FLORIDA ENTOMOLOGIST (An International Journal for the Americas)

(ISSN 0015-4040) FLORIDA ENTOMOLOGIST (An International Journal for the Americas)

Volume 67, No.2 June, 1984

TABLE OF CONTENTS 67th Annual Meeting of the Florida Entomological Society i 68th Annual Meeting-First Announcement ii

...... • '4 BENSCHOTER, C. A., AND P. C. WITHERELIr--Lethal Effects of Suboptimal Temperatures on Immature Stages of Anastrepha suspensa _.____ 189 ALI, A., D. M. SAUERMAN, AND J. K. NAYAR-Pathogenicity of Indus- trial Formulations of Bacillus thuringiensis serovar. israelensis to Larvae of Some Culicine Mosquitoes in the Laboratory ._ 193 BOCZEK, J., AND R. DAVIS-New Species of Eriophyid Mites (Acari: Eriophyidae) .______198 JONES, V. P., AND J. G. MORsE-A Synthesis of Temperature Dependent Developmental Studies with the Citrus Red Mite, Panonychus citri (McGregor) (Acari: Tetranychidae) . .... _. . .__ 213 GAGNE, R. J., AND G. M. BEAVERS-Contarinia spp. (Diptera: Ceci domyiidae) from Shoots of Slash Pine (Pinus elliotti Engelm.) with the Description of a New Species Injurious to Needles 221 LLOYD, J. E.-Evolution of a Firefly Flash Code .______228 LANDOLT, P. J.-Reproductive Maturation and Premating Period of the Papaya Fruit Fly, Toxotrypana curvicauda (Diptera: Teph- ritidae) . .__ . . .___ 240 LIN, J., AND J. D. HARPER-Isolation and Culture of Entomophthora gammae, a Fungal Parasite of Noctuid Larvae _. .. ._.. 245 ZOEBISCH, T. G., D. G. SCHUSTER, AND J. P. GILREATH-Liriomyza trifolii: Oviposition and Development in Foliage of Tomato and Common Weed Hosts .. ... 250 WITHERELL, P. C.-Methyl Bromide Fumigation as a Quarantine Treat ment for Latania Scale, Hemiberlesia lataniae (Homoptera: Diaspididae) ---.--.-.. -- ._._._. .._._. __ . .______254 BROWER, J. H., AND L. D. CLINE-Response of Trichogramma pretiosum and T. evanescens to Whitelight, Blacklight or No-Light Suction Traps .. ..__ ------______262 ELVIN, M. K., AND P. E. SLODERBECK-A Key to Nymphs of Selected Species of Nabidae (Hemiptera) in the Southeastern USA 269 Continued on Back Cover

Published by The Florida Entomological Society FLORIDA ENTOMOLOGICAL SOCIETY

OFFICERS FOR 1983-84 President .. .__ ...... __ .. C. W. McCoy President-Elect .__ . .__ . _. ._ .. __ ._ __ . M. L. Wright, Jr. Vice-President ...__ ._. ._ ...._.._.... __ .__ ... ._ ...... J. A. Reinert Secretary . .__ _.. _.. _. __ ...... _. __ . ._.. .. D. F. Williams Treasurer .. __ .__ .... .__ .__ _.. ._....._..._.. __ .. ._ ... .._...... _A. C. Knapp

J. R. Cassani J.L. Knapp D. C. Herzog Other Members of the Executive Committee _... K. Lee C. A. Morris W. L. Peters C. A. Musgrave Sutherland

PUBLICATIONS COMMITTEE

Editor ..._.. __ .. .__ .... _._ ... _._._ _....._.. .. __ .. __ .. .__ . C. A. Musgrave Sutherland Associate Editors _. .._ __ .__ .... ._...... D. C. Herzog O. Sosa, Jr. M. D. Hubbard J. R. McLaughlin J. B. Heppner H. V. Weems, Jr. Business Manager ___.._ __ .._.. _._._._.. __ .. A. C. Knapp

FLORIDA ENTOMOLOGIST is issued quarterly-March, June, September, and December. Subscription price to non-members is $20.00 per year in advance, $5.00 per copy. Membership in the Florida Entomological Society, including subscription to Florida Entomologist, is $15 per year for regular membership and $5 per year for students. Inquires regarding membership and subscriptions should be addressed to the Business Manager, P. O. Box 7326, Winter Haven, FL 33883-7326. Florida Entomologist is entered as second class matter at the Post Office in DeLeon Springs and Gainesville, FL. Authors should consult "Instructions to Authors" on the inside cover of all recent issues while preparing manuscripts or notes. When submitting a paper or note to the Editor, please send the original manuscript, original figures and tables, and 3 copies of the entire paper. Include an abstract and title in Spanish, if possible. Upon receipt, manuscripts and notes are ac knowledged by the Editor and assigned to an appropriate Associate Editor who will make every effort to recruit peer reviewers not employed by the same agency or institution as the authors (s). Reviews from individuals working out-of-state or in nearby countries (e.g. Canada, Mexico, and others) will be obtained where possible. Manuscripts and other editorial matter should be sent to the Editor, C. A. Musgrave Sutherland, 4849 Del Rey Blvd., Las Cruces, NM 88001.

This issue mailed June 30, 1984 THE 67th ANNUAL MEETING OF THE FLORIDA ENTOMOLOGICAL SOCIETY The Florida Entomological Society will hold its 67th Annual meeting on 24-27 July 1984 at the Holiday Inn, 6515 International Drive, Orlando FL 32809; telephone-1- (305) -351-3500. Room rates will be $58.00, for single, double, triple, or quadruple. Questions concerning the local arrangements should be directed to: FREDERICK L. PETITT, Chairman Local Arrangements Committee Florida Entomological Society Walt Disney World-Epcot Center-The Land P.O. Box 40 Lake Buena Vista, Florida 32830 USA Phone: 1- (305) -827-7256 To present a paper, the tear out sheet must be postmarked and sent no later than 15 MAY 1984, to: JAMES A. REINERT, Program Chairman Ft. Lauderdale Research and Education Center University of Florida 3205 S.W. College Avenue Ft. Lauderdale, Florida 33314 USA Eight minutes will be allotted for presentation of oral papers, with 2 minutes for discussion. In addition, there will be a separate session for members who may elect to present a Project (or Poster) Exhibit. The 3 oral student papers judged to be the best on content and delivery will be awarded monetary prizes during the meeting. Student authors must be Florida Entomological Society Members and must be registered for the meeting. Awards will be $125.00, 75.00 and 50.00. The 3 student display presentations judged to be the best on content and preparation will also be awarded monetary prizes during the meeting. Studentt authors must be Florida Entomological Society Members and must be registered for the meeting. Awards will be $125.00, 75.00 and 50.00.

Registration Schedule1 for Annual Meeting:

Preregistration Registration On Site

Full & Sustaining Members $35.00 $40.00 Student not in Student Contest 18.00 20.00 Student in Student Contest 13.00 15.00 Each Extra Banquet Ticket 10.00 10.00

1 Each fee includes one banquet ticket. 68th ANNUAL MEETING-FLORIDA ENTOMOLOGICAL SOCIETY First Announcement In 1985, the 68th annual meeting of the Florida Entomological Society will be held at the Ocho Rios Sheraton Hotel, Ocho Rios, Jamaica, W.I. Meeting dates are from Monday, 5 August to Thursday 8 August 1985. The Ocho Rios Sheraton is located on Mallards Beach on the north coast of Jamaica and is a complete resort facility offering excellent meeting facilities and day and night entertainment. Emmer Travel, Inc. of Gainesville, FL has been designated the official travel agency of these meetings. They will make all of your travel arrange ments including airline tickets and hotel reservations. The land/hotel pack age will be $145.00 per person (double occupancy) or $260.00 per person (single occupancy). This price includes: *4 days/3 nights hotel accommodations *all Jamaican room taxes *all service charges on the rooms *porter service at the hotel *welcome rum swizzle on arrival at the hotel *transportation from/to the Montego Bay airport Children under 17 stay FREE at the hotel when they stay in the room with 2 adults. Those under 17 must pay $22.00 for round-trip transportation from/to Montego Bay. Persons wishing to stay after the meetings will pay $70.00 per room per night, including taxes and service charges. Children under 17 staying with adults after the meetings are not charged. A $50.00 per person deposit will secure your reservations. The balance will be due by 15 June'1985. All costs are fully refundable if the reservation is cancelled prior to 21 June 1985. Thereafter, a cancellation charge of $75.00 per person on the land/hotel package will be levied. Information on airfare rates, airline schedules and meeting registration fees will be forthcoming soon. To make reservations, send check for $50.00 to Emmer Travel, Inc., 2801 SW Archer Road, Gainesville, FL 32608. Include your name, address, and r hone number. For further information, call Emmer Travel, Inc. toll free at (800) -342-2223 (Florida residents) or (800) -874-8487 (nationwide) or (904) -377-1222 (in Gainesville).

FIRST CALL FOR PAPERS Submitted papers for the 68th annual meetings of the FES should be 8 minutes long. Time alloted will be 10 minutes: 8 minutes for presentation and 2 minutes for discussion. Use the following tear sheet to submit a paper title. Return to : James A. Reinert D Oral Presentation Ft. Lauderdale Research & Project Exhibit Session Education Center D University of Florida (Formerly Poster Session) 3205 S.W. College Avenue D Student Paper Ft. Lauderdale, FL 33314 DEADLINE: 15 May 1985 1985 Meeting in Ochos Rios, Jamaica Author's Name _ Title of Paper _

Affiliation and Address _ of the First (Presenting) Author

Time Required for Presentation (Max. 10 min.) Abstract: Must be Provided. Do not use more than 75 words.

SuggestionforEveningBullSession ~_~~ _ SLIDE POLICY FOR ANNUAL MEETINGS The following slide policy will govern slide presentations at the Annual Meetings. Only Kodak Carousel projectors for 2 x 2 slides will be available. However motion picture projectors will be available by special request to the Local Arrangements Chairman prior to the date of the meeting. Authors should keep slides simple, concise, and uncluttered with no more than 7 lines of type on a rectange 2 units high by 3 units wide. All printed information should be readable to an audience of 300 persons. A previewing room will be designated for author's use. A projectionist will be available in the previewing room at least one hour before each session. Authors are expected to give the projectionist their slides in the previewing room prior to each session. Slides will be returned to the authors after each session in the meeting room. Authors are expected to organize their slides in proper order in their personal standard Kodak Carousel slide tray (no substitution, please). Only a few slide trays will be available in the previewing room from the projec tionist for hardship cases. Slides in the tray should be in correct order start ing with slot #1 of the tray and positioned correctly (position of slides to go into tray: 1. upside down, and 2. lettering readable from this position upside down and from right to left). A piece of masking tape should be placed on the slide tray by the author and the following information should be written on the tape: 1. author's name, 2. session date, and 3. presentation time. Benschoter & Witherell: Suboptimal Temperatures 189

LETHAL EFFECTS OF SUBOPTIMAL TEMPERATURES ON IMMATURE STAGES OF ANASTREPHA SUSPENSA

C. A. BENSCHOTER AND P. C. WTTHERELLl Subtropical Horticulture Research Unit Agricultural Research Service, S & E, USDA Miami, FL 33158 USA

ABSTRACT Immature stages of Anastrepha suspensa (Loew) (Diptera: Tephritidae) were exposed for variable periods of time to temperatures ranging from 1.7 to 15.6°C. Lethal effects varied inversely with temperature. Regression equations were calculated and LT50 and LT95 values (lethal time for 50% and 95% mortality) are presented. Data points for all stages showed high linearity when percentage mortalities were converted to probits and plotted against log-days (exposure time), providing correlation coefficients within the range of 0.934-0.999. Susceptibility to cold decreased with the age of the eggs at the temperatures and exposure periods tested. Less than 50% of mature larvae were killed from exposure to 15.6°C for 28 days, but 7.2°C for 7 days produced 99.8% mortality. Exposure to 15.6°C had essentially no lethal effect on pupae and 12.8°C for 28 days killed only 45.4% of them. Two-day-old pupae were more sensitive than older groups exposed to 4.4°C. By comparison, mortality of 2-day-old pupae exposed to this temperature for 3 days was 99.8% while mortality of 10-day-old pupae exposed for 15 days was 99.9%. The order of susceptibility (LT95 ) of immature stages of A. S1lspensa exposed to 7.2°C was larvae> eggs> pupae.

RESUMEN Estadios inmaduros de Anastrepha suspensa (Loew) (Diptera: Tephri tidae) fueron sometidos durante diferente peri6dos de tiempo a tempera turas de 1.7 a 15.6°C. El efecto letal vari6 inversamente a la temperatura.

Se calculo la ecuacion de regresi6n y los valores de TL50 y TL95 (tiempo letal para mortalidades de 50 y 95%). Todos los estadios mostraron alta linealidad cuando los porcentages de mortalidad se pasaron a probits y se representaron en gratica coeficientes de correlaci6n de rango de 0.934 a 0.999. La susceptibilidad al frio disminuy6 con la edad de los huevos para los periodos de tiempo y exposicion probados. Menos de un 50% de larvas maduras murieron al exponerlas a 15.6°C durante 28 dias, pero una ex posicion a 7.2°C durante 7 dias produjo una mortalidad de 99.8%. La ex posicion de las pupas a 15.6°C no tuvo efecto letal y a 12.8°C durante 28 dias solo mato el 45.4%. Las pupas de dos dias de edad fueron mas sensibles que el grupo de mayor edad expuestos a 4.4°C. En comparacion la mortalidad de pupas de dos dias de edad expuestas a esta temperatura durante tres dias fue de 99.8%, mientras que la mortalidad de pupas de 10 dias de edad expuestas durante 15 dias fue de 99.9%. En orden de suceptibilidad al frio

(TL95 ) de estadios inmaduros de A. suspensa expuestos a 7.2°C fue en orden descendiente: larva, huevo, pupa.

Data concerning the lethal effects of suboptimal temperatures on the Caribbean fruit fly, Anastrepha suspensa (Loew) are lacking. Prescott and

lMethods Development Station, USDA, APHIS, PPQ, 13601 Old Cutler Road, Miami, FL 33158 USA. 190 Florida Entomologist 67 (2) June, 1984

Baranowski (1971) determined the optimal temperatures for development of immature stages of this species over the range of 10-35°C. The USDA-APHIS-PPQ Treatment Manual (1983) lists a cold treatment for fruit infested by Anastrepha species other than the Mexican fruit fly, Anastrepha ludens (Loew). This all-inclusive group presumably includes the Caribbean fruit fly even though there is no published information on this species. This paper investigates the efficacy of using low temperature to control A. suspensa and reports the results of exposing immature stages of this fly to constant temperatures ranging from 1.7°-15.6°C.

