level of apex of scutellum. Length claval commissure Midline distance apex clavus to apex corium Midline distance apex corium to apex abdomen Metathoracic scent gland auricle short, slightly curving posteriorly. Evaporative area occupying inner two thirds of metapleuron. Forefemora strongly incrassate, armed below with at least four major spines, distally with a series of smaller spines. Labium elongate, exceeding metacoxae, attaining middle of abdominal sternum two (first visible segment), segment one almost reaching base of head, second segment extending onto anterior portion of mesosternum. Length labial seg-

100 492 Florida Entomologist 77(4) December, 1994 Figure 2. Heraeus caliginosus Slater and Baranowski New Species. ments I 1.08, II 1.20, III 0.60, IV 0.66 (approx.). Antennae conventionally terete, fourth segment narrowly fusiform. Length antennal segments I 0.58, II 1.16, III 1.10, IV Total body length TYPES. Holotype. Female. Dominican Republic: Pedernales Prov., 21 km. N. Cabo Rojo, 19.VI.1976 (R. E. Woodruff) (blacklight trap). In Florida State Collection of Arthropods. Paratype: 1 male. Dominican Republic: Barahona, 6 km. NW Paraiso, Rio Nizao, 18-02N, 17-12W, 170 m VII-1990 (C. Young, J. E. Rawlins, S. A. Thompson). In Carnegie Museum. ETYMOLOGY. Referring to the dark coloration of the body. Heraeus caliginosus n. sp. is most closely related to Heraeus guttatus (Dallas), the two species resembling one another closely in size and general body proportions. Heraeus caliginosus may readily be distinguished from H. guttatus by the elongate labium which extends well onto the second abdominal segment. In H. guttatus the labium extends, at most, only between the metacoxae. In H. caliginosus the first an-

101 Slater & Baranowski: New Species of Lygaeidae (Hemiptera) 493 tennal segment is chocolate brown and contrasts strongly with the pale yellow second segment; in H. guttatus both the first and second antennal segments are pale yellow and concolorous. Heraeus caliginosus has a quadrate white corial macula with the anterior and posterior margins parallel (somewhat irregular in the paratype) and at right angles to the body length; in H. guttatus the subdistal pale macula has irregular margins, the anterior margin being convexly, although irregularly, produced, whereas the posterior margin of the macula is concave and the entire macula is oriented slightly antero-mesad from the lateral margin. In H. caliginosus the apex of the membrane has a large conspicuous white parallel-sided patch. In H. guttatus the membrane is often pale at the end, but the pale area does not form a large evenly parallelsided patch. Heraeus guttatus has at most only a trace of a brown distal area on the mesofemora and a pale brown distal one third to one half of the metafemora, whereas H. caliginosus has the distal two-thirds of both the meso- and metafemora dark chocolate brown. Heraeus caliginosus is a very dark species predominately black in color, whereas H. guttatus is chiefly yellowish brown. Differences from the similarly colored H. concolor described below are discussed under the latter species. The other Antillean species are readily separable, H. pulchellus Barber and H. plebejus Stal are much smaller, pale testaceous to light tan species, and H. triguttatus (Guerin) is a strikingly-colored orange and black species. Heraeus concolor Slater and Baranowski, New Species DESCRIPTION. Head, pronotum, scutellum, forefemora, distal annulus on middle and hind femora, pleuron, sternum and abdomen black. Posterior pronotal lobe with humeral angles and four small spots adjacent to transverse impression yellowish. Scutellum with a darker narrow median stripe. Hemelytra variegated: clavus chiefly dark brown, but with anterior half of cubital vein and a small macula mesad of vein near base testaceous; corium with a small dark brown spot near base, a complete broad, dark, transverse fascia completely across corium on distal third, and a dark apex beyond the prominent white subapical pale spot; a small pale spot present near inner angle of corium. Membrane black or dark chocolate brown with a conspicuous white rectangular apical mesal patch and light brown veins. Second and third antennal segments sordid tan, distal end of 3rd segment dark brown as is fourth segment except for a conspicuous white subbasal annulus. First antennal segment and first labial segment darkened. Head shining, contrasting strongly with dull surface of pronotum and scutellum. Body sparsely clothed with scattered upright hairs and short decumbent silvery hairs. Head slightly declivent; neck short. Length head 0.94, width 0.84, interocular space Transverse pronotal impression deep, anterior lobe not swollen above level of posterior lobe. Length pronotum 0.98, width Scutellum lacking a median carina. Length scutellum 0.84, width Length claval commissure Midline distance apex clavus to apex corium Midline distance apex corium to apex abdomen Metathoracic scent gland auricle straight, short. Evaporative area covering at least inner half of metapleuron, its outer margin strongly rounded. Forefemora armed below with 3-4 large spines and several smaller ones. Labium extending posteriorly between mesocoxae. Length labial segments I 0.72, II 0.74, III 0.62, IV Antennae conventionally terete with hairs on segments 2 and 3 longer than diameter of segments. Length antennal segments I 0.38, II 0.80, III 0.64, IV Total body length HOLOTYPE. Female. Dominican Republic: Bayaguana. 4.IX.1991 (D. Brown) (blacklight trap). In National Museum of Natural History (NMNH).

102 The flow from this article and the following article are the same but the columns have b een disconnected. Be careful not to cover text up with the column. 494 Florida Entomologist 77(4) December, 1994 ETYMOLOGY. Referring to a similarity in coloring (to H. caliginosus). It is unfortunate that we have only a single female of this predominately black species. The color patterns of H. caliginosus and H. concolor are remarkably similar, both species having a dark head, pronotum and scutellum with a prominent pale annulus on the fourth antennal segment, a large black median stripe through the dark scutellum, a large rectangular white patch mesally at the apex of the hemelytral membrane, four small yellow spots on the posterior pronotal lobe immediately behind the transverse impression, and predominately dark femora. Heraeus concolor has pale humeral pronotal angles, whereas the humeral angles of H. caliginosus are completely dark. The subapical white corial macula of H. caliginosus is relatively block-like with both anterior and posterior margins complete, whereas in H. concolor the anterior margin of this subapical corial macula is deeply invaded by dark coloration so that it takes on a rather hook-like appearance. While the color differences discussed above could well be variable, the structural differences between these two species are greater than those between several other species of Heraeus. Heraeus caliginosus has a relatively much longer head with the eyes less produced outward from the surface of the head (Fig. 2) than does H. concolor. In H. caliginosus the head length is about 1/3 greater than the pronotal length ( ), whereas in H. concolor the pronotum is slightly longer than the length of the head ( ). Heraeus caliginosus has much longer, sweeping antennae. Antennal segments II and III are each longer than the pronotal length (length pronotum 1.0, length antennal segment II 1.16, III 1.10), whereas in H. concolor the length of the pronotum is appreciably greater than the length of either the second or third antennal segments (length pronotum 0.98, length antennal segment II 0.89, III 0.64). Heraeus caliginosus also has a much longer fourth antennal segment than does H. concolor. In H. caliginosus the length of the fourth antennal segment is more than 2 1/2 times (2.71) as great as the interocular distance, whereas in H. concolor the fourth antennal segment is less than twice as long as the interocular distance (1.95). The labium is relatively very long in H. caliginosus as noted above in the discussion of its relationship to H. guttatus. In H. concolor the labium is much shorter, extending only between the mesocoxae rather than reaching onto the abdominal sternum. The occurrence of two such similarly colored, yet structurally, different species on Hispaniola once again demonstrates the complex past history of speciation on this enigmatic island. ACKNOWLEDGMENTS We wish to extend our appreciation to Dr. R. E. Woodruff, Emeritus Entomologist (Florida Department of Agriculture and Consumer Services, Division of Plant Industry), Dr. Eduardo Osuna (Universidad Central de Venezuela) for the loan of material, Dr. Donovan Brown (Plantaciones Oscar de la Renta), Gustavo Anzerro (Finca Goya) for operating blacklight traps and Mr. Stephen Thurston for providing the illustrations. Florida Agricultural Experiment Station Journal Series No. R REFERENCES CITED DISTANT, W.L Rhynchota. Hemiptera-Heteroptera. Vol. I. Biol. Centrali-Americana. London. Supp. pp SLATER, J.A A Catalogue of the Lygaeidae of the World. Vol. II. University of Connecticut, Storrs, Ct.