MATERIALS AND METHODS All insect stages used in these experiments were taken from the labora tory colony. Eggs were collected from 1 to 2-wk-old laboratory stock flies. The num bers of eggs used in each test lot were estimated volumetrically with an eyedropper, then placed in random batches on small squares of wet blotting paper. Each blotter was placed in a petri dish which was enclosed in a plastic bag to retain moisture. After cold exposure the blotter containing the eggs was placed on a small sponge which elevated the eggs above water in the bottom of a petri dish. The water was used to trap hatched larvae which crawled off the blotter. Dilute (0.03%) sodium benzoate solution was used as a mold inhibitor in the moist environment required for eggs. After incubation at 25-27°C and 85+% RH for 4 days, larvae and unhatched eggs were counted with a dissecting microscope to determine the number of eggs treated and percent hatch. Eggs (2, 8, and 24 h old) were held at 1.7°C for 1-10 days, at 4.4°C for 3-15 days, and at 7.2°C for 6-18 days, 3 replicates per treatment. Mature larvae (8 days old) were washed from laboratory diet and placed in lots of about 200 (measured volumetrically) into 29.6-ml (8 oz) waxed paper cups containing moist vermiculite (6 ml H 2 0 per 50 ml vermiculite). Six small holes were punched in the lids of the cups to allow air exchange. Larvae were exposed to 7.2, 10.0, 12.8 and 15.6°C for 7, 14, 21 and 28 days, 8 replicates per treatment (except 3 replicates at 7.2°C). Larvae that were able to pupate and emerge as adult flies were counted as survivors. Pupa selected randomly from a single lot were measured volumetrically into test aliquots of 175-200 and placed in wax paper cups described for larvae. At 3, 6, and 10 days of age, they were exposed (3 replicates) to 7.2, 10,12.8 and 15.6°C for 7, 14,21 and 28 days. In separate tests, pupae 2, 4, 6, 8 and 10 days old were exposed about 300 per test unit (5 replicates) to 4.4°C for 3, 7, 11 and 15 days. Survival was based on the number of adult flies that emerged following treatment. Data were corrected for natural mortality in controls by Abbott's formula (1925). Then, percent mortality values were transformed to probits and ex posure periods (days) to logarithms. The transformed data were used to calculate time mortality regression equations and LT50 and LT95 values. The linear regression equation took the form of Y = a + bx where Y = probit mortality, A = y intercept, b = slope, and x = log days. Visual examination of the plotted data (not presented) revealed that, in some instances, the lowest data points did not align with higher points which otherwise showed Benschoter & Witherell: Suboptimal Temperatures 191 good linearity. These lower points were not used in computing the regression lines. This tendency of point deviation at the lower end of the mortality line is clearly demonstrated by data of Baker (1939) regarding effects of low temperature on fruit flies infesting fruits in Hawaii.

RESULTS AND DISCUSSION When data points were plotted on graph paper, sensitivity to cold gen erally decreased with age of the eggs for all temperatures tested (Table 1). At 107°C, 8 days exposure produced 100% mortality of 2- and 8-h-old eggs, and 10 days of exposure resulted in 99.87% kill of 24-h eggs. At 4.4°C, 100% kill of 2- and 8-h eggs resulted from exposures of 9 and 12 days, respec tively, while 15 days exposure produced 99.87% mortality of 24-h eggs. Complete mortality was not attained in any of the 7.2°C tests. Ten (10) days of exposure destroyed 99.5% of the 2-h eggs and 99.2% of 8-h eggs, while at 14 days it produced 98.6% mortality of 24-h eggs. Results for larvae exposed to 7.2, 10, 12.8 and 15.6°C for 7, 14, 21 and 28 days are presented in Table 2. Lethal effects varied inversely with tem perature. A temperature of 7.2°C was so severe that 99.8% mortality re sulted from 7 days exposure. Consequently, not enough data points were available to provide a regression equation for this temperature. At the other extreme, the lethal effect of 15.6°C was very slight, with less than 50% mortality after 28 days exposure. Response of pupae to cold exposure was similar to that for the other stages, with pupal mortality increasing as temperature decreased. Results are shown in Table 3. At 4.4°C, 2-day-old pupae were the most sensitive to

TABLE 1. EFFECT OF LOW TEMPERATURE ON HATCHABILITY OF EGGS OF A nastrepha suspensa, 3 REPLICATES.

Regression equation Age of eggs No. eggs LT (days)' (Y=a+bx) (h) treated 50% 95% a b

107°C' 2 5561 1.6 3.4 4.07 4.82 8 5005 2.3 5.4 3.32 4.56 24 4686 2.2 5.0 3.42 4.62 4.4°C3 2 2976 1.3 4.7 4.69 2.94 8 2998 3.7 6.2 0.99 7.12 24 3748 5.4 9.1 -0.51 7.48 7.2°C4 2 6272 2.3 6.1 3.65 3.84 8 6251 2.7 7.2 3.39 3.82 24 7677 8.3 10.7 -3.30 9.03

'LT = lethal time in days at indicated temperatures for 50% and 95% mortality. 28 exposure periods ranging from 1-10 days. 35 exposure periods ranging from 3-15 days. 45 exposure periods ranging from 2-14 days. 192 Florida Entomologist 67 (2) June, 1984

TABLE 2. EFFECT OF SUBOPTIMAL TEMPERATURES ON SURVIVAL OF MATURE (8-DAY-OLD) LARVAE OF Anastrepha suspensa.

Regression Treatment1 No. equation temperatures larvae LT (days)" (Y=a+bx) (OC) treated 50% 95% a b

7.2 2671 <5.0 <7.0 - 3 - 3 10 4938 4.9 10.3 1.49 5.10 12.8 6309 10.5 31.0 1.44 3.49 15.6 6195 22.1 3.30 1.27

1 Exposure periods were 7, 14, 21 and 28 days. 8 replicates, except 3 replicates at 7.2 ° C. "LT = lethal time in days for 50% and 95% mortality. 3Insufficient data points for regression analysis. TABLE 3. EFFECT OF SUBOPTIMAL TEMPERATURES ON SURVIVAL OF PUPAE OF Anastrepha suspensa.

Regression No. equation Age of pupae pupae LT (days) 1 (Y=a+bx) (days) treated 50% 95% a b

4.4°C" 2 5879 0.25 (hr) 6.25 (hr) 7.28 1.08 4 6329 7.6 21.0 1.71 3.74 6 6132 2.8 6.9 3.07 4.28 8 6730 3.6 7.2 1.97 5.47 10 7336 5.7 9.8 -0.15 6.85 7.2°C3 3 2208 11.2 18.8 -2.71 7.35 6 2022 4.7 lOA 1.82 4.76 10 2209 7.5 13.3 -0.81 6.64 10°C3 3 2270 18.0 33.3 -2.71 6.15 6 2035 12.9 20.8 -3.92 8.02 10 2270 17.7 30.0 -3.94 7.17 12.8°3 3 2258 40.6 156.6 0.47 2.81 6 2054 31.9 85.5 -0.80 3.86 10 2321 46.0 196.4 0.64 2.62 15.6°3 3 2222 6 2061 10 2327

1 LT = lethal time in days (except where shown in hours) at indicated temperatures for 50% and 95% mortality. "Exposure periods were 3, 7, 11 and 15 days. 5 replicates. 3Exposure periods were 7, 14, 21 and 28 days. 3 replicates. No lethal effects at 15.6 0 C. Benschoter & Witherell: Suboptimal Temperatures 193

cold, their longevity being measured in hours rather than days. Comparing results for pupae in the other age groups, 6-day-old pupae were uniformly more susceptible to cold than younger or older individuals at all temperatures tested. Mortality of pupae in the 12.8°C tests was very moderate with less than 50% kill of 6-day old pupae exposed for 28 days. Response of pupae at 15.6°C was negligible and was rated as "no effect". Results of exposing immature stages of A. suspensa to suboptimal tem peratures indicated that temperatures of 12.8°C or higher would have little practical value as a control measure against this insect. The order of ) susceptibility (LT95 of immature stages of A. suspensa exposed to 7.2°C was larvae> eggs> pupae. The data points in all of the tests had high correlation coefficients (0.934 0.999) indicating a high degree of linearity. Because of this, the regression equations may be used to estimate exposure peribds (at a given temperature) for any desired level of mortality, or to predict mortalities at other (lower) temperatures not actually tested.

REFERENCES CITED ABBOTT, W. S. 1925. A method of computing the effectiveness of an insecti cide. J. Econ. Ent. 18: 265-7. BAKER, A. C. 1939. The basis for treatment of products where fruit flies are involved as a condition for entry into the United States. USDA Circ. No. 551. 7 p. PRESCOTT, J. A. III, AND R. M. BARANOWSKI. 1971. Effects of temperature on the immature stages of Anastrepha suspensa (Diptera: Tephritidae) . Florida Ent. 54: 297-303. USDA-APHIS-PPQ Treatment Manual. 1983. Sec. VI, T107: 23-25. .. . • ••••••• • • • PATHOGENICITY OF INDUSTRIAL FORMULATIONS OF BACILLUS THURINGIENSIS SEROVAR. ISRAELENSIS TO LARVAE OF SOME CULlCINE MOSQUITOESl IN THE LABORATORY ARSHAD ALI University of Florida, IFAS, Agricultural Research and Education Center, P. O. Box 909, Sanford, FL 32771 USA AND DONALD M. SAUERMAN AND JAI K. NAYAR Florida Medical Entomology Laboratory 200 9th Street, S.E. Vero Beach, FL 32962 USA

ABSTRACT In the laboratory, Teknar®, a flowable concentrate formulation, and Bactimos® and Vectobac®, 2 wettable powder formulations of Bacillus thuringiensis serovar israelensis (Bti) were tested against laboratory main-

lDiptera: Culicidae. Benschoter & Witherell: Suboptimal Temperatures 193

lDiptera: Culicidae. 194 Florida Entomologist 67 (2) June, 1984

tained late 3rd and early 4th instars of Aedes taeniorhynchus (Wiedemann), Anopheles albimanus Wiedemann, Culex nigripalpus Theobald, Cx. restuans Theobald, Cx. salinarius Coquillett, Cx. quinquefasciatus Say, and Wyeomyia vanduzeei Dyar and Knab. Aedes taeniorhynchus was susceptible (LC9o = 0.114 to 0.772 ppm) to the test formulations. Culex nigripalpus was generally the most susceptible to each formulation, followed by Cx. salinarius, Cx. quinquefasciatus, and Cx. 'restuans. The 3 formulations of Bti were relatively ineffective against An. albimanus and Wy. vanduzeei. Considering the relative potencies of the test formulations of Bti, Bactimos® invariably was the most effective formulation and Teknar® the least against the species tested under laboratory conditions.

RESUMEN En el laboratorio, Teknar®, una formulaci6n fluida concentrada, y Bactimos® y Vectobac®, 2 formulaciones humedecibles de polvo de Bacillus thuringiensis serovar israelensis (Bti) fueron probadas contra tarde tercero y temprano cuarto estadio mantenidos en ellaboratorio de Aedes taeniorhyn chus (Wiedemann), Anopheles albimanus Wiedemann, Culex nigripalpus Theobald, Cx. restuans Theobald, Cx. salinarius Coquillett, Cx. quinque fasciatus Say, and Wyeomia vanduzeei Dyar and Knab. Aedes taeniorhynchus fue susceptible (LC9o = 0.114 a 0.772 ppm) a las formulaciones probadas. Culex nigripalpus fue generalmente la mas sus ceptible a cada formulaci6n, seguida por Cx. salinarius, Cx. quinque fasciatus, y Cx. restuans. Las 3 formulaciones de Bti fueron relativamente ineficientes contra An. albimanus y Wy. vanduzeei. Considerando la relativa potencia de las formulaciones de Bti probadas, Bactimos® invariablemente fue la formulaci6n mas efectiva, y Teknar® la menos efectiva contra las especies probadas bajo condiciones de laboratorio.

A particularly potent isolate of the entomogenous bacterium, Bacillus thuringiensis Berliner,'was discovered by Goldberg and Margalit (1977) in soil of sewage lagoons in Israel. This isolate, corresponding to a new sero type H-14, was designated the variety israelensis by de Barjac (1978). In the past 6 years, B. thuringiensis serovar.israelensis (Bti) has been tested in the laboratory and field in different parts of the world against at least 70 mosquito species (Anonymous 1982) and has proven to be a highly effective mosquito larvicide. Some commercial producers have developed a variety of increasingly toxic and potent formulations of this biocide for use against mosquitoes and some other aquatic pest and vector insects. At present, sev eral wettable powder (WP), fiowable concentrate (FC), granular (G), and slow or sustained release formulations of the microbial insecticide are avail able and more are being formulated for testing and eventual marketing. Some of these formulations are currently registered for use against mos quitoes and simuliid black flies in the United States. This paper reports laboratory evaluations of 3 industrial formulations of Bti against 7 species of mosquitoes maintained in the laboratory.

MATERIALS AND METHODS The formulations evaluated were 2 WPs, Bactimos® (Biochem Products, Montchanin, DE) and Vectobac® (Abbott Laboratories, North Chicago, IL), Ali et al.: Pathogenicity of Bt israelensis 195

.and a FC, Teknar® (Sandoz, Inc., San Diego, CA). The reported potencies of Bactimos® and Vectobac®, respectively, were 3500 Aedes aegypti (AA) IUfmg, and 2000 AA IUfmg, while Teknar® contained 1500 AA IUfmg. For laboratory assays, a mixture of late 3rd and early 4th instars of Aedes taeniorhynchus (Wiedemann), Anopheles albimanus Weidemann, Culex nigripalpus Theobald, Cx. restuans Theobald, Cx. salinarius Coquillett, Cx. quinquefasciatus Say, and Wyeomyia vanduzeei Dyar and Knab were used. These species were maintained at the Florida Medical Entomology Laboratory at Vero Beach, Florida. The procedures of bioassays were generally the same as described by Mulla et al. (1982). Twenty larvae were placed in a 120-ml disposable cup containing 100 ml of tap water (pH 6.8 ± 0.2). Distilled water (pH 6.9 ± 0.2) was used for Wy. vanduzeei because of the high larval mortality of this species occurring in tap water (Nayar et al. 1979). The Bti formulations were suspended in distilled water by using a commercial blender for 1 to 2 min to make a 1% stock suspension (wtfvol) of each formulation. The suspension was maintained with a magnetic stirrer for making subsequent serial dilutions and transfers for treatments. Each formulation was tested on 3 or more different occasions. Four or 5 triplicate concentrations of a formulation were applied each time and 3 cups were left untreated as con trols. Stock suspensions and their serial dilutions were freshly prepared on each occasion. After a 24-h exposure period in a controlled temperature (27 ± 1°C) holding room, larval mortality was recorded. The corrected mortal ity of a species at different concentrations of a formulation was subjected to log-probit regression analysis.

RESULTS AND DISCUSSION

Larvae of Ae. taeniorhynchus were susceptible (LC90 = 0.114 to 0.772 ppm) to the 3 test formulations of Bti (Table 1). The activity of these formulations against Ae. taeniorhynchus did not correspond with their rela tive potencies. For example, Bactimos® was 2.33X more potent than Teknar® and 1.75X more potent than Vectobac®, but showed 6.77X and 2.98X better activity against Ae. taeniorhynchus than Teknar® and Vectobac®, respec tively. Similarly, Vectobac® was 1.33X more potent than Teknar® but showed 2.27X better activity than Teknar® against Ae. taeniarhynchus. Among the species of Culex, Cx. nigripalpus was most susceptible to Bti, followed by Cx. salinarius, Cx. quinquefasciatus, and Cx. restuans. The LC90 values of Bactimos® against these Culex species ranged from 0.117 to 0.552 ppm, while those of Vectobac® and Teknar® varied from 0.234 to 1.192 ppm and 0.265 to 1.721 ppm, respectively. All 3 formulations of Bti were relatively ineffective against An. albimanus with LC90 values ranging from 2.683 ppm (Bactimos®) to 11.446 ppm (Vec tobac®). Similarly, Wy. vanduzeei also remained relatively unaffected; the

LC90 value of Teknar® against this species was 4.706 ppm. There is no previous laboratory work on the activity of Bti against larvae of Cx. nigripalpus, Cx. salinarius, and Wy. vanduzeei. However, the levels of larvicidal activity of the 3 formulations of Bti against Ae. taeniorhynchus and ex. quinquefasciatus observed in this study are generally similar to the activity of comparable Bti formulations tested against the same species in earlier investigations (Ali et al. 1981, Mulla et al. 1982, Purcell 1981, Van f-" TABLE 1. SUSCEPTIBILITY OF LABORATORY REARED MOSQUITO LARVAE" (LATE 3RD AND EARLY 4TH INSTARS) TO TWO WETTABLE ~ POWDERS (BACTIMOS® AND VECTOBAC®) AND ONE FLOWABLE CONCENTRATE (TEKNAR®), INDUSTRIAL FORMULATIONS 0") OF Bacillus thuringiensis SEROVAR. israelensis IN THE LABORATORY, 1981-82.

24-h lethal concentration (ppm) b b 2 Species Formulations LC50 C.L. LC BO C.L. Slope R

.Iiedes taeniorhynchus Bactimos 0.059 0.054-0.064 0.114 0.094- 0.134 5.59 0.92 ~ ".... Vectobac 0.131 0.112-0.149 0.340 0.282- 0.398 3.20 0.98 C Teknar Vectobac 0.192 0.169-0.214 0.645 0.534- 0.755 2.49 0.97 <:"i- 0") Teknar 0.318 0.216-0.421 1.062 0.698- 1.426 2.94 0.86 -.;'J ..-.. Culex restuans Bactimos 0.096 0.077-0.115 0.307 0.248- 0.367 4.15 0.86 !':l Vectobac 0.362 0.315-0.408 1.192 1.041- 1.343 2.51 0.96 -- Teknar 0.812 0.762-0.863 1.721 1.478- 1.964 4.12 0.99 Culex salinarius Bactimos 0.059 0.044-0.073 0.169 0.126- 0.211 3.14 0.95 Vectobac 0.129 0.120-0.137 0.286 0.242- 0.330 3.97 0.87 Teknar 0.138 0.117-0.158 0.265 0.244- 0.286 4.76 0.93 Wyeomyia vanduzeei Bactimos 0.348 0.280-0.416 0.916 0.813- 1.020 3.04 0.97 s::~ Vectobac 0.571 0.507-0.634 1.201 0.965- 1.437 4.54 0.96 l::i Teknar 1.294 1.148-1.440 4.706 3.174- 6.239 2.84 0.93 SO f-" ~ "Maintained at the Florida Medical Entomology Laboratory at Vero Beach, FL. 00 b95% confidence limits. Jo!:>. Ali et al.: Pathogenicity of Bt israelensis 197

Essen and Hembree 1980). The LC90 value of the 2 species for any com parable formulation was <1.0 ppm (Mulla et al. 1982). It is obvious that An. albimanus and Wy. vanduzeei were relatively less susceptible to Bti under laboratory conditions. While Wy. vanduzeei may have inherent tolerance to Bti, the tolerance of Anopheles is probably due to its surface feeding habit (Mulla et al. 1982, Nugud and White 1982). Float ing formulations of Bti might be more effective against Anopheles larvae. From this study, it is evident that Bti offers a good potential for the con trol of Ae. taeniorhynchus, Cx. nigripalpus, Cx. quinquefasciatus, Cx. salinarius, and Cx. restuans. Among the formulations tested under laboratory conditions, Bactimos® invariably was the most effective material against these species and Teknar® the least when their relative potencies are taken into account. Florida Agricultural Experiment Stations Journal Series No. 4777.

REFERENCES CITED ALI, A., R. D. BAGGS, AND J. P. STEWART. 1981. Susceptibility of some Florida chironomids and mosquitoes to various formulations of Bacillus thuringiensis serovar. israelensis. J. Econ. Ent. 74: 672-7. ANONYMOUS. 1982. Data sheet on the biological control agent Bacillus thuringiensis serotype H-14 (de Barjac 1978). WHO/VBC/79.750, Rev. 1,46 p. DE BARJAC, H. 1978. Une nouvelle variete de Bacillus thuringiensis tres toxique pour les moustiques: B. thuringiensis var. israelensis serotype 14. C. R. Acad. Sci. (Paris) 286D: 797-800. GOLDBERG, L. J., AND J. MARGALIT. 1977. A bacterial spore demonstrating rapid larvicidal activity against Anopheles sergentii, Uranotaenia unguiculata, Culex univitattus, Aedes aegypti, and Culex pipiens. Mosq. News 37: 355-8. MULLA, M. S., B. A. FEDERICI, AND H. A. DARWAZEH. 1982. Larvicidal efficacy of Bacillus thuringiensis serotype H-14 against stagnant water mosquitoes and its effects on nontarget organisms. Environ. Ent. 11: 788-95. NAYAR, J. K., P. A. PIERCE, AND J. S. HAEGER. 1979. Autogeny in Wyomyia vanduzeei in Florida. Entomologia expo appl. 25: 311-6. NUGUD, A., AND G. WHITE. 1982. Evaluation of Bacillus thuringiensis sero type H-14 formulations as larvicides for Anopheles arabiensis (species B of the An. gambiae complex). Mosq. News 42: 36-40. PURCELL, B. 1981. Effects of Bacillus thuringiensis var. israelensis on Aedes taeniorhynchus and some nontarget organisms in the salt marsh. Mosq. News 41: 476-84. VAN ESSEN, F. W., AND S. C. HEMBREE. 1980. Laboratory bioassay of Bacillus thuringiensis against all instars of Aedes aegypti and Aedes taeniorhynchus larvae. Mosq. News 40: 424-31. 198 Florida Entomologist 67 (2) June, 1984

NEW SPECIES OF ERIOPHYID MITES (ACARI: ERIOPHYOIDEA) JAN BOCZEKl AND ROBERT DAVIS Stored-Product Insects Research and Development Laboratory Agricultural Research Service, USDA Savannah, Georgia 31403 USA

ABSTRACT Nine species of eriophyid mites are described, 3 from Poland: Aceria, ma,lva,cea,rum n.sp., Aculus ma,lvae n.sp., and Epitrimerus tanaceti n.sp.; 5 from Brazil: Aculus pita,ngae n.sp., A. ca,ulifiorus n.sp., A. ca,ta,ppae n.sp., A. sola,ni n.sp., and Phyllocoptes casea,ria,e n.sp.; and, 1 from Colombia: Ca,la,Ca,ruB guerreroi n.sp.

RESUMEN Son descritas nueve especies de acaros eriofidos, tres de las cuales fueron colectadas en Polonia: Aceria malvacearum n. sp.; Aculus malva,e n. sp.; Epitrimerus ta,na,ceti n. sp.; cinco en Brasil: Aculus pitanga,e n. sp.; A. caulifiorus n. sp.; A. ca,tappa,e n. sp.; A. solani n. sp. y Phyllocoptes ca,seariae n. sp.; y una en Colombia: Calaca,rus guerreroi n. sp.

The eriophyids (Eriophyoidea) are undoubtedly the most numerous of the plant feeding mites although, at present, only 1859 species are known (Davis et al., 1982). Nine new species of eriophyid mites are described in this paper, 3 from Poland: Aceria, ma,lvacearum n.sp., Aculus malvae n.sp., and Epitrimerus tana,ceti n.sp.; 5 from Brazil: Aculus pitrangae n.sp., A. ca,ulifiorus n.sp., A. ca,ta,ppae n.sp., A. solani n.sp., and Phyllocoptes caseariae n.sp.; and, 1 from Colombia: Ca,lacarus guerreroi n.sp. Type materials are deposited at the Department of Applied Entomology, Agricultural University of Warsaw, Warsaw, Poland.

Aceria malvaceMum Boczek and Davis, NEW SPECIES (Fig. 1)

FEMALE: 258 p,m (range of 8 specimens 204-301 p,m) long; 56 p,m wide; wormlike; yellowish. Rostrum 19 p,m long; rostral seta 10 p,m long; chelicerae 18 p,m long. Dorsal shield 38 p,m long, 40 p,m wide, without lobe over rostrum, with short median, 1 admedian line and 3 submedian lines on each side. Dorsal tubercles on rear shield margin; 26 p,m apart; with dorsal setae 44 p,m long, reclinate and diverging. Foreleg 66 p,m long, tibia 8 p,m long; claw 10 p,m long; feather-claw 9 p,m long, 8-9 rayed. Hindleg 50 p,m long; hind ,coxal setae 28 p,m apart. First pair of coxae with longitudinal ornamenta tion. Thanosome with about 65 microtuberculate rings; microtubercles tri angular. Lateral setae 40 p,m long, on sternite 13; 1st ventral setae 56 p,m long, on sternite 26; 2nd ventral setae 30 p,m long; caudal setae 41 p,m long;

'Present Address: Department of Applied Entomology, Agricultural University of Warsaw. Nowoursynowska 166. Poland. Boczek & Davis: New Eriophyids 199

~ 1 AP

Fig. 1. Aceria malvacearum n. sp.: AP-internal female genitalia; C claws; D-dorsal view; DA-anterior dorsal view; GF-external female genitalia; and SA-anterior lateral view. accessory setae 3 p,m long. Female genital coverflap 20 p,m long, 24 p,m wide; with about 14 longitudinal striae; genital setae 23 p,m apart, 18 p,m long. MALE: Unknown. NYMPH II: 165 p,m long; shield 32 .am long; dorsal setae 20 p,m long; dorsal tubercles 20 .am apart, abdomen with 63 rings. HOST: Malva sylvestris L. (Malvaceae). RELATION TO HOST: Vagrant on undersurface of the leaves. 200 Florida Entomologist 67 (2) June, 1984

TYPE MATERIAL. Holotype: female on slide, Poland, Warsaw-Lazienki, 22-VI-1982, D. Zalewska. Paratypes (7): females on slides, same data as holotype. This species is close to Aceria elacanthi Keifer and Can be distinguished by the shield pattern, size and location of the dorsal tubercles and the shape of the featherclaw. The dorsal shield A. elacanthi is subtriangular, with a broken median line and one pair of admedian lines; the dorsal tubercles are large, situated slightly ahead of rear shield margin; the featherclaw is 4-rayed. In the new species dorsal shield is rhomboidal, with a complete median line and 3 pairs of admedian lines; the dorsal tubercles are small, and on rear shield margin; the featherclaw is 8-9 rayed.

Aculus malvae Roczek and Davis, NEW SPECIES (Fig. 2)

FEMALE: 186 ~m (range of 12 specimens 180-203 ~m) long; 82 ~m wide; fusiform; yellowish-white. Rostrum 24 ,urn long; curved down; rostral seta 6 ~m long. Shield 49 ~m long; 75 ~m wide, with broad anteriorly rounded lobe over rostrum, with minute granules; shield pattern of only 2 admedian lines curving laterad at both ends. Dorsal tubercles on rear margin of shield; 45 ~m (32-51 ~m) apart; dorsal setae 18 ~m long, reclinate and diverging. Foreleg 57 pm long, with granulated tibiae and tarsi j tibia 11 /lm long, claw 6 ~m long, knobbed featherclaw 5-rayed, 5 ~m long. Hindleg 54 ~m long, tibia 9 #LID long. Both pairs of coxae ornamented. Hind coxal setae 26 pm apart. Thanosome with about 40 dorsal rings, generally microtuberculate; and about 78 sternites with mictrotubercIes. Tergal microtubercles elliptical, ventral ones almost rounded. Lateral setae 32 pm long, on sternite 56. Telosome with 5 rings; setae 30 pID long; microtubercles elongated; caudal setae 47 pm long; accessory setae 2 pm long. Female genital coverflap 12 .urn long, 24 pm wide with 12 longitudinal striae; genital setae 17 ,urn apart, 20.am long. MALE: 179 ~m long; shield 48 ~m long; J(enitalia 23 ~m wide. NYMPH II: 154 ~m long, shield 43 ~m long; dorsal setae 12 ~m long, 30 pm apart; abdomen with 54 microtuberculate tergites. HOST: Malva sylvestris L. (Malvaceae). RELATION TO HOST: Vagrant on undersurface of the leaves. TYPE MATERIAL. Holotype: Female on slide, Poland, Warsaw-Lazienki, 22-VI-1982, D. Zalewska. Paratypes (6): females (5) and male (1) on slides, same data as holotype. This species is close to Aculus fockeui (Nalepa and Troussart) and can be distinguished by the presence of shield spines, the shape of female genitalia and the featherclaw. "In A. fockeui the shield lobe has 2 distinct spines; the female genitalia has 16-18 longitudinal striae; and the feather claw is 4-rayed. In the new species the shield lobe is without spines; the female genitalia has 12 striae; and the featherclaw is 5-rayed.

Epitrimerus tanaceti Boczek and Davis, NEW SPECIES (Fig. 3)

FEMALE: 172 ~m (range of 11 specimens 172-203 ~m) long; 63 ~m wide; straw-yellow; fusiform. Rostrum 17 ,urn long; chelicerae 20 flm long. Dorsal shield 50 ~m long; 56 ~m wide. Shield smooth, only laterally with some Boczek & Davis: New Eriophyids 201

Fig. 2. Aculus mal'L'ae n. sp.: AP-internal female genitalia; C-clawsj DA-anterior dorsal view; GF-external female genitalia; GM-external male genitalia; S-lateral view; and SA-anterior lateral view. granulation. Dorsal tubercles on rear margin of shield; 20 .urn apart, with dorsal setae 8 .urn long, pointed vertically and curved centrally. Foreleg 45 .urn long; tibia 7 .urn long; claw 6 .urn long, knobbed; featherclaw 4-rayed, 5 ,urn long. Hindleg 40 ,urn long. Both pairs of coxae with linear troughs. Hind coxal setae 23 ,urn apart. Thanosome with 36 (36-41) tergites and about 70 sternites. First 3 tergites covered with delicate elongated microtubercles, 202 Florida Entomologist 67 (2) June, 1984 .~ AP

Fig. 3. Epitrimerus tanaceti n. sp.: AP-internal female genitalia; C claws; D-dorsal view; DA-anterior dorsal view; ES-lateral view of tergite-sternite juncture; GF-external female genitalia; and SA-anterior lateral view.

further tergites have such microtubercles only in central and lateral parts forming dorsally 3 longitudinal bands. Lateral setae 20 J.!m long on sternite 14; 1st ventral setae 46 J.!m long, on sternite 31; 2nd ventral setae 14 J.!m long, on sternite 49. Telosome with 5 microtuberculate rings and 22 J.!m long setae. Caudal setae 22 J.!m long; accessory setae 3 J.!m long. Female genital Boczek & Davis: New Eriophyids 203 coverfiap 13 Itm long, 24 Itm wide, with 12 longitudinal striae; genital setae 14 Itm apart, 25 Itm long. MALE: Unknown. NYMPH II: 105 Itm long; with microtuberculate rings. HOST: Tanacetum vulgare L. (Compositae). RELATION TO HOST: Vagrant on undersurface of the leaves. TYPE MATERIAL. Holotype: Female on slide, Poland, Warsaw-Powsin Park, 13-VIII-1982, D. Zalewska. Paratypes (3): Females on slides, same data as holotype. This species is close to Epitrimerus jaceae Liro and can be distinguished by the shape and pattern of the dorsal shield and the tergites. In E. jaceae the dorsal shield is longer than wide, with a few broken admedian lines; the tergites are smooth. In the new species, the dorsal shield is as long as wide with one short admedian line and some granulations laterally; the tergites have microtubercles in central and lateral regions forming a central longi tudinal band.