103 Slater &Baranowski: Oxycarenus hyalinipennis, West Indies 495 THE OCCURRENCE OF OXYCARENUS HYALINIPENNIS (COSTA) (HEMIPTERA: LYGAEIDAE) IN THE WEST INDIES AND NEW LYGAEIDAE RECORDS FOR THE TURKS AND CAICOS ISLANDS OF PROVIDENCIALES AND NORTH CAICOS JAMES A. SLATER 1 AND RICHARD M. BARANOWSKI 2 1 Dept. Ecology & Evolutionary Biology University of Connecticut Storrs, Ct University of Florida Institute of Food and Agricultural Sciences Tropical Research and Education Center Homestead, Florida ABSTRACT A breeding population of Oxycarenus hyalinipennis (Costa) is reported for the first time from the West Indies. Its distribution is discussed and information to distinguish it from other closely related species is given. Several additional lygaeids are reported for the first time from the islands of North Caicos and Providenciales. Key Words: Cotton, introduced, West Indies, Caicos Islands. RESUMEN Se reporta por primera vez en las Indias Occidentales una población reproductiva de Oxycarenus hyalinipennis. Se discute su distribución y se brinda información con el fin de distinguir esta especie de otras especies relacionadas. Se reportan por primera vez varias especies adicionales de lygaeidos en las islas de Caicos del Norte y Providenciales. The cotton seed bug, Oxycarenus hyalinipennis (Costa), is a common and widespread species in the Old World tropics. It extends from the European side of the Mediterranean throughout Africa where it has frequently been reported as injurious to cotton. Samy (1969) stated that it is a cosmopolitan species occurring in the Palearctic, Oriental and Neotropical regions. It is not entirely accurate to state that it is cosmopolitan as it is unknown in the Nearctic Region and, as will be discussed below, is introduced in the Neotropics. In Africa, it is an abundant widespread species and often causes staining of cotton. The intensive study by Kirkpatrick (1923) remains the definitive work. Kirkpatrick and other workers have listed a long series of plants upon which O. hyalinipennis has been found (see Slater 1964), but breeding records appear to be largely restricted to plants of the order Malvales. Samy (1969) summarized this literature asserting that probably the true host plants are confined to species of Malvaceae, Sterculiaceae and Tiliaceae. He mentioned the following genera: Abutilon, Cola, Eiodendron, Gossypium, Malva, Sphaeralcea, Hibiscus, Pavonia, Sida, Dombeya, Sterculia and Triumfetta. This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

104 496 Florida Entomologist 77(4) December, 1994 In the summer of 1991, the junior author and his wife collected on North Caicos five nymphs of a lygaeid under a species of Pluchea (Asteraceae) that they had not previously seen in the West Indies. Despite several hours of assiduous ground searching and sweeping in the same area on subsequent days, no additional specimens were taken. No malvaceous plants were evident in the immediate area. These specimens subsequently molted into adults and proved to be O. hyalinipennis. On another collecting trip to North Caicos, one adult of O. hyalinipennis was collected on wild cotton within 1 km of the previous find. More recently O. hyalinipennis was collected at Clarence Town, Long Island, Bahamas. These records represent the first known breeding populations of this potentially destructive insect in the West Indies. Although Henry et al. (1983) indicated that O. hyalinipennis had been intercepted at a U.S. airport on a citrus leaf in baggage from the Dominican Republic, plus a number of additional interceptions at U.S. ports of entry, the above record is the only one that could possibly have originated from a Western Hemisphere source. Oxycarenus hyalinipennis has been established in South America for many years. Although Samy (1969) claimed a cosmopolitan distribution, there is no doubt that this species is an introduced member of the Neotropical lygaeid fauna. The genus Oxycarenus is a large and diverse one in the Old World tropics (approximately 50 species of which at least 33 are Ethiopian) but, other than O. hyalinipennis, is not known to occur in the Western Hemisphere. Oxycarenus hyalinipennis was present in Brazil as early as 1917 (Costa Lima, 1922). Kormilev (1950) summarized the history of the species in the Western Hemisphere. Costa Lima (1940) stated that specimens were taken in northeastern Brazil in 1917 on lagarta rosa and that there were records of damage to cotton bolls. Slater (1964) listed published records for Argentina, Paraguay, and Bolivia, as well as Brazil. It is, therefore, not surprising that this common and mobile species should appear in the West Indies, but it is surprising that the first record should come from the small island of North Caicos. This suggests that although we have not collected it in our rather extensive surveys on many islands in the West Indies, it will eventually prove to be more widespread in the islands. There are several closely related species of Oxycarenus that can readily be confused with O. hyalinipennis and several of these feed on Malvaceous plants. To aid in accurate identification of this species, the following comments are included to enable others workers to readily separate it from closely related species that have black heads and pronota and pale colored hemelytra. Several species, such as O. albidipennis Stal, O. pallidipennis (Dallas) and O. congoensis Samy, may readily be recognized by the orange-red coloration of the first five abdominal segments. Samy (1969) described the species O. nigricornis, which he listed as widespread in Africa. He separated it from O. hyalinipennis because the antennae of this species were completely black whereas he believed O. hyalinipennis always had pale second antennal segments. Slater (1972) synonymized O. nigricornis with O. hyalinipennis stating that the antennal coloration of O. hyalinipennis is variable within populations and ranges from completely black through shades of dark brown to almost completely pale. Oxycarenus bokalae Samy is the species most likely confused with O. hyalinipennis. Both are similarly colored, but O. hyalinipennis has the clavus either completely or in large part pale testaceous to white, whereas in the clavus, O. bokalae is almost uniformly dark brown to black. The pygophore opening also differs in the two species. In O. bokalae the opening is broad with the side margins arcuate and triangularly tapering to a sharply or bluntly pointed distal end whereas in O. hyalinipennis the pygophore opening tapers evenly to a triangular point. Slater (1972) figured both conditions. It