Aculus pitangae Boczek and Davis, NEW SPECIES (Fig. 4)

FEMALE: 185 Itm (range of 23 specimens 166-191Itm) long; 60 Itm wide; beige; fusiform. Rostrum 18 Itm long; rostral seta 6 Itm long. Shield sub triangular, with lobe over rostrum, 42 Itm (36-44 Itm) long, 65 Itm wide. Shield pattern with 2 distinct admedians, submedians forming a ridge which parallels shield margin, beginning and ending at the base of dorsal tubercles. Dorsal tubercles on rear margin of shield: 41 Itm apart, dorsal setae 12 Itm long, reclinate and slightly diverging. Foreleg 48 Itm long; genu 5 Itm long with setae 22 Itm long; tibia 7 Itm long, with seta 2 Itm long. Claw 6 Itm long, knobbed; featherclaw 4-rayed, 6 Itm long. Hindleg 42 Itm long, genu 4 Itm long, tibia 5 Itm long. Coxae ornamented, with small lines; hind coxae with dots; with setae 22 Itm apart. Thanosome with 29 (26-34) tergites and 63 sternites. Tergites with 4 longitudinal bands of microtubercles reaching telosome. Sternites with oval, minute microtubercles. Lateral setae 12 Itm long, on sternite 15; 1st ventral setae 44 Itm long, on sternite 30; 2nd ventral setae 11ltm long, on sternite 48. Telosome with 6 rings and 16 Itm long setae. Caudal setae 60 Itm long; accessory setae absent. Genital coverfiap 11 Itm (10-12 Itm) long; 18 Itm (16-19 Itm) wide; with 12 longitudinal striae; genital setae 12 Itm apart, 12 Itm long. MALE: Unknown. HOST: Eugenia unifiora L. (Myrtaceae). RELATION TO HOST: Vagrants on undersurface of the leaves. TYPE MATERIAL: Female on slide, Brazil, Piracicaba, Sao Paulo, 19-IX 1980, J. Boczek. Paratypes (22): Females on slides, same data as holotype. This species is close to Aculops eugeniae Keifer (1977) described from the same host plant from Florida. However, it differs in the shield pattern, in the tegrite microtubercles, and from the n. sp. in its relationship to the host. In A. eugeniae the shield has a broken median and 4 curved admedian lines; the tergites have microtubercles their entire length; and the mite causes bead galls on the upper surface of leaves. The new species has a dorsal shield with 2 pairs of admedian lines only; the tergites have microtubercles 204 FloridrL Entomologist 67 (2) June, 1984

Fig. 4. Aculus pitangae n. sp.: AP-internal female genitalia; C-claws; DA-anterior dorsal view; ES-lateral view of tergite-sternite juncture; GF-external female genitalia; S-lateral view; and SA-anterior lateral view. Boczek & Davis: New Eriophyids 205 forming 3 longitudinal bands; and they are vagrants on undersurface of the leaves not causing any apparent damage.

Aculus cauliflorus Boczek and Davis, NEW SPECIES (Fig. 5)

FEMALE: 193.um (range of 15 specimens 166-197 .urn) long; 65 .urn wide, yellowish; fusiform. Chelicerae 16 .urn long. Dorsal shield subtriangular with rounded lobe over rostrum; with dorsal tubercles 32 .urn apart, situated on rear margin of shield. Dorsal setae 10 pm long, reclinate and diverging. Dorsal shield with 2 indistinct broken submedian lines. Thanosome with 18 smooth tergites and about 70 microtuberculate sternites. Microtubercles oval.

Fig. 5. Aculus cauliflorus n. sp.: AP-internal female genitalia; C claws; D-dorsal view; DA-anterior dorsal view; ES-Iateral view of tergite-sternite juncture; GF-external female genitalia; GM-external male genitalia; and SA-anterior lateral view. 206 Florida Entomologist 67 (2) June, 1~84

Lateral setae 27 /.tm long on sternite 12; 1st ventral 56 pm long, on sternik 30; 2nd ventral 8 I'm long on sternite 49. Caudal setae 48 I'm long; accessory setae 4 pm long. Forelegs 50 ,am long; genu 6 .am long with seta 19 .am long; tibia 9 I'm long with seta 4 I'm long; claw 6 I'm long, knobbed; featherclaw 4-rayed. Hindleg 45 I'm long; genu 6 I'm long with seta 6 I'm long. Telosome with 5 rings and 20 I'm long setae. Female genital coverflap 16 I'm (16-20 I'm) long, 22 I'm (20-24 I'm) wide, with 12 longitudinal striae; genital setae 16 I'm apart, 12 I'm long. Coxae slightly granulated; hind coxal setae 23 I'm apart. MALE: 154 I'm long; shield 40 I'm long; dorsal setae 6 I'm long; abdomen with 23 tergites; genitalia 15 I'm wide; genital setae 14 I'm apart. HOST: AcnistU8 cauliflorus Schott. (Solanaceae). RELATION TO HOST: Vagrants on undersurface of the leaves. TYPE MATERIAL. Holotype: Female on slide, Brazil, Piracicaba, Sao Paulo, 19-IX-1980, J. Boczek. Paratypes (14): Females (13) and male (1) on slides, same data as holotype. This species is close to Aculus schlechtendali (Nalepa) and can be dis tinguished by the shape and the pattern of dorsal shield, presence of micro tubercles on tergites and the shape of female genital coverflap. In A. schlechtendali the dorsal shield is semicircular, with median and some admedian lines j the tergites are smooth; and the female genital cQverflap has striae in one rank. In the new species the dorsal shield is subtriangular with one pair of broken admedian lines; the tergites have microtubercIes laterally; and the female genital coverflap has striae in 2 separate ranks.

Aculu8 catappae Boczek and Davis, NEW SPECIES (Fig. 6)

FEMALE: 166 I'm (range of 15 specimens 151-196 I'm) long; 62 I'm wide, amber; fusiform. Rostrum 20 .urn long, with seta 7 /LID long; chelicerae 16 jtm long. Shield 41 I'm (32-42 I'm) long, 48 ~m wide. Dorsal shield with median line fragmented and largely indistinct on posterior 1/3 of shield. Admedian lines complete in most specimens 'and forming a circular pattern on posterior 1/3 of shield. First and 2nd submedian lines slightly fragmented but dis tinguishable from anterior margin to dorsal tubercles. Additional submedian lines fragmented and presenting appearance of a hairy pattern on shield margins. Dorsal tubercles 27 I'm apart, on rear margin of shield; dorsal setae 12 I'm long, pointed reclinate and diverging. Foreleg 54 I'm long; genu 6 I'm long with seta 21 I'm long; tibia 10 I'm long with seta 5 I'm long; claw 7 I'm long, knobbed featherclaw 7-rayed. Hindleg 51 I'm long. Thanosome with 52 (39-54) tergites and about 80 sternites. Both tergites and sternites microtuberculate; microtubercles elliptical. Lateral setae 22 ,urn long, on sternite 13; 1st ventral setae 64 I'm long, on sternite 27; 2nd ventral setae 16 I'm long, on sternite 48. Telosorne with 6 rings and 32 I'm long setae. Caudal setae 80 ,um long; accessory setae 4 jtm long. Female genital cover flap 13 I'm (10-13 I'm) long, 21 I'm (16-24 I'm) wide, with about 12 longi tudinal ribs; genital seta 15 I'm apart, 20 I'm long. MALE: 151 I'm long; shield 36 I'm long. HOST: Terminalia catappa L. (Combretaceae) RELATION TO HOST: Vagrant On undersurface of the leaves. TyPE MATERIAL. Holotype: Female on slide, Brazil, Piracicaba, Sao Paulo, Roczek & Davis: New Eriophyids 207

Fig. 6. Aculus catappae n. sp.: AP-internal female genitalia; C-claws; DA-anterior dorsal view; ES-lateral view of tergite-sternite juncture; GF--external female genitalia; S-lateral view; and SA-anterior lateral view. 19-IX-1980, J. Boezek. Paratypes (12): Females (10) and males (2) on slides, same data as holotype. This species resembles Acerimina terminaliae Keifer, and can be dis tinguished by the number of eoxal setae, featherelaw rays and shape of 208 Florida Entomologist 67 (2) June, 1984

genital coverflap. In A. terminaliae there is only one pair of forecoxal setae; the featherclaw is 6-rayed; and the female genital coverflap has diagonally lateral lines basally and is somewhat lobed laterally. In the new species there are 2 pairs of forecoxal setae; the featherclaw is 7-rayed; and the genital coverflap has striae of even length, and no lateral lobes.

Aculus solani Boczek and Davis, NEW SPECIES (Fig. 7)

FEMALE: 231 ~m (range of 21 specimens 170-237 ~m) long; 56 ~m wide, amber; spindleforrn. Chelicerae 15 ,am long. Dorsal shield subtriangular, 42 ~m (40-44 ~m) long, with median line visible as 2 short fragments on pos terior 1/2 of shield. Medial, admedial, and submedial lines complete on anterior 1/4 of shield. A ridge formed by portions of the submedial lines, beginning and ending at the base of the dorsal tubercles, parallels the shield margin. Dorsal tubercles 40 .am apart, on rear margin of shield; dorsal setae 14,um long; reclinate and diverging. Forelegs 36 .am longj femur 10 ~m long; tibia 10 ~m long with setae 7 ~m long; claw 8 ~m long, slightly knobbed; featherclaw 6 ~m long, 4 rayed. Hind legs 32 ~m long. Forecoxae with 1st setae 12 ~m long; 2nd setae 32 ~m long; hindcoxal setae 47 I'm long. Thanosome with 22 (22-24) smooth tergites and about 64 sternites. Sternites with oval microtubercles. Lateral setae 20 jLm long, on sternite 6; 1st lateral setae 85 ~m long, on sternite 26; 2nd lateral 22 ~m long, on sternite 42. Telosome with 5 microtuberculate rings and setae 30 ~m long. Female genital coverflap 18 p.m long, 24 ,urn wide, with 16 furrows; genital setae 22 ,urn apart, 13 ~m long. MALE: 200 p.m long; shield 44 /Lllllong, wiLh dvr;:;al seLae 14 ",m lung, with 28 tergites. Male genitalia 16 ~m wide. HOST: Solanum nigrum L. (Solanaceae). RELATION TO HOST: Vagrant on both surfaces of the leaves. TYPE MATERIAL. H o1otype: Female on slide, Brazil, Piracicaba, Sao Paulo, 18-IX-1980, J. Boczek. Paratypes (13): Females (11) and males (2) on slides, same data as holotype. This species is close to Aculus rhamnivagrans (Keifer), and can be dis tinguished by the shield shape and the internal female genitalia. In A. rhamnivagrans the shield is subtriangular with a long and wide lobe over rostrum; the spermathecae are circular. In the new species the shield is semicircular with a very broad, short lobe over rostrum; the spermathecae are oval.

Phyllocoptes caseariae Boczek and Davis, NEW SPECIES (Fig. 8)

FEMALE: 190 ~m (range of 16 specimens 180-220 ~m) long; 67 ~m wide; 72 I'm thick; amber; spindleform. Rostrum 24 ~m long, curved downward obliquely. Chelicerae 17 ~m long. Shield 45 I'm (44-48 ~m) long, with short lobe over rostrum; shield oval, surface with two declivities along median line at anterior of shield and near its midpoint. Admedial lines visible sur rounding both declivities. Submedial lines extending from anterior margin to base of dorsal tubercles. Overall surface appearance is that of several fractured irregular placed rectangles. Dorsal tubercles 30 (29-32 ~m) Boczek & Davis: New Eriophyids 209

J) ) DA #!ff {f ES

Fig. 7. Aculus solani n. sp.: AP-internal female genitalia; C-claws; DA-anterior dorsal view; ES-lateral view of tergite-sternite juncture; GF-external female genitalia; GM-external male genitalia; S-lateral view; and SA-anterior lateral view. 210 Florida Entomologist 67 (2) June, 1984

Fig. 8. Phyllocoptes caseariae n. sp.: AP-internal female genitalia; C claws; DA-anterior dorsal view; ES-Iateral view of tergite-sternite junc ture; FG-external female genitalia; GM-external male genitalia; S lateral view; and SA-anterior lateral view. apart, 6 ~m long, 11 ~m ahead of rear margin of shield, erect and convergent. Foreleg 50 I'm long; femur 9 I'm long, genu 6 I'm long with setae 32 I'm long, tibia 10 ,um long with setae 5 ,urn long; featherclaw 5·rayed, 6 p:m long, claw knobbed. Hindlegs 41 I'm long; tibia 8 I'm long; tarsus 6 I'm long. Thanosome Boczek & Davis: New Eriophyids 211 with 35 (31-36) broader tergites and about 59 sternites. Lateral ends of tergites with spiny microtubercles and sternites with rounded microtubercles. Lateral setae 12 ,urn long on sternite 5; 1st ventral seta 44 ,urn long on sternite 37. Telosome with 5 rings and 20 ,urn long setae. Caudal setae about 64 ,urn long, accessory setae absent. Female genital coverflap 18 ,urn (18-24 ,urn) wide, with 14-16 elongate furrows; setae 22 ,urn apart, 12 ,urn long. MALE: 152 ,urn long; dorsal shield 40 ,urn long; dorsal setae 5 ,urn long, 24 ,urn apart; 8 ,urn from rear margin of shield; chelicerae 14 ,urn long; male genitalia 18 ,urn wide, genital setae 7 ,urn long. NYMPH II: 140 ,urn long; 56 ,urn thick; shield 30 ,uum long; dorsal setae 4 ,urn long; dorsal tubercles 9 ,urn apart, with 40 rings with spiny micro tubercles. HOST: Casearia sylvestris Sw. (Flacourtiaceae). RELATION TO HOST: Vagrants on undersurface of leaf. TYPE MATERIAL. H olotype: Female on slide, Brazil, Piracicaba, Sao Paulo, 19-IX-1980, J. Boczek. Paratypes (14) : Females (13) and male (1) on slides, same data as holotype. This species is close to Phyllocoptes abaenus Keifer and can be distin guished by the shield shape, the shape of microtubercles of the tergites and by the number of featherclaw rays. In P. abaenus the shield has a long broad lobe; the sternites have oval microtubercules their entire length; and the featherclaw is 4-rayed. In the new species the shield is almost semicircular; the sternites have spiny microtubercles only laterally; and the featherclaw is 5-rayed. This is the first eriophyid mite described from the plant family Flacourtiaceae.