105 Slater &Baranowski: Oxycarenus hyalinipennis, West Indies 497 should be noted that, despite its relatively recent description, O. bokalae is also widely distributed in Africa and has been reported from cotton (Samy, 1969). The following Lygaeidae are new records for the North Caicos and Providenciales Islands. North Caicos: Ochrimnus laevus Brailovsky. Paromius longulus (Dallas). Oedancala cladiumicola Baranowski & Slater. Craspeduchus pulchellus (F.) Ozophora umbrosa Slater (previously reported from West and South Caicos (Slater,1987)). Pseudopachybrachius vinctus (Say). Nysius raphanus Howard. Providenciales: Ochrimnus laevus Brailovsky. Oncopeltus fasciatus (Dallas). Ozophora divaricata Barber. Pseudopachybrachius basalis (Dallas). Craspeduchus pulchellus (F.) Lygaeus bahamensis Barber & Ashlock (previously reported by Barber and Ashlock (1960), from South and West Caicos). ACKNOWLEDGMENT Florida Agricultural Experiment Station Journal Series No. R REFERENCES CITED BARBER, H.G., AND P.D. ASHLOCK The Lygaeidae of the Van Voast-American Museum of Natural History expedition to the Bahama Islands Proc. Entomol. Washington 62: COSTA LIMA, A Char. e Quint 25: (2): COSTA LIMA, A Insectos do Brasil Hemipteros II. Rio de Janeiro: Escuela Nacional de Agronomia pp HENRY, T.J., AND BIOLOGICAL ASSESSMENT SUPPORT STAFF Pests not known to occur in the United States or of limited distribution. No. 38 Cottonseed Bug Hemiptera: Lygaeidae Oxycarenus hyalinipennis (Costa) USDA Animal and Plant Health Inspection Service, Plant Protection Quarantine APHIS pp. KIRKPATRICK, T.W The Egyptian cotton seed bug (Oxycarenus hyalinipennis (Costa). Its Bionomics, damage and suggestions for remedial measures. Bull. Minist. Agric. Egypt Techn. Scient. Serv. 35: 107 pp. KORMILEV, N La subfamilia Oxycareninae Stal en la Argentina con la description de una especie nueva (Hemiptera: Lygaeidae). Anales Soc. Cient. Argentina 149: SAMY, O A revision of the African species of Oxycarenus (Hemiptera: Lygaeidae). Trans. Royal Entomol. Soc. London 121S: Pt. 4: SLATER, J.A A Catalogue of the Lygaeidae of the World. Vol. I. U. Connecticut, Storrs, Ct. SLATER, J.A The Oxycareninae of South Africa (Hemiptera: Lygaeidae. U. Connecticut Occ. Papers Biol. Scien. Ser. 2 (7): SLATER, J.A A revision of the Ozophora umbrosa complex in the West Indies. J. New York Entomol. Soc. 95:

106 498 Florida Entomologist 77(4) December, 1994 WHITE EYE AND YELLOW LARVA: MUTANTS IN ANOPHELES STEPHENSI LISTON (DIPTERA: CULICIDAE) N. JAYA SHETTY 1,2, GAYATHRI PRABHAKAR 1, SUDHIR KARL NARANG 3, P. DAVID FOGLESONG 2, AND DENNIS J. JOSLYN 2 1 Centre for Applied Genetics, Bangalore University, J.B. Campus, Bangalore, , India 2 Department of Biology, Rutgers University, Camden, NJ United States Department of Agriculture, Agricultural Research Service, Biosciences Research Laboratory, P. O. Box 5674, State University Station, Fargo, ND ABSTRACT Two spontaneous mutants, white eye (w) and yellow larva (y), were isolated and characterized from the Bangalore (southern India) and Poona (western India) strains, respectively, of the malarial mosquito, Anopheles stephensi (Liston) (Diptera: Culicidae). The w mutation is a sex-linked recessive. The second mutant, y, is an autosomal recessive. Key Words: Genetic mutants, Anopheles stephenci, malaria. RESUMEN Fueron aislados y caracterizados dos mutantes espontáneos, ojo blanco (w) y larva amarilla (y), a partir de las cepas de Bangalore (sur de India) y Poona (oeste de India), respectivamente, del mosquito de la malaria, Anopheles stephensi (Liston). La mutación w es recesiva y ligada al sexo. El segundo mutante, y, es autosomal recesivo. Genetic studies of mosquitoes, especially of species and strains which are vectors, continue to be an essential component of genetic control strategies aimed at disrupting the transmission of diseases. These alternative strategies for control require genetic characterizations of geographically isolated strains because any one control mechanism must operate throughout the range of the target mosquito species. Therefore, genetic profiles of different geographical isolates can be used to predict the potential success of laboratory-altered strains intended for release into native populations of mosquito vectors. Our laboratories are currently performing genetic studies towards the eventual development of strains for genetic control of the vectors of infectious diseases. These studies include the following: genetic fingerprinting of laboratory and natural populations of malaria- and encephalitis-carrying mosquitoes; restriction mapping of mitochondrial genomes from diverse geographical isolates; characterization of disease-refractory strains; and the isolation and characterization of Mendelian mutants. This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

107 Shetty et al.: Anopheles stephensi Mutants 499 Considerable progress has been made on the genetics and cytogenetics of Anopheles stephensi Liston ( Diptera: Culicdae). These studies have been reviewed by Kitzmiller (1976), Narang & Seawright (1982), and Parvez et al. (1985). We report here two morphological mutants of the mosquito, An. stephensi, one of the most important vectors of malaria throughout India and in the Middle East. Both mutants are easily distinguished from the wild-type phenotype, and the viability of the mutants is as good as that of the wild-type. Therefore, they are excellent genetic markers and can be used for the design of strains for genetic control. The first mutant, white eye (w), from the Bangalore (southern India) strain of An. stephensi is sex-linked and recessive in females and hemizygous in males, as revealed by our crossing experiments. It may or may not be related to white eye reported for other strains of An. stephensi. The second mutant, yellow larva (y), is autosomal and recessive and comes from the Poona (western India) strain. White eye contrasts readily with the brownish-black wild-type eye color in this mosquito, and yellow larva is easily distinguished from the wild-type straw-tan color. MATERIALS AND METHODS Rearing Both the Bangalore (southern India) and the Poona (western India) strains of An. stephensi were utilized in this study. All mosquito stages were maintained at 26 ± 1 C and at a relative humidity of 75% ± 5% with a photoperiod of 14:10 (L:D). Adults were kept in 20 cubic-inch cages containing 10% sucrose. Females were bloodfed on mice, and all eggs were deposited hours after blood feeding in an enamel bowl with water and a lining of filter paper. Larvae were reared in 25 x 30 x 6 cm white enamel pans containing tap water and fed commercial bakers yeast. Isolation of Mutants The white eye (w) mutant appeared spontaneously in the Bangalore colony of An. stephensi. Its phenotype is an entirely white eye in larvae, pupae, and adults that could be distinguished easily from the wild-type with the naked eye. The yellow larva (y) mutant appeared spontaneously in the Poona colony of An. stephensi. Its phenotype is an entirely yellow larva except for black eyes. It is manifested in early third instars and persists throughout the fourth instar and pupal stage with the fourth instar showing the most conspicuous color. Yellow larva (y) is also easily distinguishable from the wild-type with the naked eye, and both mutants are as viable and easy to maintain as their wild-type counterparts. Design of Crosses Pure colonies of w and y were obtained by crossing mutant adults inter se. Several generations of such crosses involving 25 adults of each sex in 8 cubic-inch cages were necessary to obtain sufficient numbers of pure-breeding mutant mosquitoes. For the inheritance studies, mass matings were made between the pure-bred mutants and pure-bred wild-types. Portions of the F 1 males and females were backcrossed to stocks of the pure-bred mutants, and the F 2 backcross progeny were scored for phenotypes and ratios. The remaining F 1 males and females were inbred to produce the F 2 generation.