Calacarus guerreroi Boczek and Davis, NEW SPECIES (Fig. 9)

FEMALE: 196,um (range of 30 specimens 166-196 ,urn) long; 76 ,urn wide; straw yellow; fusiform. Rostrum 36 ,urn long; seta 20 ,urn long. Chelicerae 40 ,urn long. Dorsal shield 50 ,urn long; 72 ,urn wide. Shield pattern in form of network of ridges and granulation laterally. Shield lobe over rostrum wide, short, rounded anteriorly. Dorsal tubercles and setae absent. Foreleg 66 ,urn long, genu 5 ,urn long with seta 36 ,urn long; tibia 15 ,urn long, with seta 2 ,urn long. Claw 6,um long, knobbed featherclaw 6 ,urn long, 4-rayed. Hindleg 58 ,urn long; genu 4 ,urn long; tibia 12 ,urn long. Thanosome with 54 (49-64) tergites and about 97 sternites. Tergites forming central ridge and two adcentral ridges with wax points. Microtubercles present on all sternites and usually on some proximal tergites laterally. Telosome of 8 rings and setae 21 ,urn long; caudal setae 98 ,urn long; accessory setae absent. Telosomal micro tubercles elongated. Lateral setae 36 ,urn long, on sternite 19; 1st ventral setae 40 ,urn long, on sternite 47; 2nd ventral setae 36 ,urn long, on sternite 80. Female genital coverflap 22 ,urn long, 28 ,urn wide; epigynium with about 12 longitudinal striae; genital setae 18 ,urn apart, 12 ,urn long. MALE: 154 ,urn long; with dorsal shield 56 ,urn long; genitalia 18 ,urn wide. HOST: Manihot esculenta Crantz (Euphorbiaceae) (Cassava, Tapioca). RELATION TO HOST: On the uppersurface of leaf; they prefer the basal leaves of the plant. Infested leaves curl upward, and gradually shrink. TYPE MATERIAL. H olotype: Female on slide, Colombia, Cali, 2-VIII-1982, 212 Florida Entomologist 67 (2) June, 1984

Fig. 9. Calacarus guerreroi n. sp.: SP-internal female genitalia; C claws; DA-anterior dorsal view; FG-external female genitalia; GM-ex ternal male genitalia; S-lateral view; and SA-anterior lateral view.

J. M. Guerrero. Paratypes (20): Females (13) and males (2) on slides, same data as holotype. This species is close to Calacarus microrostrus Chakrabarti and Gosh and can be distinguished by the shape of female genitalia and the number of featherclaw rays. In C. microrostrus the genital overflap is smooth; the Boczek & Davis: New Eriophyids 213 featherclaw is 7-rayed. In the new species the genital coverflap has 12 longi tudinal striae; the featherclaw is 4-rayed. We are pleased to name this mite for Mr. Jose Maria Guerrero, Centro Internacional de Agricultura Tropical, Apartado Aereo 6713, Cali, Colombia. This is the first species of eriophyid mite found on cassava (Bellotti & Schoonhoven 1978), an economical subsistence crop plant grown in about 90 developing countries.

ACKNOWLEDGEMENT We would like to acknowledge the assistance of Dr. C. H. W. Flechtman, ESALQ Zoologia, Piracicaba SP, Brazil, in collecting the Brazilian mites; and of Miss Danuta Zalewska, laboratory assistant, Department of Applied Entomology, Agricultural University of Warsaw, Warsaw, Poland, for draw ing the mites.

REFERENCES CITED BELLOTTI, A., AND A. VAN SCHOON HOVEN. 1978. Mite and insect pests of cassava. Ann. Rev. Ent. 23: 39-67. DAVIS, R., C. H. W. FLECHTMAN, J. BOCZEK, AND H. BARKE. 1982. Catalogue of eriophyid mites (Acari: Eriophyoidea). Agric. Univ. Warsaw Press, Warsaw, Poland. 254 p. KEIFER, H. H. 1977. Eriophyid studies C-13. California Dept. of Agric., Jan. 20,1977.24 p. .. . • •• • • • • • •• . ~ A SYNTHESIS OF TEMPERATURE DEPENDENT DEVELOPMENTAL STUDIES WITH THE CITRUS RED MITE, PANONYCHUS CITRI (McGREGOR) (ACARI: TETRANYCHIDAE) V. P. JONES AND J. G. MORSE Dept. of Entomology University of California Riverside, CA 92521 USA

ABSTRACT A literature review revealed 12 temperature-humidity studies performed on the citrus red mite, Panonychus citri (McGregor) (Acari: Tetranychi dae). An analysis of these studies using a heat unit scale indicated that they were in surprisingly close agreement especially in view of the different host plants and experimental techniques utilized. Theoretical thresholds for de velopment, degree days necessary to complete development, and temperature dependent mortality for the egg, immature and adult female stages are presented. This synthesis implies that further research with the citrus red mite is needed in the area of variable temperature developmental rates, fecundity and mortality rates.

RESUMEN Una revision de la bibliografia revelo 12 estudios de temperatura- Boczek & Davis: New Eriophyids 213 featherclaw is 7-rayed. In the new species the genital coverflap has 12 longi tudinal striae; the featherclaw is 4-rayed. We are pleased to name this mite for Mr. Jose Maria Guerrero, Centro Internacional de Agricultura Tropical, Apartado Aereo 6713, Cali, Colombia. This is the first species of eriophyid mite found on cassava (Bellotti & Schoonhoven 1978), an economical subsistence crop plant grown in about 90 developing countries.

RESUMEN Una revision de la bibliografia revelo 12 estudios de temperatura- 214 Florida Entomologist 67 (2) June, 1984 humedad sobre Panonychu8 citri (McGregor) (Acari: Tetranychidae). Un amilisis de estos estudios usando una unidad de calor de escala indico que estaban en sorprendente acuerdo, especialmente en vista de las diferentes plantas hospederas y tecnicas experimentales utilizadas. Limite teorico de desarrollo, grados diarios necesarios para completar el desarrollo, mortalidad de los huevos dependiente de la temperatura, y las etapas de los inmaduros y las hembras adultas son presentados. Esta sintesis implica que mas in vestigaci6n sobre P. citri es necesaria en el area de grados de desarrollos por temperatura variable, fecundidad, y grados de mortalidad.

The citrus red mite, Panonychu8 citri (McGregor), is one of the most important pests of citrus worldwide (Jeppson et al. 1975). In California alone, the citrus red mite was estimated to cause an economic crop loss of approximately 15.9 million dollars in 1977 (Buxton 1977). Its importance has led to a number of studies dealing with the relationship between tem perature and the rate of development (Table 1). However, none of these studies have defined parameters, such as degree days or developmental thresholds, which are necessary for temperature dependent modeling efforts. In this review, we provide a summary and analysis of these studies and define several modeling parameters from a selected subset of all work done.

MATERIALS AND METHODS Data were initially grouped into 3 categories which corresponded to the egg, immature, and adult life stages. The data were summarized (Table 2) and experimental methods reviewed. Experiments were eliminated which utilized sub-optimal host plants, such as papaya (Maity and Chakrabarti 1978), or involved temperature regimes which were not sufficiently defined for this type of analysis (English and Turnipseed 1941). Data from studies marked with an * in Ta.ble 2 were not used in the analysis. Since there was apparently no high temperature mediated decrease in developmental rates, data for t4e egg and immature stages were graphed as (developmental time) -1 vs. (temperature). Linear regression was used to obtain least squares estimates of the slope and x-intercept. Degree days for development and theoretical developmental thresholds were determined as (slope)-l and (x-intercept), respectively (Arnold 1959). If more than one humidity was tested per temperature, humidities which produced the fastest development time were selected. Only data for the female stage were considered.

RESULTS AND DISCUSSION High mortali,ty was found to occur in both the egg and immature stages at a constant temperature of ca. 30°C (23.7 and 20.6 degree days per day, respectively) (Fig. lA,B). Based primarily on the work of Keetch (1971), Munger (1963), and observations in greenhouse experiments where tem peratures peaked in excess of 30°C for short periods without population de crease (Jones 1983), we feel that a sharp increase in mortality occurs only after 1-3 days constant exposure to temperatures over 30°C. The 5 studies used for the egg development regression were in very good agreement (R2 of 0.91, Fig. lC). The developmental threshold was found to Jones & Morse: Developmental Studies on Citrus Red Mite 215

TABLE 1. TEMPERATURE STUDIES" WITH Panonychus citri.

Authors Host Comments

Prendergast (1938) Grapefruit Data incomplete. Mortality Citrus paradisi in all cases unacceptably high. English and Turnipseed Orange No raw data given. Mean (1941) Citrus sinensis temperature regressed against developmental thresholds with only summary equations presented. Fukuda and Shinkaji Tangerine Egg stage data only. 6 hu- (1954) Citrus reticulata midities tested per temperature. Shinkaji (1959) Tangerine Post-embryonic stage data Citrus reticulata only. 4 humidities tested per temperature Munger (1963) Lemon Data on high temperature Citrus limon mortality conflicts with other authors. Tanka and Inoue (1970) Citrus sp. Beavers and Hampton Lemon (1971) Citrus limon Keetch (1971) Lemon 3 humidities tested per tem- Citrus limon perature. Chakrabarti and Maity Papaya Unfavorable host plant. De- (1978) Carica papaya velopmental times don't differ significantly with temperature changes. Maity and Chakrabarti Papaya Same data, as Chakrabarti (1978) Carica papaya and Maity (1978) Yusuda (1978) Citrus sp. Saito (1979) Satsuma mandarin Citrus unshiu

nAIl studies except English and Turnipseed (1971) done at constant temperature. be 6.3°C with 116.9 degree days necessary for mean egg development. For immature development, the developmental threshold was 9.4°C with 96.0 degree days (R2=0.66, Fig. 1D) found to be the mean developmental period. For the preovipositional period and adults stage, linear regressions were run for both period duration vs temperature (R2=0.60, 0.64, respectively) and (period)-l vs temperatures (R2=0.45, 0.57). We chose to represent the data in terms of stage duration vs temperature although other factors may be important such as host plant differences or experimental conditions (note especially linear relationship between adult stage duration and temperature

TABLE: 2. (Continued) ~ <:> ;;Z a:. Developmental period (days) % Mortality i:I;) Im- Egg- Im- Eggs/ %% R'=> Author °C Egg mature Adult Adult Egg mature Female Female RH ~ <:> ~ a:. Tanka and Inoue '. (1970) tl 15 14.7 10.6 25.3 25.3 8.3 22.0 66.9 50-90 a:. - ~ 20 8.5 9.7 16.0 18.2 8.7 - 17.9 81.5 50-95 a:. ~ 25 6.7 4.2 10.2 10.9 7.4 - 17.9 66.9 60 <:> 5.3 3.4 6.0 8.7 4.9 19.8 53.6 40-85 ~ 30 - ~ 35 - - 4.6 - 100.0 - 3.4 - 50-70 a:. ;;Z Beavers and Hampton ~ ~ (1971) 26 5.3 5.8 11.5 11.1 6.0 10.0 28.8 52.5 52.5 ~ ~ ~ Keetch (1971) ~ 17.0 11.0 15.0 17.0 26.0 - - 42.3 - 60 ..,.~ 54.7 a:. 22.0 7.0 8.0 13.0 15.0 - - - 60 i:I;) 27.0 5.0 7.0 10.0 12.0 - - 51.3 - 60 <:> aMaityand Chakrabarti ~ (1978) (J..,. 23.6 14.0 9.2 14.0 15.0 17.5 17.9 30.0 66.7 64 ~ 26.7 13.4 9.0 13.4 14.5 10.5 25.0 29.0 66.7 51 g 30.6 12.5 8.8 12.5 14.1 8.3 33.1 36.5 66.7 49 ~ Yusuda (1978) a:. 25 - - - 10.9 - -- - .60-90 ~ Saito (1979) ..,.~ 25 5.8 6.1 23.0 12.0 36.9 70.8 ~ - - 52 a:.

t>:.l "Data not used in the analysis. ~ ~ TABLE 3. ADDITIONAL DE''ELOPMENTAL DATA BY LIFE STAGE FOR Panonych'llS citri. t.:> >-' 00 Period in days Proto- Deuto- Pre- Post- Author °C Larvae nymph nymph Nymph oviposition oviposition Shinkaji (1959) 20 3.6 3.3 3.9 7.2 3.4 22 3.3 2.9 3.0 5.9 2.6 - '""l 25 .2.7 2.3 2.4 4.7 2.0 - 0 28 2.1 2.2 1.9 4.1 2.1 - -j. 30 2.1 2.1 1.4 3.5 1.3 - .,... 32 1.6 1.1 1.4 2.5 1.3 - Tanka and Inoue (1970) ~,.... 15 3.5 - - 7.1 3.0 3.0 0 20 3.4 - - 6.2 2.6 - ~ 25 1.0 - - 3.1 1.2 0.8 0 30 1.3 - - 2.0 1.0 0.3 -0 35 - - - - 1.4 1.4 .,"'.,.... Keetch (1971) '" (j) 17 5.0 5.0 6.0 11.0 - - "'l ~ 22 3.0 3.0 2.0 5.0 - - t.:> 27 3.0 3.0 1.0 4.0 - - ~ Beavers and Hampton (1971) 26 1.9 1.6 2.3 3.9 1.0 Maity and Chakrabarti· (1978) 23.6 2.3 3.3 3.6 6.9 2.3 3.6 .... 26.7 2.3 3.1 3.6 6.7 2.2 3.4 ~ 30.6 2.2 3.2 3.4 6.6 2.1 3.0 ::l Saito (1979) .'" 25 1.9 1.9 2.3 4.2 1.5 2.1 >-' 00 aData not used in the analysis. '" "'" Jones & Morse: Developmental Studies on Citrus Red Mite 219

>- 100 t:100 >- • ~ ° S 80 ~ 80 0 « 0 ~ 60 260 o w 2 40 g:; 40 0 I- '"w 20 :i 20 ° ° ° 0 1ft " • ,,~(\ ~ 'b o 0~---'8~-~16--"b.J..I.qJ24?=--'~3f:.2---'4~0----''8 0 o 0!1 0. 8 16 24 32 40 A TEMPERATURE (OC) ~B TEMPERATURE (OC) « I- z .40 ;i .40 w_ R2, 0.66 I- - 27 ~~.32 <>. - 2>- ~« • g; ~24 ~~24 • 0' ~ ~ ...~O-Q- w ~~.16 ~ ~.16 OlJ a--~ W « ~O/ 0:« ~ '" .08 • .p ~ "'.08 ~e' 2 ~ 0 ~--7'~_---:'~--:'-:----:::':----::-'::--' ~ 0 O 8 16 24 32 40 0 8 16 24 32 40 C TEMPERATURE (OC) 0 TEMPERATURE (OC) ,v; !J 40 >~ 5 2 0 R , 0.60 :::>- w_0 ~ 4 g; 32 0: 0 <>'0 ~o ... - 3 'c5 e24 0'" W z<>. ., ° ~ ~ 16 2 1-« S?z ~ 1-0«- "'""'-0 ° • t;; 8 ",t: I ~~ :::> :::>. 0 8 16 24 32 40 E TEMPERATURE (OCl F TEMPERATURE toC) Fig. 1. Effects of temperature on various aspects of Panonychus citri development. a) Effect on egg mortality. b) Effect on mortality of the im mature stages. c) Effect on developmental rate of eggs. d) Effect on develop mental rate of immatures. e) Effect on the duration of the pre-ovipositional period of adult females. (6 lemon, 0 tangerine, • unspecified Citrus sp., A satsuma mandrin). when the data is separated by host plant and data source-Fig. IE and Table 2). Regressions for preovipositional period and adult stage duration vs tem- /\ /\ perature were y=4.76-0.llx and y=44.38-1.17x, respectively. Egg production showed no apparent relationship to female lifespan or temperature. The average egg production for data below 30°C was 38.0 eggs per female (S.E.=5.14) from all studies (because of the sharp decline in egg production at temperatures over 30°C, data taken over this temperature were eliminated from this part of the analysis). The sex ratio was in the range of 65-84% female in all cases except for the studies of Beavers and Hampton (1971) where the sex ratio was 52% female and Tanka and Inoue (1970) where sex ratio was 53.6% female at 30°C. The average sex ratio from all studies was 71 % female. Although it is difficult to interpolate from constant temperature labora tory studies to studies of population dynamics in the field, the correlation of population increases and decreases observed by Jeppson et al. (1957) and McMurtry (1969) can, at least partially, be explained by temperature medi ated changes in development and mortality. The purpose of this paper is not to provide an indepth analysis of temperature mediated changes in citrus red mite population dynamics, but rather to orient future research to areas 220 Florida Entomologist 67 (2) June, 1984

where data necessary for a modeling effort aimed at the citrus red mite is inadequate. Our analysis has shown that temperature dependent studies from around the world are in surprisingly close agreement and that adequate data are available on constant temperature developmental rates of the citrus red mite. Based on these results, future research projects are needed in the following areas: (1) comparison of constant temperature developmental rates with variable temperature rates, (2) determination of accurate mortality data under variable temperature regime, especially high temperature mortality,. and (3) influence of host plant nutrition on mite development.