108 500 Florida Entomologist 77(4) December, 1994 RESULTS The mode of inheritance of white eye (w) and yellow larva (y) in An. stephensi was determined using classical Mendelian crosses and analyses. White eye was determined to be a sex-linked recessive in the homogametic female and hemizygous in the heterogametic male. Yellow larva was found to be an autosomal recessive trait with full penetrance and uniform expression in both sexes, but the specific autosome involved (2N = 6) remains undetermined. The data in Table 1 summarizes the crosses between the white-eyed strain and the wild-type strain. In cross 1 in which white-eyed males were crossed with wild-type females, the F 1 progeny consisted of all wild type individuals. In the reciprocal crosses (cross 2) in which white-eyed females were crossed with wild-type males, F 1 progeny consisted of wild-type females and males of the mutant phenotype. Backcrosses of the F 1 females from both crosses 1 and 2 mated with white males resulted in wild-type and white-eyed females and males in a 1:1:1:1 ratio (crosses 4 and 5). When F 1 males from cross 1 were mated to white females, the progeny consisted of wild-type females and white-eyed males (cross 3), whereas white-eyed females crossed to F 1 males from cross 2 resulted in all white-eyed progeny (cross 6). In both crosses F 1 progeny were inbred to obtain F 2 individuals. In the first cross among the F 2 progeny all females were wild-type while in males both wild-type and white-eyed phenotypes were found (cross 7). In the second cross among the F 2 progeny both wild type and white-eyed phenotypes were found in a 1:1:1:1 ratio (cross 8). The results of the crosses between yellow larva and wild type are given in Table 2 which indicates all appropriate Chi-square values supporting our interpretation of TABLE 1. MODE OF INHERITANCE OF MUTATION WHITE-EYE (w) IN ANOPHELES STEPHENSI. X, Y = X AND Y CHROMOSOMES, RESPECTIVELY; w = WHITE EYE; + = WILD TYPE. Presumptive Parental Genotypes Progeny Phenotypes Cross + w + w 1 X+ Xw X+ (Wild type) Y- (White) Xw X+ Xw (White) Y- (Wild type) Xw X+ Xw (White) Y- (Wild type) Xw Xw X+ (Wild type) Y- (White) Xw Xw X+ (Wild type) Y- (White) Xw Xw Xw (White) Y- (White) Xw X+ X+ (Wild type) Y- (Wild type) Xw Xw X+ (Wild type) Y- (White)

109 Shetty et al.: Anopheles stephensi Mutants 501 TABLE 2. MODE OF INHERITANCE OF MUTATION YELLOW LARVA (y) IN ANOPHELES STEPHENSI. *NOT SIGNIFICANT. Number of Larvae Crosses Wild Type Yellow χ 2 Cross No. Total Total wild type y y y y y y y + y + y y y + y + yellow yellow + + wild type yellow y y wild type y y wild type y y yellow + + wild type y + wild type y + wild type * yellow * yellow * wild type * wild type * wild type * the mode of inheritance of y as an autosomal recessive. None of the resulting F 1 mosquitoes in crosses 1 and 2 could be distinguished from the wild-type parents. The dominance of the wild-type was complete. The F 1 heterozygotes were then backcrossed with the mutants. The results of the crosses (3, 4, 5, and 6) fit the expected 1:1 ratio of wild-type to mutants. The F 1 adults were inbred to yield F 2 generations. These crosses (7 and 8) also yielded the expected 3:1 ratio of wild-type to mutants. DISCUSSION An. stephensi is an important vector which has developed resistance to insecticides. Therefore, it is mandatory that alternative strategies for its control be developed. Genetic control is one such strategy which requires basic genetic characterizations. We have recently reported several studies in these areas of our ongoing characterizations of An. stephensi (Gayathri & Shetty, 1989, 1992a, 1992b; Shetty & Gayathri, 1989; Bhaskar & Shetty, 1992; Rao & Shetty, 1992). The two mutants, w and y, described in this study represent excellent markers for the extension of these kinds of studies. Traditionally, such morphological mutants have been used to construct special genetic load strains containing chromosomal translocations or inversions (Pal & Whitten, 1974; Joslyn, 1980). In such strains, genetic markers indicate the presence of the chromosomal aberrations through either altered linkage relationships or position effects of genes located close to chromosomal breakpoints. Genetic markers such as w and y also can be used in monitoring the pro-