ACKNOWLEDGMENTS The authors would like to thank Mr. Arata Iseki for translating several of the articles written in Japanese.

REFERENCES CITED ARNOLD, C. Y. 1959. The determination and signiflcance of the base tem perature in a linear heat unit system. American Soc. Hort. Sci. 74: 430-45. BEAVERS, J. B., AND R. B. HAMPTON. 1971. Growth, development, and mating behavior of the citrus red mite (Acarina: Tetranychidae). Ann. Ent. Soc. America 64: 804-6. BUXTON, G. M. 1977. Estimated damage and crop loss caused by insect and mite pests. California Dept. of Food and Agric. 12 p. CHAKRABARTI, S., AND S. P. MAlTY. 1978. Effect of temperature and relative humidity on the life cycle of Panonychus citri (McGregor) (Acarina: Tetranychidae). Sci. and Culture 44: 233-4. ENGLISH, L. L., AND G. F. TURNIPSEED. 1941. The influence of temperature and season on the citrus red mite (Paratetmnychus aitri). J. Agr. Res. 62: 65-77., FUKUDA, J., AND N. SHINKAJI. 1954. Experimental studies on the influence of temperature and rela.tive humidity upon the development of citrus red mite. (Metatetmnychus aitri McGregor). (1) On the influence of temperature and relative humidity upon the development of the eggs. Bull. Tokai-Kinki Agric. Exp. Sta. (Horticulture). 2: 160-71. JEPPSON, L. R., C. A. FLESCHNER, M. J. JESSER, AND J. O. COMPLIN. 1957. Influence of season and weather on citrus red mite populations on lemons in southern California. J. Econ. Ent. 50: 293-300. JEPPSON, L. R., H. H. KEIFER, AND E. W. BAKER. 1975. Mites injurious to economic plants. Univ. of California Press, Berkeley. 614 p. JONES, V. P. 1983. Pest resurgence of, and sampling plans for, the citrus red mite, Panonychus citri (McGregor) (Acari: Tetranychidae). Ph.D. Dissertation. Univ. of Calif., Riverside. 104 p. KEETCH, D. P, 1971. Ecology of the citrus red mite, Panonychus aitri (McGregor), (Acarina: Tetranychidae) in South Africa. 2. The in fluence of temperature and relative humidity on the development and life cycle. J. Ent. Soc. South Africa 34: 103-8. MAlTY, S. P., AND S. CHAKRABARTI. 1978. Biological studies on Panonychus aitri (Acari: Tetranychidae). Indian J. Acar. 2: 55-9. McMuRTRY, J. A. 1969. Biological control of citrus red mite in California. Proc. 1st International Citrus Symposium 2: 855-62. MUNGER, F. 1963. Factors affecting growth and multiplication of the citrus red mite, Panonychus aitri. Ann. Ent. Soc. America 56: 867-74. Gagne & Beavers: Contarinia from Slash Pine Shoots 221

PRENDERGAST, D. 1938. Studies of the biology of the citrus red mite, Paratetranychus citri (MeG.). Ph.D. Dissertation, Univ. California, Riverside. 65 p. SAITO, Y. 1979. Comparative studies on life histories of three species of spider mites (Acarina: Tetranychidae). App. Ent. Zool. 14: 83-94. SHINKAJI, N. 1959. Experimental studies on the temperature and relative humidity influencing the development of the citrus red mite (Panony chus [Metatranychus] citri) (McGregor). (2) On the influence of temperature and relative humidity on the post-embryonic development, adult life span, and egg-laying potency of the citrus red mite. Bull. Hort. Div. Tokai-Kinki Agr. Exp. Sta. 5: 129-42. TANKA, M., AND K. INOUE. 1970. Fundamental studies on the utilization of natural enemies in citrus groves in Japan. II. The method of predic tion of outbreaks of the citrus red mite, Panonychus citri (MeG.). Bull. Hort. Res. Sta., Japan, Ser. D. 6: 1-40. YUSUDA, M. 1978. Reproduction and factors influencing dispersal of the citrus red mite, Panonychus citri (McGregor) (Acarina: Tetrany chidae) under laboratory conditions. Japanese J. Appl. Ent. Zool. 22: 12-7. •••••••••••• • CONTARINIA SPP. (DIPTERA: CECIDOMYIIDAE) FROM SHOOTS OF SLASH PINE (PINUS ELLIOTTII ENGELM.) WITH THE DESCRIPTION OF A NEW SPECIES INJURIOUS TO NEEDLES RAYMOND J. GAGNE Systematic Entomology Laboratory, IIBIII, Agricultural Research Service USDA, c/o U.S. National Museum of Natural History Washington, DC 20560 AND GREGORY M. BEAVERS 3103 McCarty Hall University of Florida Gainesville, FL 32611

ABSTRACT A new species, Contarinia acuta Gagne (Diptera: Cecidomyiidae), which damages developing needles of slash pine, Pinus elliottii Engelm., in Florida, is described. It is distinguished from other North American species of Contarinia infesting pine, and diagnostic illustrations are provided for the larvae of the new species and those of 3 other, unnamed species of Contarinia associated with needles of slash pine.

RESUMEN Una nueva especie, Contarinia acuta Gagne (Diptera: Cecidomyiidae), que dana las hojas en desarrollo del pino Pinus elliottii Engelm. en la Flor ida, es descrita. Es distinguida de otras especies Norteamericanas de Con tarinia que infestan pinos, e ilustraciones diagn6sticas son proveidas para larvas de la nueva especie y de otras 3 especies sin nombrar de Contarinia asociadas con las hojas del pino P. elliottii. Gagne & Beavers: Contarinia from Slash Pine Shoots 221

During research into the cause of needle damage on slash pine, Pinus elliottii Engelm., 4 distinct species of Contarinia were found associated with new shoots. A characteristic needle damage is caused by Contarinia acuta Gagne n. sp. described in this paper. Larvae of the other 3 species are less common than C. acuta and their particular role on the needles is still un known. They were not reared to the adult stage, which is advisable before they can be named, but they are briefly described with illustrations so that they can be distinguished in future studies. Characteristic brown lesions (ca. 4 mm long) are formed on needles as a result of the feeding of C. acuta larvae. Feeding occurs inside the fascicle sheath of basally elongating needles (Fig. 1). After needle elongation has been completed, the needles become dry and brittle at the lesion site and frequently break (Fig. 2). Mature larvae drop to the ground and pupate in the surface litter. In northern Florida, larvae have been collected as early as mid-February, with 6 generations having been recorded in 1 year and up to 4 between May and September (Beavers, in preparation).

Contarinia acuta Gagne, NEW SPECIES Fig. 3-12

ADULT. Head. Eyes large, about 5 facets long at vertex, the facets circular, closely approximated except at midheight of eye where they lie slightly farther apart. Occiput rounded, without peak. Frontoclypeal setae 8-10. Labella hemispherical in frontal view with 4-5 setae. Palpus 4-seg mented. Male antennal flagellomere 3 (Fig. 3) binodal, bicircumfilar, the circumfilar loops not attaining the next distal node. Female antennal flagel lomeres 1-3 as in Fig. 4; circumfilar loops slightly bowed (Fig. 5). Thorax. Mesoscutal row with sparse setae in mostly single row inter spersed with scales. Mesanepisternum bare. Mesepimeron with 5 setae. Wing length male, 158-164 mm (162, avg. of 5), female, 173-200 mm (190 avg. of 5); RS slightly bowed apically, joining C behind wing apex. C broken, the wing margin indented at juncture with R5. Claws barely shorter than empodia. Male abdomen (Fig. 8). Tergites 1-7 rectangular with basal pair of trichoid sensilla, a caudal, single row of sparse setae, 2-4 lateral setae, and o scales; tergite 8 weakly sclerotized caudally but with at least 1-2 caudal setae laterally, 0 lateral setae, 2 basal trichoid sensilla. Sternites 2-6 rec tangular with pair of basal, closely approximated trichoid sensilla, a mostly single, caudal row of setae, and mixed setae and setiform scales grouped near midlength; sternite 7 as for preceding except caudal row mostly double; sternite 8 with midlength and caudal groups of setae continuous on caudal half of sclerite, and basal trichoid sensilla not as closely approximated as on sternites 2-7. Genitalia (Fig. 9 (dorsal) -10 (ventral): cerci short, broadly rounded with 2 caudoventral setae; hypoproct not deeply divided, the lobes broad, with 2 long setae. Aedeagus longer than hypoproct, very narrow, attenuate, and pointed at apex; gonocoxal apodeme broad; gonopod stout; gonostylus broadest near midlength, tapering to apex, setulose throughout. Female abdomen (Fig. 6). Tergites 1-7 and sternites 2-7 as in male but setae more numerous except for fewer laterals on tergites. Tergite 7 about .14 length distal section of ovipositor. Tergite 8 longer than 7 with pair of Gagne & Beavers: Contarinia from Slash Pine Shoots 223

Fig. 1-2. 1) Larva of C. acuta feeding on adaxial surface of needle re moved from fascicle. 2) Characteristic damage of C. acuta. 224 Florida Entomologist 67 (2) June, 1984

4 5

Fig. 3-7. Contarinia acuta: 3, 3rd () antennal flagellomere; 4, 1st through 3rd ~ flagellomeres; 5, 3rd ~ flagellomere; 6, ~ postabdomen; 7, ~ cerci (dorsolateral) . Gagne & Beavers: Contarinia from Slash Pine Shoots 225

() "} 0 0 o • o 0 o v o 0 .. 0

" 0 o o

/ ,,/ COO ? ; I o 0 I I o I I o I I I -_\_-~ 9 10 Fig. 8-10. Contarinia acuta: 8, (; abdominal segments 6-8; 9 (; terminalia (dorsal); 10, same (ventral). basal trichoid sensilla and short caudal setae. Cerci (Fig. 7) short, setulose throughout. LARVA (last instar). Length 2.3-2.7 mm (2.6 mm, avg. of 20). Body smooth except for anteroventral spicule fields on abdominal segments. Spatula (Fig. 11) clove-shaped, the anterior edge deeply incised, the lobes rounded. Papillae with short setae. Spiracles of abdominal segment 8 at posterior margin. Terminal papillae as in Fig. 12. HOLOTYPE, (;, from slash pine shoots, Gainesville, FL, 28-VII-1981, G. M. Beavers, deposited in Florida State Collection of Arthropods, Gainesville. Paratypes (all from Florida and shoots of slash pine unless otherwise noted 226 Florida Entomologist 67 (2) June, 1984 and deposited in Florida State Collection and U.S. National Museum of Natural History; GMB - G. M. Beavers; JLF ~ J. L. Foltz; RCW ~ R. C. Wilkinson): 6 ~, 1 ~, 3 pupal exuvia, Gainesville, 13-VIII to 2-XI-81, GMB; 3 ~ from pitch canker diseased terminals, 30-VII-1974, Flagler Co., RCW; 3 ~, emerged from soil under slash pine, Newberry, 17-VIII-1979, RCW & JLF; 5 pupae, Alachua Co., 16-VII-1982, GMB; 9 larvae, Trenton, 30 to 31-VII-1979, RCW & JLF; 3 larvae, pitfall traps, Newberry, 5-IX-1979, RCW & JLF; 2 larvae, Newberry, 30-VII-1979, RCW & JLF; 3 larvae, Gainesville, 26-VII-1979, RCW & JLF; 6 larvae, Gainesville, 7-VII-1981, GMB; 7 larvae, pitch canker diseased terminals, Flagler Co., 27-VIII-1975, RCW; 10 larvae, Flagler Co., 30-VII-1974; and 12 larvae from slash pine needles, Alachua Co., 28-VII-1981, GMB. Contarinia acuta is so-named for its extremely fine-pointed aedeagus. Adults will readily key to Contarinia in Gagne (1981), and it is the only species in that genus with such an attenuate aedeagus. It is otherwise dis tinguished from all other North American Contarinia on pine by the almost equally long claws and empodia (all other cecidomyiids associated with pine characteristically have much longer empodia than claws), the wide male hypoproct, and the completely setulose female cerci. Contarinia baeri (Prell) is the only other Contarinia known on pine in eastern North America, but that species has long empodia relative to the claws, differently-shaped male and female terminalia (unpub.), and its larva has each of the corniform terminal papillae at the end of a long caudal process (Skuhravy 1973). The larva of C. aeuta can be separated from those of the other 3 species of Contarinia associated with it on slash pine needles with the help of Fig. 11-8. In other Contarinia such differences urc diagnostic and will presum ably be so here also. Contarinia sp. A. (Fig. 13-4) is distinguished by a field of short spines almost surrounding the terminal corniform papillae. In addition, its spatula is least incised anteriorly of all 4 species. Contarinia sp. E. (Fig. 15-6) is rather similar to C. acuta except that the spatula is not as deeply incised, resulting in slightly differently shaped anterior lobes. Con tarinia sp. C. (Fig. 17-8) ha13 almost connate terminal corniform papillae and its spatula is the largest of the 4 species. Larvae of Thecodiplosis also have somewhat approximated corniform terminal papillae also, but without associated adults we cannot definitely say to which of the 2 genera species C belongs. For the present the larvae of species C are assigned to Contarinia sensu lato. Specimens of the 3 unnamed species of Contarinia from slash pine are recorded as follows: Contarinia sp. A.: 2 larvae, slash pine needle, Newberry, FL, 5-X-1981, GMB; 3 larvae, pitfall trap, Alachua Co., FL, 15-VII-1981, GMB; 1 larva, slash pine needle, Alachua Co., FL, 1-IX-1982, GMB. Contarinia sp. E.: 2 larvae, pitfall traps, Newberry, FL, 25-1-1980, R.C.W. & J.L.F.; 2 larvae, slash pine needles, DeLand, FL, 7-X-1982, GMB. Conturinia sp. C.: 2 larvae, pitfall traps, Newberry, FL, 25-1-1980, R.C.W. & J.L.F.; 1 larva, slash pine needle, Alachua Co., FL, 17-1X-1982, GMB; 4 larvae, Perry, FL, 2-III-1977, R.C.W. We are grateful to Drs. John L. Foltz and Robert C. Wilkinson, Depart ment of Entomology and Nematology, University of Florida, for initiating this study, providing research facilities, and submitting specimens for identification. Gagne & Beavers: Contarinia from Slash Pine Shoots 227

.::]:.

• \ _r..' ( ... ~~.

13 ~.

'~" :~.n..' .. .

;:x:?\•

15 16

~- ',' :. .------...... D .. , :. )( ., ,. ~ ( ! . ~ 17 18

Fig. 11-18. Larval spatulae with associated papillae and terminal seg ments (dorsal) of Contarinia spp. on slash pine shoots: 11-12, Contarinia acuta; 13-14, Contarinia sp. A; 15-16, Contarinia sp. B; 17-18, Contarinia sp. C. 228 Florida Entomologist 67 (2) June, 1984

REFERENCES CITED BEAVERS, G. M. In preparation. Bionomics of four species of Contarinia (Diptera: Cecidomyiidae) on slash pine needles. Univ. Florida, Gainesville, M.S. Thesis. GAGNE, R. J. 1981. Cecidomyiidae. Pages 257-92 In McAlpine, J. F. et al. Manual of Nearctic Diptera. Vol. 1. Research Branch Agriculture Canada Monograph No. 27. vi & 674 p. SKUHRAVY, V. 1973. "Needle blight" and "needle droop" on Pinus sylvestris L. in Europe and P. resinosa Ait. in North America (Diptera: Cecidomyiidae). Z. Angew. Ent. 72: 421-8. • • • • • • • • •• • •• EVOLUTION OF A FIREFLY FLASH CODE JAMES E. LLOYD Department of Entomology and Nematology University of Florida Gainesville, FL 32611 USA

ABSTRACT The flash code of Photinus ignitus (male flashes once, female delays long and then flashes) appears to have evolved from a code like that of Photinus macdermotti (male flashes twice, female delays briefly and flashes) by the omission of the second flash of the male pattern and the "connection" of the historic between-flash interval of the male pattern and the female short delay, thus producing the female long delay. Signaling variations and situations necessary for, and that might have led to, such a change in the hypothetical ancestor occur in today's macdermotti.

RESUMEN El c6digo de luces de Photjnus ignius (macho: 1 luz; hembra: pausa larga, luz) aparentamente ha evolucionado de un c6digo parecido a el de Photinus macdermotti (macho: 2 luces; hembra: pausa corta, luz). Ha omitido la segunda luz del patron del macho y ha unido el intervalo historico entre luces del patron del macho con la pausa corta de la hembra, asi pro duciendo la pausa larga de la hembra. Variaciones de senales y situaciones que permitiria y podria producir tal cambio en el antecesor hipotetico se observa en macdermotti en poblaciones actuales.

Firefly flashing behavior does not fossilize, but through a comparison of related living species we can see the sorts of changes that occur in evolution, and often get some notion of how and why these changes came about. Such a comparative approach may suggest specific observations and experiments that should be made, and focus attention on overlooked but significant details of the behavior of living species. I have made observations on and experi mented with the flashing behavior of several species in the Photinus con sanguineus group (Green 1956, Lloyd 1969) and on species with which they have pertinent interactions. I can suggest a reasonable scheme for the evolution of the code of Photinus ignitus from a precursor like the code of

228 Florida Entomologist 67 (2) June, 1984

Photinus macdermotti-ignitus males emit a single flash and their females delay several seconds before flashing their answer; macdermotti males emit 2 flashes and their females delay about 1 sec after the second flash before responding (Fig. la, b; Lloyd 1966, 1969). There are at least 9 Nearctic species in the consanguineus group of Photinus and probably more remain to be discovered (Lloyd 1969, p. 38; and unpublished). The flash patterns of mate-seeking males of 6 of the species usually or always, depending upon the species, are composed of 2-5 flashes, and those of 2 species are single flashes (Fig. la-g). Another is diurnal and has given up flash communication for pheromones (Lloyd 1973a, 1977). The most reasonable and conservative assumption as to the signal of the most recent common ancestor of the group is that its males had a multi-flash pattern. This is because most of the known living species in the group, the only available sample to draw inferences from, have such patterns, and there is no compelling evidence to the contrary. The 1-flash patterns of ignitus and aquilonius are thus derivative, as is the loss of adult luminosity in indictus. The apparent closest relatives and virtual sibling species (see Lloyd 1969) of ignitus, the (l-flash) firefly whose code evolution I seek to explain, are: consanguineus, greeni, macdermotti, and an unnamed species, "Barber's photinus." In the codes, of these species, male flash patterns are composed of 2 flashes, and female flash answers are delayed about 1 s after the second (P2) male flash. The intervals between the 2 flashes in the male patterns vary among the 4 species (Fig. 1b-e), and females will not answer paired flashes of inappropriate timing (Lloyd 1966, 1969, unpublished; Buck and Buck 1972) . If the P2 (second) flash of macdermotti's pattern is omitted the resulting (male-female) code is nearly that of ignitus (cf. Fig. 1a and 1b,h). But why would the females of some ancestral "macdermotti" population, whose males emitted 2-flash patterns, have become responsive to 1-flash patterns? Would there not have been strong selection pressure in the contexts of re productive isolation and sexual selection to avoid answering incorrect or inadequate patterns? If females accepted 1-flash patterns could they be ex pected to delay 1 s beyond where the missing P2 flash would have been?; or might they delay 1 s after P1? Why should males of this ancestor have be gun to emit 1-flash rather than 2-flash patterns? Field observations and lab experiments provide some tentative answers.

METHODS AND MATERIALS Field observations were made at numerous, diverse localities in eastern U.S., mostly in Florida. Field and lab flashing behavior was electronically recorded as described by Lloyd (1973b). Lab experiments used females from Gainesville, FL (Oct.-Nov. 1979), kept on a 16L:8D photoperiod with the dark phase beginning at 2 PM local time, about 30 min before experiments began. Females were held in individual shell vials (8 dram, airtight), corked, with a 5-mm cube of fresh apple to maintain humidity and provide water and possible nourishment. Artificial flash patterns were produced with an incandescent bulb shining along a 3-cm long, 6.7-mm-diam., covered lucite rod whose exposed end had been filed and rounded to produce even illumina tion. The flasher was aimed at subjects (in their vials) from distances of 30-50 cm, and moved a few centimeters from time to time when responses 230 Florida Entomologist 67 (2) June, 1984

a f

A ,: i 1LII II II II,:.Jt II 11 II II ~ ~ d'P1 ~ ~ ~ ~

I I I I I J I I I I I I I 0 1 2 3 0 0.5 1 1.5 b g

1 A ~ 0'P1 d'P2 ~ Jd' 'i'

I I I I I I I, I I I I ( I , 0 1 2 3 0 1 2 C h m~:lJ

"'. ~ d'P1 d'P2 'i''--- d' P1 'i' I I I J I I I I I I I I, 0 1 2 0 1 2 3 d i

fi A ~ A ~ A ~ ~ l~ d'P1 d'P2 O'P1 0'P2 I , I I I I! I I I I I I I, 0 1 2 0 1 2 3

e J A A ~ &'P1 &'P2 d'P1 Iv d'P2 ~v I, I I I r I r ! I I I I I 0 1 2 0 1 2 3 Lloyd: Firefly Flash Code 231 waned. Simulations of the 2-flash macdermotti pattern were made uniform (±<1%) by a stable electronic timing generator (Alton Electronics) whose flash interval was verified and monitored by the electronic recording system with which many of the experiments were recorded. Some experiments were timed with a split-timer, digital stopwatch (Siliconix ET110). The flash interval between P1 and P2 was approximately that measured over the past several years in free flying macdermotti males (Fig. 2). Flash duration was ea. 0.1 s. Test females (n=13) were collected as adults by their responses to penlight simulations of their males' pattern and probably included both virgins and nonvirgins; eggs in dissected females numbered: 1, 4, 6, 16, 16, 47, 70, and 78, suggesting that some had already laid eggs. Wing (1982) has found through field study of marked individuals that females may mate at least 3 times, and will attract males and copulate on consecutive evenings. Experimental protocol was as follows: individual females were presented with series of 5 (designated A-E) flash patterns at 5-10 s intervals; 10 series were presented in succession with 1 min of darkness interposed between series. Patterns in each series were either properly-timed (for ambient temp., Fig. 2) macdermotti patterns, or experimental, single flashes. The position of the 13, 1-flash (=X) presentations in the various series from 1-10 are as follows: series 1, no X; 2, X at E; 3, no X; 4, X at Ej 5, X at Ej 6, X at D and E j 7, X at E j 8, X at D and E; 9, X at C and E; 10, X at C, D, and E. There were 410 series presented, and in addition, a few longer series of patterns.

RESULTS Responsiveness to Experiment Flashes. When presented single flashes at the end of and within series of 2-flash patterns timed like those of males of their species, macdermotti females often answered, and, as anticipated, al most always at a long delay (see below). There was a range of response levels: one female answered only 6% (3/49) of the 1-flash patterns but answered 98% (169/173) of the preliminary and interspersed 2-flash (mac dermotti) patterns. At the other extreme one female answered 71% of the 1-flash patterns (36/51; 2-flash patterns = 95% 139/146), and another 89% (34/38; 2-flash patterns = 100% 113/113). (Totals in the preceding introductory summary include data from a few modified procedures not in cluded in the remainder of this report.) No overall increased or reduction in

Fig. 1. Flash codes of species in the Photinus consanguineus group (a-g), and other flash interactions. Flash forms drawn from chart traces of elec tronic recordings and, along with time intervals, are adjusted to about 23° C. Horizontal axis, time in sec; vertical axis, relative intensity. Variations in height among flashes not significant. Codes of: (a) P. ignitus, (b) P. mac dermotti, (c) P. consanguineus, (d) P. greeni, (e) "Barber's Photinus," (f) P. lineellus, indicating variation occurring in a male's pattern, (g) P. aquilonius. (h) An ignitus-like code is produced when a macdermotti female answers a single flash (compare with 'a'). (i) P. macdermotti female mis takenly answered after P1 j note identical delays after P1 and P2. (j) False injection (Iv) and false female response (Q v) of Photuris versicolor female aggressive mimic, a predator) j note different delays to P1 and P2 (see text, Lloyd 1981 and in prep). 232 Florida Entomologist 67 (2) June, 1984 •,, 3.2 "- ""- ~ 2.8 "- <.i Q) (f) c " ;::; 2.4 c;; "

~ 2.0 (\J Observation C-, = mac value ~ C- d"P1-P2 Interval -- 'b 1.6

1.2

15 20 25 30 Temperature (oG) Fig. 2. PIP2 line for determining mac~value across temperature range. Line was determined from PIP2 intervals of flying, flashing males. Values indicated at lower temperatures are questionable (see text). Divide observa~ tion by PIP2 interva~ indicated on y axis to determine mac-value. response level from the beginning of the 50-pattern test to the end waS ap parent: comparisons of response to early VB late experiments in the tests disclosed no difference (56% response to 2E and 4E vs 9C and 10C). (There was a slight increase (i=O.24 "mac-value," see text; ca. 0.04 s at 24°C) in delay when the first 12 responses to 2-flash patterns of 9 females were com pared with their last 12 responses on tests. Seven of 9 females showed an increase, on 12 of the 14 tests.) From test to test individuals sometimes changed response level dramatically (e.g. ~ 1, n~7 tests, range 0-100%; ~5, n~3 tests, range 0-77%; ~4, n-7 tests, range 15-77%). Other females showed greater constancy among their tests (e.g. ~ 3, n=4 tests, range 46 100%; ~ 13, n~3 tests, range 77-100%; ~ 2, n~2 tests, both 8%). The arrangement of patterns within the various series provided answers to specific questions: Would females answer single flashes if they had seen 4,3, or 2 macdermotti patterns first? Yes, and the mean response levels (both pooled and of means) were 50-59% (and to controls, 95-98%; and re sponses to the 3 presentations were not significantly different (Table 1, nos. 1-3). Would females answer l-flash patterns that were preceded by 1-flash patterns-Leo answer more than one experimental pattern in succession? Yes (Table 1, nos. 4, 6, 8), but if they did not answer the first 1-flash pattern TABLE 1. RESPONSE LEVELS OF FEMALES TO EXPERIMENT PATTERNS WITH DIFFERENT ANTECEDENTS.

Pool2 High3 Low4 X'S5 of c;? c;? Antecedents! X n X n X n X s Xu

1 4 mac X 0.55 82/149 0.91 10/11 0.08 1/13 0.50 0.32 11.5 (13 c;? c;? ) C· 0.95 79/83 1.0 8/8 1.0 10/10 0.92 0.17 4.5 2 3 mac X 0.59 41/70 1.0 6/6 0 0/6 0.52 0.39 5.8 .,....~ (12 c;? c;?) C 0.97 213/220 1.0 18/18 1.0 18/18 0.97 0.06 18.3 <:> ~ 3 2 mac X 0.55 41/74 1.0 0.17 0.54 0.35 R.. 7/7 1/6 6.2 '. (12 c;? c;? ) C 0.98 292/298 1.0 29/29 1.0 24/24 0.98 0.04 24.8 ~..... ~ 4 3 mac, X 0.66 27/41 0.80 4/5 0.38 3/8 0.60 0.39 4.6 (':) 1 single C 1.0 21/21 1.0 2/2 1.0 3/3 1.0 00 2.6 ~ answered ~ ~ (9 c;? c;? ) ~ 5 3 mac, X 00 0/32 00 0/8 00 0 3.6 ;;:r-o 1 single C 0.94 15/16 1.0 4/4 1.0 0 2.0 C'J not answered <:> R.. (10 c;? c;? ) (':) 6 2 mac, X 0.70 14/20 0.61 0.41 2.3 1 single C 1.0 21/21 1.0 0 2.7 answered (9 c;? c;? ) 7 2 mac, X 0.11 2/19 0.17 1/6 0.13 0.33 2.1 1 single C 0.93 14/15 1.0 4/4 0.95 0.13 2.1 not answered (9 c;? c;? ) l\:l CI:i CI:i ~ ~ H::-

'rABL:8 1. (Continued) ~ "- C 8 2 mac, x 0.77 10/13 0.57 0.53 1.9 .,...""i: 2 single R.. answered :;:l (7~ ~) ~ 9 2 mac, x 00 0/22 00 0/9 00 00 2.2 C""'" 2 single ~ C "- with <2 not C answered (Q (10 ~ ~ ) ~. ""'"cr;, ~ 'Stimulus events in series preceding presentation of experiment pattern. E.g. [4 mac, 1 single] indicates 4 macdermotti flash patterns and one I-flash ,-, pattern were seen by the experimental subject before the experiment (l-flash) pattern was presenkd. t-.? 2Pooled data, aggregating all responses of all females in all series with given antecedents to experiment pattern. '-" "High single female of females receiving 5 or more stimuli preceded by given antecedents. 'Low single female of females receiving 5 or more stimuli preceded by given antecedents. "Statistics of response level means (mean and standard deviation) and mean number of experimental sequences presented to individual females in the group• •Controls are response given to macdermotti (2-flash) patterns in other series with antecedents identical to the experiment presentation under considera tion, except in 4 which had no parallel series; the "control" sequence used (fr.>m line 9) was [2 mac, 1 single], the proper control for 6, but note that re sponses to 4 and 6 were similar.