110 502 Florida Entomologist 77(4) December, 1994 duction and maintenance of genetic sexing systems (Curtis et al., 1976; Kaiser et al., 1978; Shetty, 1987; Baker et al., 1981; Weller & Foster, 1993). More fundamentally, such genetic markers are necessary for expanding the linkage maps being established for An. stephensi (Parvez et al., 1985). Although the relatedness of w in the Bangalore strain to the w previously reported (Aslamkhan, 1973) in the Pakistani strain is unclear, the occurrence of the same phenotypic mutant in such widely separated geographical isolates is, nevertheless, significant. The extent of genetic homogeneity among native strains of An. stephensi will affect the ability of strains designed for genetic control to reduce natural populations of this vector. Previous studies of mutations in eye and larval colors in An. stephensi indicate considerable genetic variability in this species. For example, in addition to the sexlinked white eye mutant reported for the Pakistani strain (Aslamkhan, 1973), an autosomal, colorless mutant was reported by Sharma et al. (1977). These and other authors also described rosy eye (Aslamkhan & Gul, 1979) and red eye (Sharma et al., 1979). Chestnut eye (Rather et al., 1983), scarlet, pigmentless, and red-spotted mutants (Parvez et al., 1985) have also been reported in A. stephensi. Larval color mutants of An. stephensi include green (Subbarao & Adak, 1981; Suguna, 1981; Gayathri & Shetty, 1993), golden-yellow (Adak et al., 1990), black (Adak et al., 1990; Suguna, 1981; Shetty & Gayathri, unpublished data), stripe (Sakai et al., 1981; Shetty & Gayathri, unpublished data), greenish brown (Sharma et al., 1979), and brown (Shetty et al., unpublished data). Of particular interest is the occurrence of the white eye mutation in the Bangalore strain as a sex-linked locus. This was reported previously for the Pakistani strain in the north (Aslamkhan, 1973) as well. Tests for allelism will be needed to verify whether the same locus is present in these two widely separated populations. If the w locus is the same in both strains, then some degree of genetic homogeneity in An. stephensi may exist throughout its range, and this would be useful for future genetic control efforts. If w involves different loci between strains, then the taxonomic status of such strains needs to be clarified in greater detail as suggested by Subbarao et al. (1987) and Shetty et al., (unpublished). In the study by Subbarao et al. (1987) the authors refer to different races, historically, of An. stephensi and relate these forms to vectorial capacity. Nevertheless, they considered their data as consistent with An. stephensi having ecological variants. Characterizations by classical cytogenetics along with the molecular approaches of DNA fingerprint analyses and characterizations of mitochondrial genomes are in progress in our laboratories. These integrated approaches could provide much information about the taxonomic status of geographical isolates and strains of An. stephensi. ACKNOWLEDGEMENTS This work was supported by a grant from U.G.C.-Departmental Special Assistance, the Indian Council of Medical Research, New Delhi, and a fellowship from the Fulbright Foundation to N.J. Shetty. REFERENCES CITED ADAK, T., S.K. SUBBARAO, AND V.P. SHARMA Genetics of golden-yellow larva in Anopheles stephensi. Mosq. News. 6: ASLAMKHAN, M Sex chromosomes and sex determination in the malaria mosquito, Anopheles stephensi. Pak. J. Zool. 5:

111 Shetty et al.: Anopheles stephensi Mutants 503 ASLAMKHAN, M., AND R. GUL Inheritance of the sex-linked mutant rosy, an allele of white in the malaria mosquito, Anopheles stephensi. Pakistan J. Sci. 31: BAKER, R.H., R.K. SAKAI, AND K. RAANA Genetic sexing for a mosquito sterilemale release. J. Hered. 72: BHASKAR, P., AND N. J. SHETTY Susceptibility status of Anopheles stephensi Liston to insecticides. J. Com. Dis. 26: CURTIS, C.F., J. AKIYAMA, AND G. DAVIDSON A genetic sexing system in Anopheles gambiae species A. Mosq. News 36: GAYATHRI, D.K., AND N. J. SHETTY Polytene chromosomes of Anopheles stephensi Liston - a malaria vector. Vigna Bharathi 12: 1-8. GAYATHRI, D.K., AND N.J. SHETTY. 1992a. Chromosomal translocations and inherited semisterility in the malaria vector Anopheles stephensi Liston. J. Com. Dis. 24: GAYATHRI, D.K., AND N.J. SHETTY. 1992b. Chromosomal inversions in Anopheles stephensi Liston - a malaria mosquito. J. Cytol. Genet. 27: GAYATHRI, D.K., AND N.J. SHETTY Genetics of a larval colour mutant in Anopheles stephensi, a malaria vector. J. Cytol. Genet. 28: JOSLYN, D The state of the art of genetic control of mosquitoes. Proc. New Jersey Mosq. Contr. Assoc. 67: KAISER, P.E., J A. SEAWRIGHT, D.A. DAME, AND D.J. JOSLYN Development of a genetic sexing system for Anopheles albimanus. J. Econ. Entomol. 71: KITZMILLER, J.B Genetics, cytogenetics and evolution of mosquitoes. Adv. Genet. 18: NARANG, S., AND J.A. SEAWRIGHT Linkage relationship and genetic mapping in Culex and Anopheles, pp in W.W.M. Steiner, N.J. Tabachnick, K.S. Rai and S. Narang [eds.]. Recent developments in the genetics of insect disease vectors. Stipes Publ. Co., Champaign, IL. PAL, R., AND M.J. WHITTEN. [eds.] The use of genetics in insect control. Elsevier/ North Holland, Amsterdam, Holland. pp PARVEZ, S.D., K. AKHTAR, AND R.K. SAKAI Two new mutations and a linkage map of Anopheles stephensi. J. Hered. 76: RAO, G.D.E., AND N.J. SHETTY Effect of insecticide resistance on reproductive potential in Anopheles stephensi Liston, a malaria mosquito. Int. J. Toxicol. Occup. Environ. Health 1: RATHOR, H.R., S. RASHID, AND TOQIR Genetic analysis of a new sex-linked mutant - chestnut eye an allele of the white locus in the malaria vector Anopheles stephensi. Mosq. News. 43: SAKAI, R.K., M.P. IQBAL, AND R.H. BAKER The genetics of stripe, a new morphological mutant in the malaria mosquito, Anopheles stephensi. Canadian J. Genet. Cytol. 16: SHARMA, V.P., T.R. MANI, T. ADAK, AND M.A. ANSARI Colorless-eye, a recessive autosomal mutant of Anopheles stephensi. Mosq. News 37: SHARMA, V.P., S.K. SUBBARAO, M.A. ANSARI, AND R.K. RAZDAN Inheritance pattern of two new mutants, red-eye and greenish brown larvae, in Anopheles stephensi. Mosq. News. 39: SHETTY, N.J Genetic sexing system for preferential elimination of females in Culex quinquefasciatus. J. Am. Mosq. Contr. Assoc. 3: SHETTY, N.J., AND D.K. GAYATHRI Genetic control of mosquitoes: mating competitiveness of translocation heterozygote males of Anopheles stephensi Liston - a malaria vector in laboratory cage trials, pp in P.P. Reddy, [ed.], Chromosome Damage by Environmental Agents. SUBBARAO, S.K, AND T. ADAK Linkage relationship between three autosomal mutants and functional relationship beween two eye colour mutants in Anopheles stephensi. Indian J. Malar. 18:

112 504 Florida Entomologist 77(4) December, 1994 SUBBARAO, S.K., K. VASANTHA, T. ADAK, V.P. SHARMA, AND C.F. CURTIS Eggfloat ridge number in Anopheles stephensi: ecological variation and genetic analysis. Med. Vet. Entomol. 1: SUGUNA, S.G The genetics of three larval mutants in Anopheles stephensi. Indian J. Med. Res. 73: WELLER, G.L., AND G.G. FOSTER Genetic maps of the sheep blowfly Lucilia cuprina: Linkage-group correlations with other dipteran genera. Genome 36:

113 Scientific Notes 505 AN EASILY-CONSTRUCTED TEMPORARY CAGE FOR STUDYING ANIMALS IN THE FIELD MARTHA DUNHAM Department of Entomology University of Kentucky Lexington, KY Controlled experiments are easy to perform on animals in the laboratory, but unnatural conditions can cause behavioral artifacts. Studying animals in their natural environment reduces artifacts, but makes precise control difficult. Large outdoor enclosures provide some of the advantages of each method of study; living conditions can be controlled to a great extent, yet laboratory artifacts are less likely to affect the results. Recent advances in the study of dragonfly behavior, in particular, have been made possible by the use of large flight cages (Michiels & Dhondt 1988, 1989). Return rates for dragonflies captured and released into the wild can be as low as 36% (Hilton 1983, 1984). Experimentally-treated individuals released into a cage, on the other hand, can generally be recaptured (Michiels & Dhondt 1988). Studies of longevity, maturation processes (Michiels & Dhondt 1989, Dunham 1993b) and the effects of population density (Dunham 1993a) are also possible in an enclosure. In this note, I describe a method of constructing a large cage. The design is simple, the cage is easy to build in a short time, and it is sturdy enough to withstand mph winds (pers. obs.). I have successfully used this cage to observe behavior of Pachydiplax dragonflies at a pond (Dunham 1993b, 1994). In addition, Calopteryx damselflies survived several weeks at a section of a small stream enclosed by this cage (J. Waage, Brown Univ., pers. comm.). These large insects were unable to escape from the cage. Although, in general, small birds were unable to enter the cage, on occasion they did get under the net if it was not sealed at ground level. If desired, minor modifications could be made to keep birds in or out. Michiels & Dhondt (1988, 1989) built a taller enclosure covering the same area. Figure 1. The completed flight enclosure, 10 m x 20 m x 3 m. This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

114 506 Florida Entomologist 77(4) December, 1994 The materials needed to construct the cage are listed below. All measurements are given in both English and Metric units because construction materials in the U.S. are still measured in English units. PVC pipes, 3 cm (1 1/4 in) schedule 40 (thick-walled) PVC joints (if pipe comes without bell ends), 3 cm (1 1/4 in) EMT (galvanized metal electrical conduit), 2 cm (3/4 in) Nylon rope, 1.5 cm (1/2 in) Wooden stakes, 3 x 3 x 60 cm (1 in x 1 in x 2 ft) Cable ties (also called tie wraps), 18 cm (7 in) PVC glue Sledge hammer Pipe cap, 2.5 cm (1 in) and short length of 2.5 cm (1 in) pipe Netting, 360 m 2 (12 x 20 m plus endpieces); (4000 ft 2 or 40 x 65 ft plus endpieces). Netting should be UV resistant, not nylon; 10% shadecloth works well. Sew together using nylon twine and large-eyed (tapestry) needle. S-hooks, 2.5 cm (1 in) If a PVC midpole is used: Drill bit, 1 cm (3/8 in) Carriage bolts, 7.5 x 1 cm (3 in x 3/8 in) Wing nuts, 1 cm (3/8 in) Instructions for building a 10 x 20 x 3 m enclosure are as follows. Larger enclosures can easily be constructed by adding more supports. For an enclosure 20 m long, 9 cross-supports provided the sturdiest support, but 8 will still hold up the cage under most conditions. The midpole (connecting the centers of the cross-supports) can be rope or PVC pipe. PVC is sturdy but heavy. Rope is lighter and less expensive. 1. Glue together sections of PVC pipe to form 8 or 9 cross-supports 12 m (40 ft) long. When bent into position, these will form the framework for a cage about 3 m high at the center. 2. While the glue is drying, cut sections of EMT into stakes 1.5 m (5 ft) in length. These stakes are needed to hold the cross-supports in place. Drive the stakes into the ground in 2 rows 10 m apart. If needed the end cap can be placed over the end of the pipe to prevent flaring of the pipe as it is driven into the ground. For 9 cross-supports, the interval between stakes is 2.5 m (8.125 ft); for 8 cross-supports, the interval between stakes is 2.85 m (9.25 ft). The exact height of the EMT stakes above ground is unimportant, but leave no more than one-half of the EMT protruding. 3. Attach the rope or PVC midpole to the cross-supports before erecting the enclosure. Tie or bolt the cross-supports (Fig. 2) at the appropriate distance apart to form the midpole. If the rope is used, make sure it is pulled taut so that the distance between cross-supports is the same as that between the stakes. 4. Slip ends of cross-supports over the ends of the EMT. The PVC ends will rest on the ground. 5. Pull rectangular piece of netting across top of cage. 6. Use cable ties to attach netting to PVC near the ground. 7. Sew hemicircular pieces of netting to netting covering the top to enclose the ends of the cage. The netting can be left sewn together when the cage is disassembled so that the next time it is erected this step will be unnecessary. Leave a section open for the entrance. The entrance can be held closed using S-hooks. 8. Stake edges of netting to ground, or cover edges with dirt. The latter may be necessary to keep birds out (or in).

115 Scientific Notes 507 Figure 2. Detail of PVC midpole bolted to PVC cross-support. 9. Secure the midpole at ends of the cage with ropes. If the cage sags (especially likely if a PVC midpole is used) put a support under the upper end of each restraining rope. My heartfelt thanks to J. Ballard, who helped at every stage of the design and building of the flight enclosure. N. Michiels also provided advice and assistance on construction, and on this manuscript. J. Waage, N. Michiels, C. Small, D. Lafferty, O. Reese, and A. Taylor helped build successive generations of the enclosure. This work was funded by the Animal Behavior Society and the Sigma Xi Society. SUMMARY An inexpensive and simple method is given for building a large outdoor cage. This type of enclosure can be erected quickly on most terrain, including over water. REFERENCES CITED DUNHAM, M.L. 1993a. Fighting and territorial behavior in the dragonfly Pachydiplax longipennis. Brown University, Providence, R. I. Ph.D. dissertation. DUNHAM, M. 1993b. Changes in mass, fat content, and water content with growth in adult Pachydiplax longipennis (Odonata: Libellulidae). Canadian. J. Zool. 71: DUNHAM, M The effect of physical characters on foraging in Pachydiplax longipennis (Burmeister) (Anisoptera: Libellulidae). Odonatol. 23:

116 508 Florida Entomologist 77(4) December, 1994 HILTON, D.F. J Territoriality in Libellula julia Uhler (Anisoptera: Libellulidae). Odonatologica 12: HILTON, D.F.J Reproductive behavior of Leucorrhinia hudsonica (Selys) (Odonata: Libellulidae). J. Kansas. Entomol. Soc. 57: MICHIELS, N.K. AND A.A. DHONDT Direct and indirect estimates of sperm precedence and displacement in the dragonfly Sympetrum danae (Odonata: Libellulidae). Behav. Ecol. Sociobiol. 23: MICHIELS, N.K. AND A.A. DHONDT Effects of emergence characteristics on longevity and maturation in the dragonfly Sympetrum danae (Anisoperta: Libellulidae). Hydrobiologia 171:

117 508 Florida Entomologist 77(4) December, 1994 A PRELIMINARY CHECKLIST OF THE ANTS (HYMENOPTERA: FORMICIDAE) OF EVERGLADES NATIONAL PARK BETTY FERSTER 1 AND ZACHARY PRUSAK 2 1 Boston University Biology Department, Boston, MA University of Central Florida, Department of Biology, Orlando, FL Everglades National Park encompasses 602,616 ha within Dade, Monroe and Collier counties of southern, peninsular Florida. The Park contains varied habitat types including rockland pine, mangrove swamp, hardwood hammock, freshwater slough, freshwater marl prairie, cypress swamp, and coastal prairie as well as marine and estuarine habitats (Everglades National Park official map and guide 1993). The Everglades may function as a last refuge for rare and rapidly disappearing natural communities. It also includes areas that were once managed for such diverse uses as cattle ranching and sugar cane production. The ant fauna, therefore, is likely to be rich in indigenous species of natural habitats as well as species typical of disturbed sites, including exotics. The Everglades museum had representatives of only one species of ant (Camponotus abdominalis floridanus) from within park boundaries. There is no published, comprehensive list of ants collected from within this unique area. This study provides a preliminary catalogue of ant species of the Everglades and serves as groundwork for more thorough studies of ant ecology in the Park. Ants were collected from: rockland pine along Research Road (Dade Co.); Palma Vista Hammock, a hardwood hammock near Anhinga Trail (Dade Co.); the disturbed area surrounding the buildings of the Dan Beard Research Facility (Dade Co.); wetland prairie at Hole-in Doughnut, 1 km SW of the research facilities (Dade Co.); wet flatwoods at Long Pine Key (Dade Co.); an interpretive trail that runs into wetland prairie south of Pay-Hay-Okee (Dade Co.); Rowdy Bend, a mangrove swamp north of Flamingo (Monroe Co.); coastal wetland surrounding the interpretive trail at Eco Pond south of Flamingo Bay (Monroe Co.); and disturbed habitat within historically hardwood hammock at Flamingo (Monroe Co.). Collections were made from 1 June to 31 June 1992 during both day and night. Dead twigs and sticks were split open to uncover ant nests, bark was peeled from dead trees, and fallen logs were overturned to search for ants. Foraging ants were collected when found, and nests were excavated to collect nest series. Alates and foragers attracted to lights were collected at night. This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

118 Scientific Notes 509 Leaf litter collected from hammock sites was placed in Berlese funnels for ant extraction. Vouchers were deposited in the Everglades museum. A literature search revealed ant species collected by other investigators within the Everglades. Additionally, the Archbold Biological Station database was searched for ants collected within the Everglades and deposited in Archbold Biological Station and the Smithsonian collections. Forty-seven species of ants were collected from nine locations in six habitat types within the park (Table 1). Our field efforts combined with the literature and database TABLE 1. ANT SPECIES COLLECTED FROM EVERGLADES NATIONAL PARK, THE HABITATS IN WHICH THEY WERE FOUND AND SOURCES. HH = HARDWOOD HAMMOCK; RP = ROCKY PINELAND; CP = COASTAL PRAIRIE; MG = MANGROVE; WP = WESTLAND PRAIRIE; DI = DISTURBED; FS = FRESHWATER SLOUGH. Species HH RP CP WP MG DI FS Aphaenogaster fulva 9 5 A. mariae 9 5 A. miamiana 1 6 Brachymyrmex depilis B. minutus 6 B. obscurior Camponotus caryae 9 3 C. abdominalis floridanus C. decipiens 1, 6 6 C. (Colobopsis) impressus C. pavidus 6 C. planatus 16 1 C. tortuganus 1 1 Cardiocondyla emeryi 16 1 C. nuda C. venustula 16 1 C. wroughtonii 16 1, 6 Crematogaster ashmeadi C. sp. nr. ashmeadi (undescribed) 8 C. atkinsoni 1 6 C. minutissima 1, 4 6 Cyphomyrmex minutus C. rimosus 16 8 Dolichoderus pustulatus 10 5 Dorymyrmex bureni 1 6 Eurhopalothrix floridana 6 Forelius pruinosus 1 Hypoponera opaciceps 3 6

119 510 Florida Entomologist 77(4) December, 1994 TABLE 1. (CONTINUED) ANT SPECIES COLLECTED FROM EVERGLADES NATIONAL PARK, THE HABITATS IN WHICH THEY WERE FOUND AND SOURCES. HH = HARD- WOOD HAMMOCK; RP = ROCKY PINELAND; CP = COASTAL PRAIRIE; MG = MANGROVE; WP = WESTLAND PRAIRIE; DI = DISTURBED; FS = FRESHWATER Species HH RP CP WP MG DI FS H. opacior 1, 6 H. punctatissima 16 1 Leptogenys manni 6 Leptothorax torrei 1 L. allardycei 7 Linepithema humile 16 1 Monomorium floricola M. minimum 9 5 M. pharaonis Myrmecina americana 12 5 Odontomachus brunneus 13 4 O. ruginodus Paratrechina bourbonica P. guatemalensis P. longicornis , 6 P. wojciki 6 6 Pheidole dentata 1, 8 1, P. dentigula 6 P. floridana P. megacephala 16 1 P. moerens 16 1, P. morrisi 5 Platythyrea punctata 1 1 P. cubaensis P. ejectus P. elongatus 6, P. mexicanus 16 1, P. pallidus P. seminole 1 1 P. simplex 14 3, 5 6 Quadristruma emmae Smithistruma dietrichi 1, Solenopsis abdita 1 6 S. geminata 1, 6, 8 8 S. invicta , 6 S. tennesseensis 6 6 Strumigenys eggersi 16 1, 6

120 Scientific Notes 511 TABLE 1. (CONTINUED) ANT SPECIES COLLECTED FROM EVERGLADES NATIONAL PARK, THE HABITATS IN WHICH THEY WERE FOUND AND SOURCES. HH = HARD- WOOD HAMMOCK; RP = ROCKY PINELAND; CP = COASTAL PRAIRIE; MG = MANGROVE; WP = WESTLAND PRAIRIE; DI = DISTURBED; FS = FRESHWATER Species HH RP CP WP MG DI FS S. gundlachi S. louisianae 6 6 Tapinoma litorale 1 1 T. melanocephalum T. sessile 6 Tetramorium caldarium , 6 T. simillimum 16 1 Wasmannia auropunctata Xenomyrmex floridanus Zacryptocerus varians Present study. 2 Nielsson et al Smith Smith Wheeler Ants of Florida database, Archbold Biological Station collection and Smithsonian collection. 7 Smith Koptur Questionable record, specimens unavailable for verification (M. Deyrup, pers. comm.) 10 Listed as Dolichoderus plagiatus subsp. pustulatus in Wheeler Listed as Ponera opaciceps in Smith Listed as Myrmecina graminicola Latr. subsp. americana Emery var. Brevisponosa Emery in Wheeler Listed as Odontomachus haematodes subsp. insularis Guerin in Smith Listed as Pseudomyrma flavidula in Smith 1930, and Wheeler Listed as Xenomyrmex stolli subsp. rufescens Wheeler in Smith Exotic species. searches revealed a total of 75 species. Twenty-six species were exotic. Fifteen of the exotic species originated from old world tropics and eleven originated from new world tropics. Because unequal amounts of time were spent at each site and collecting methods varied with each habitat type, these data cannot be used as a measure of habitat species abundance. The Everglades is a large ecological preserve located at the southern tip of peninsular Florida. It includes both pristine habitat and historically man-modified areas, and lies between two areas of relatively well-studied ant faunas: 1) the Florida Keys (Wilson 1964, Deyrup et al. 1988, Deyrup 1991), and 2) areas of southern Florida north of the Everglades (Smith 1930, Wheeler 1932, Smith 1933, Nielsson et al. 1971, Deyrup & Trager 1986). Thirty-five percent of the species that have been collected within the Everglades were exotics and none of these species were restricted to disturbed habitats. However, because our collections from natural areas contained disturbances such as roadsides and trails, these are effectively disturbed areas. The proportion of exotic species that were found in the Everglades was similar to the proportion of exotic species that have been found in the Florida Keys (Deyrup 1991), and similar to the proportion of exotic ant species found in residential Dade county (Deyrup 1991).

121 512 Florida Entomologist 77(4) December, 1994 The present study did not extend the ranges for any native or exotic species. No new species were discovered. Many exotics were expected because of the neotropical climate and proximity to centers of commerce and human traffic. Future collections in pristine habitats, less prone to the invasions of some exotics, would be useful additions to our knowledge of the ant assemblage of the Everglades. Mark Deyrup (Archbold Biological Station) supervised ant identifications as well as providing encouragement, support and endless hours of entertainment. No acknowledgment could exhibit the amount of gratitude and respect the authors have for Dr. Deyrup. Alfredo Begazo, Jamie Prusak, Walter Meshaka, Jr., Lloyd R. Davis, Jr., Marcia Moretta, Elizabeth A. Capaldi and an anonymous reviewer provided helpful suggestions to improve this manuscript. SUMMARY Forty-seven species of ants were found in six habitat types in Everglades National Park during nine collecting trips in June A search of both the literature and a database of Florida ants are combined with our efforts to form a preliminary list of 75 species of ants from the park. REFERENCES CITED DEYRUP, M Exotic ants of the Florida Keys, pp in W. Hardy Eshbaugh [ed.], Proceedings of the 4th Symposium on the Natural History of the Bahamas. San Salvador, Bahamas. DEYRUP, M., AND J. TRAGER Ants of the Archbold Biological Station, Highlands County, Florida. Florida Entomol. 69: DEYRUP, M., J. TRAGER, N. CARLIN, AND G. UMPHREY A review of the ants of the Florida Keys. Florida Entomol. 71: KOPTUR, S Plants with extrafloral nectaries and ants in Everglades habitats. Florida Entomol. 75: NATIONAL PARK SERVICE Everglades, Official map and guide. U.S. Dept. of the Interior. Washington, D. C. NIELSSON, R.J., A.P. BAKTAR, AND H.A. DENMARK A preliminary list of ants associated with aphids in Florida. Florida Entomol. 54: SMITH, D.R Superfamily Formicoidea. pp in K. V. Krombein, P. D. Hurd, D.R. Smith, and B.D. Burks [eds.], Catalogue of Hymenoptera in America north of Mexico, vol. 2., Smithsonian Institution Press, Washington, D.C. SMITH, M.R A list of Florida ants. Florida Entomol. 14: 1-6. SMITH, M.R Additional species of Florida ants, with remarks. Florida Entomol. 17: WHEELER, W.M A list of the ants of Florida with descriptions of new forms. J. New York Entomol. Soc. 40: WILSON, E.O The ants of the Florida Keys. Breviora. 210: 1-14.

122 Scientific Notes 513 A CONTAINER FOR ECLOSION AND HOLDING ADULT INSECTS PRIOR TO MASS RELEASE J.M. SIVINSKI, C.O. CALKINS, AND R. BARANOWSKI Insect Attractants, Behavior, and Basic Biology Research Laboratory, Agricultural Research Service, U.S. Department of Agriculture, Gainesville, FL The mass release of adult insects requires containers for eclosing pupae and, at times, for holding adults for several days (e.g., Tanaka boxes used in the release of sterile fruit flies illus. in Holler, et al. 1984). These units can be both bulky and expensive. Casual observation of large numbers of confined insects has shown that the majority of their time is spent resting on cage walls or on objects in the cage. The large air volume of a cage is essentially wasted space. With this in mind, a container in the shape of a bag roughly the size of a pillow case was designed for eclosing and holding large numbers of adult braconid parasites. The eclosion bag is mostly walls, takes up little room, and is easy to move about, both within the rearing/eclosion facility and during transportation to release sites. The bags are made from pieces of 32 x 32 nylon mesh screen (Lumite, Inc., Gainesville, GA.) that are sewn together and closed continuously across the bottom and the length of one side by velcro strips. For our purposes, we found that a bag 60 cm wide and 90 cm long was ideal, but smaller as well as larger bags (up to 2m long) have been constructed, and insects have been successfully maintained in them. Earlier designs had rounded edges on the bottom to prevent insects from accumulating in corners and milling. The parasite being held in our research, Diachasmimorpha longicaudata (Ashmead), did not display this behavior to any significant extent, and bags with square-edges proved easier to sew and cheaper to produce. Should these bags be adapted for use with insects in which milling is typically a problem, e.g., tephritid fruit flies, the inclusion of rounded bottom corners might be considered. During the holding/maturation period, parasites were fed a solution of honey and water that was poured into 30 cm polyethylene tubes with an inner diam of 10 mm. The tube ends were closed with 3 cm cotton wicks. The diluted honey seeped through the wick and provided a feeding surface for the insects. To prevent either dripping or incomplete absorbance, we found it necessary to vary the proportions of honey and water with changes in temperature and relative humidity. The tube was held along the upper margin of the bag with a large (#5) binder clip. Cages were then hung from lines, sometimes at two levels in rooms of normal height (~ 2.5 m). An S-hook or open paper clip fitted through the binder clip and then over the line made an effective hanger. Parasitized Caribbean fruit fly pupae (Anastrepha suspensa (Loew)) were poured evenly on the bottom of the bag. A typical volume was 375 ml, which gave rise to approximately 5000 adult parasites. For our needs, adults were maintained in the bags for five days after the first eclosion. They were then taken by vehicle to the release sites. Lines strung in the back of a van provided a convenient method of suspending them during transportation, although bags could be laid flat and stacked several deep with no apparent ill effects on the parasites. At the release site, the feeding tube was removed, the velcro opened, and the pupal remains poured into a bucket. The bag was spread open and shaken in the air. This was a particularly useful technique in releasing D. longicaudata, which have a rela- This article is from Florida Entomologist Online, Vol. 77, No. 4 (1994). FEO is available from the Florida Center for Library Automation gopher (sally.fcla.ufl.edu) and is identical to Florida Entomologist (An International Journal for the Americas). FEO is prepared by E. O. Painter Printing Co., P.O. Box 877, DeLeon Springs, FL

123 514 Florida Entomologist 77(4) December, 1994 Fig. 1. A bag cage being fitted with a feeding tube.