~ ~ I:l ~etl f-4 ~ 00 H::- Lloyd: Firefly Flru;h Code 235 they were not likely to answer another one following it (Table 1, nos. 5, 7, 9). There was not a significant difference in response level among 3, compared, antecedent conditions (involving 2 vs 3 macdermotti patterns and 1 vs 2 1 flash patterns; Table 1, nos. 4, 6,8). Would females answer more than three I-flash patterns in succession? Yes; 5 of the 13 females answered 3 or more consecutive I-flash patterns. The limit of female responsiveness to patterns in succession was "probed" beyond the measure encorporated in the protocol (series 10, C-E) by sometimes continuing to present I-flash patterns after 10E, at 5-10 s intervals. High values for continuous answering of these (extra) single-flash patterns for different females were 15, 14, 6, 5, and 4. The 2 high females with 15 and 14 responses, in other post 10E series an swered 13 and 10, respectively. One was given 146 patterns after the 50 pattern procedure. Each of the 16 times she stopped responding to I-flash patterns, she immediately answered a 2-flash pattern and 1-5 subsequent I-flash patterns. Response levels of females on individual tests and in total, during their experimental careers and including as well as excluding zero test scores, were examined for correlations with: days since capture, days till death, number of eggs present at death (""'" at capture), and number of tests given. Only "days until death" showed any significant correlation, and this was im proved by excluding zero-score tests and 2 very high scores that occurred one day before death in 2 individuals (r = 0.63; Fig. 3). Female response to I-flash patterns gradually dropped between 12 and 4 days before death (y = 0.28 + .06x), and then appears to drop sharply. If there actually is a sharp break in the response curve at ca. 3 days it would explain and incorporate several low and zero scores. Timing of Female Responses. Female macdermotti flashed answers to approximately 1800 I-flash and 2-flash stimulus patterns during this study. In all but 26 their answers followed the P2 position, and roughly by 1 s whether the P2 flash was present or not; thus, when timed from PI the delay ranged around 3 s (.:;;:,d9°C, Fig. 1h). Delays varied predictably with tem perature, as do other firefly flash parameters, including the delay of ignitus females. In order to combine and compare measurements (of delays, etc.) made at different temperatures, delays were divided by the male P1P2 in terval measured at various temperatures in free flying males over the past several years (Fig. 2). The resulting number is referred to as a mac-value (=mv; Lloyd 1981). A total of 789 delays from 11 females were measured. The mean female delay to I-flash patterns (n=120) was about 3% longer than that to 2 flash patterns (n= 669; 1.33 vs 1.29 mv), and this bias direction occurred in 8 of the 11 females. This difference is not significant if the responses of all females (n= 11) are compared (t= 1.50, DF = 20). However, if the responses of only the 8 are compared, the difference (1.37 vs 1.30 mv; 5% longer) is significant between 0.01 and 0.001 (t=3.76, DF=14). The mean female delays to 2-flash patterns that were presented immediately before I-flash patterns were like those given to all other 2-flash patterns (before: i = 1.29, 8=0.044; all: i=1.29, 8=0.039; t=O.O, DF=20). The delays elicited by 2 flash patterns immediately following I-flash (experimental) patterns were like those after all other 2-flash patterns (after: i = 1.30, 8 = 0.046; all: i=1.29, 8=0.039; t=0.553, DF=20). 236 Florida Entomologist 67 (2) June, 1984 too o .. • • o • .80 • • • • Q) • • > Q) -I Q) • • • • en .60 c: o enCo Q) • c: .40 Q) • • Ci5 y =.28 + .06x E • Q) LL r =.63 .20 •• • •• •

o 2 468 10 12 Days Until. Death Fig. 3. Change in female response level as she ages, i.e. as a function of "days until death." Note that about 4 days before death response level prob ably drops precipitously. Five °values and 2 very high values on day 1 (circles) were excluded from calculations to determine line.

There was more variation among the means of female responses to 1 flash patterns (8=0.075) than 'among those to 2-flash stimuli (8=0.039; F=3.73, DF=20, p<0.05). When two 1-flash patterns were answered in succession (presentations 6D,E; 8D,E; 10C,D), the delay following the second in the series was not significantly longer (7 ~ ~, 25 responses; 8 = 7.3%; t=1.70, DF=48). Although extensive information was obtained on variations in female delays-from stimulus to stimulus, among different tests on individual fe males, and among females,-it is largely uninterpretable at the moment. (But experiments and field observations over the past several years have suggested that it is possible that females convey information to males by variations in thE;ir timing from pattern to pattern.) Durations of female response flashes emitted after 1-flash patterns were like those following 2-flash patterns (6~ ~, 34 vs 62 flashes measured; X "'=' 0.14 s, 20-23° C; t=O.316, DF=10). Delay lengths of female responses during their experimental careers showed no significant correlations with days since capture, days until death, or number of eggs present at death. Of the 26 flashes that females emitted after the P1 flash, 15 were meas ured. They were timed like flashes emitted after P2 flashes (differences not Lloyd: Firefly Flash Code 237 significant, t=O.16, DF=23). Particular attention was given to this timing because signal mimicking predators and competing males of this species in ject "extra" flashes between P1 and P2 of an approaching male (Lloyd 1979, 1981). It was conceivable that female macdermotti flashes occurring after P1 could have some connection with the injected flashes of males and predators. This does not seem to be the case (Fig. 1, cf i with j ).

DISCUSSION The experiments demonstrate that female macdermotti will answer 1 flash patterns after they have seen 2-flash patterns characteristic of their species, and that some females, or females sometimes, will answer several in succession. These experiments also demonstrate that female macdermotti answer single flashes at long delays, thus producing a flash interaction similar to that of ignitus (Fig. 1a,h), the species whose flash code is of evolutionary interest here. Since vegetation can sometimes obscure one flash of a male 2-flash pattern, it seems reasonable that females would sometimes answer incomplete patterns, thus helping the approach of a male they have already identified as a member of their species. But why should females rarely (26/ """,1800) treat a single stimulus flash as P1 and not P2? I believe part of the answer is that firefly females have a timing mechanism that measures flash pattern intervals (repetition cycles). This could explain why competing males of Photinus pyralis that hover and flash together near a responsive female do not flash immediately after a female response, but instead wait a few seconds and then flash, thus, in near synchrony (Maurer 1968, Lloyd 1979). Such a timer could also have some role in a more complex coding used to detect the predaceous mimicries of Photuris females (Lloyd 1983). The time-measuring mechanisms of macdermotti females of primary im portance here are those that measure the interval of the 2 flashes in the males' 2-flash pattern and the delay after P2. To recognize males of her species, a macdermotti female measures the P1-P2 interval. If this interval is appropriate, and P2 is seen during the critical "window," a timer counts down to trigger the response flash at the appropriate time. P. macdermotti females that answer a 1-flash pattern at the long delay have, in one simple model, coupled the 2 timers. Additional experiments should permit more to be said about the factors that influence "decision making" at the junction of these timers. Why would males of ignitus' ancestor have begun to emit 1-flash rather than 2-flash patterns? (Note that their females were already responsive to 1-flash patterns after initial contact and identification.) At first they would have emitted both 1-flash and 2-flash patterns, depending upon circum stances. There are 3 situations in which 1-flash patterns would have been adaptive: (1) The typical species pattern was altered to avoid attracting other males that would have competed for the female. At first this could have been an advantage, but eventually, in evolutionary time, it would have been countered by competitor recognition of the "secret code" approach, and females should have encouraged the counter (sons would get "genes" for the alternative mate-finding tactic). (2) Males might omit P2 upon close approach to a responding light to 238 Florida Entomologist 67 (2) June, 1984 cause a signal-mimicking predator to reveal her identity. The omission of P2 would probably cause Photuris females to stop answering or to answer incorrectly either because: their neural mechanisms for answering after the P2 flash were not like those of macdermotti females, whose machinery meas ures intervals between P1 and P2, and Photuris could not time from P1; or because they "computed" the single flash as that of another species they preyed upon and answered in its code. Experiments with Photuris versicolor females (Florida), predators of macdermotti, indicate that they answer 1 flash patterns with a short «1 sec) delay, not a long one (Lloyd 1981, un published) . (3) Photuris females attack flashing males in the air, using the flashes to aim their attacks, and they sometimes switch to this predatory tactic from signal mimicry, when approaching males hesitate (Lloyd and Wing 1983). A 2-flash ignitus ancestor might have omitted P2 on close approach to a fe male, or even during searching, because it provided a target for predators that approachd P1, during the P1P2 interval, and struck at P2. Several factors could have been responsible for finally fixing the 1-flash pattern in ignitus, but after viewing aerial attack behavior by various Photuris species and noting its efficiency and frequency of occurrence at certain times, and the effects it seems to have on macdermotti behavior (Wing 1982, Lloyd and Wing 1983), I believe it could well have been the critical factor in fixing ignitus' 1-flash pattern. Complete acceptance of 1-flash patterns by females would follow the change in males if the risks to males were very great.

EPILOGUE The high level of response to 1-flash patterns found in some females was not anticipated. On numerous occasions over the past several years females in the field and lab had been presented with single flashes and none had flashed in response. Therefore, the experimental protocol used in this study was not designed to determine whether females would answer 1-flash pat terns without prior 2-flash stimulation. This was tested on females taken from the same locality as those used above, in the spring (n=5) and fall (n=3) of 1983. They were presented 15 single flashes in 3 groups of 5 (in the protocol described above), then an increasing number of normal (2 flash) patterns was interspersed in subsequent groups of the tests. Response level to normal patterns was 0.78, lower than in previous testing in which normal patterns were more numerous (cf. 0.93-1.0, Table 1). Only 5 of 661 single flashes were answered at a time delay that indicated females judged them to be P1's. These responses occurred after normal patterns had been presented and not when females had seen only single flashes. Thus, some priming by the normal (species-typical) male pattern is necessary to elicit long-delayed responses to single flashes. Of more interest is the fact that 3 females answered a total of 50 1-flash patterns at a time delay that indicated that they judged them to be P2's. This is being studied.

ACKNOWLEDGMENTS I thank John Alcock, John Sivinski, Tom Walker, Steve Wing, and an anonymous referee for commenting on the manuscript; Ngo Dong for dis secting the fireflies; Ray Littel for statistical consultation; Laura Reep for Lloyd: Firefly Flash Code 239 drawing the figures; Jack Schuster for composing the Resumen; and Barbara Hollien for typing the manuscript. This study was supported by my Depart ment and N.S.F. Grant DEB-7921744. Florida Agricultural Experiment Sta tion Journal Series No. 4527.

REFERENCES CITED BUCK, J. B., AND E. BUCK. 1972. Photic signaling in the firefly Photinus greeni. BioI. Bull. 142: 195-205. GREEN, J. W. 1956. Revision of the Nearctic species of Photinus (Lampy ridae: Coleoptera). Proc. California Acad. Sci. 28: 561-613. LLOYD, J. E. 1966. Studies on the flash communication of Photinus fireflies. Univ. Michigan Mus. Zool. Misc. ColI. 130: 1-95. ---. 1969. Flashes, behavior and additional species of Nearctic Photinus fireflies (Coleoptera: Lampyridae). Coleop. Bull. 23: 29-40. ---. 1973a. Chemical communication in fireflies. Environ. Ent. 1: 265-6. ---. 1973b. Fireflies of Melanesia; bioluminescense, mating behavior, and synchronous flashing. Environ. Ent. 2: 991-1008. ---.. 1977. Bioluminsecence and communication. Pages 164-83 in T. A. Sebeok, ed. How animals communicate. Indiana Univ. Press, Blooming ton. ---,. 1979. Sexual selection in luminescent beetles. Pages 393-42 in M. and A. Blum, eds. Sexual selection and reproductive competition in in sects. Academic Press, New York. ---,. 1981. Firefly mate-rivals mimic their predators and vice versa. Nature 290: 498-500. ---. 1983. Bioluminescence and communication in insects. Annu. Rev. Ent. 28: 131-60. ---, AND S. R. WING. 1983. Nocturnal aerial predation of fireflies by light-seeking fireflies. Science 222: 634-5. MAURER, U. M. 1968. Some parameters of photic signaling important to sexual species recognition in the firefly Photinus pyralis. M.S. Thesis, SUNY, Stony Brook, New York. WING, S. R. 1982. The reproductive ecologies of three species of fireflies. M.S. Thesis, Univ. of Florida, Gainesville. 240 Florida Entomologist 67 (2) June, 1984

REPRODUCTIVE MATURATION AND PREMATING PERIOD OF THE PAPAYA FRUIT FLY, TOXOTRYPANA CURVICAUDA (DIPTERA: TEPHRITIDAE)

PETER J. LANDOLT' Subtropical Horticulture Research Unit, USDA, ARS 13601 Old Cutler Road Miami, FL 33158 USA

ABSTRACT Female papaya fruit flies, Toxotrypana curvicauda Gerstaecker, obtained from field-infested papayas, Carica papaya L., required 5 to 6 days to de velop mature ovaries comparable in size to those of Dvipositing field-collected females. Younger females, with ovaries less than 0.25 em long, were absent from papaya groves, sJ.1ggesting that female papaya fruit flies do not search out host plants until mature. In mating tests females mated when 5 days old or older, while males mated at any age (one to 6 days after emergence) al though mating rates were less with one-day-old males.

RESUMEN Hembras de Ia mosca de Ia papaya, Toxotrypana curvicauda Gerstaecker, obtenidas de papayas, Carica papay L., infestadas en el campo necesitaron 5 6 6 diRs para desarrollar ovarios maduros parecidos en tamafio a aquellos de las hembras adultas ponedoras de huevos coleccionadas en el campo. No Se encontraron en el campo hembras mas j6venes, con ovarios de menos de de 0.25 cm, de longitud, sugiriendo esto que la hembra de la mosca de la papaya no husea planta hospedera hasta au madurez. En pruebas de acopla miento las hembras se acoplaron al alcanzar 5 dias de edad 0 mas, mientras que los machos se acoplaron a cualquier edad (de uno a 6 dias despues de emerger) aunque los niveles de acoplamiento fueron menores en machos de un dia de edad.

Tephritid fruit flies generally are reproductively immature upon emer gence as adults and require a maturation period before mating and oviposi tion occur. Laboratory-reared Anastrepha suspensu (Loew) begin mating at 10 days of age (Mazomenos et al. 1977), coinciding with maximum develop ment of male pheromone glands (Nation 1974) and attraction of females to males (Nation 1972). Similar premating periods are 3 to 4 days in Ceratitis capitata (Wiedemann) (Mourikis 1965), 12 to 14 days in Dacus tryoni (Froggatt) (Fletcher 1969), and 7 to 8 days in Rhagoletis pomonella (Walsh) (Prokopy et al. 1972). Many tephritids require a proteinaceous food source (usually honeydew) during this period for development to sexual maturity (Bateman 1972). It was reported by Landolt and Hendrichs (1983) that female papaya fruit flies, Toxotrypana curvicauda Gerstaecker, observed in field cages mated and oviposited only when 6 days old or older, suggesting an obligatory pre-

lInsect Attractants. Behavior & Basic Biology Research Laboratory, USDA, ARS, P.O. Box 14565, Gainesville, FL 82604 USA. Landolt: Toxotrypana curvicauda 241 mating period. This paper describes the growth of the ovaries of female papaya fruit flies after emergence and the mating propensity of males and females of different ages. The primary purpose was to determine if such an obligatory pre-mating period exists in the papaya fruit fly and how long it lasts.

MATERIALS AND METHODS Flies used in laboratory tests were obtained from infested papaya fruits, Carica papaya L., collected in a commercial grove in Dade Co., Florida from February through July 1982. Mature larvae from these fruit pupated in moist soil placed 7 cm deep in 8 oz wax-cartons. These cartons of soil with pupae were kept at 26-28°C and 60-80% RH until adult emergence. Emerged flies were sorted by sex daily (at 8-9 AM EST) and placed in (24 cm) 3 screen cages with sugar cubes and a wet sponge. Fifteen female papaya fruit flies of each age from 1 to 8 days old (1 day = 6 to 30 hold) were dissected to measure ovarian development. Dissections were made under 80% ethanol using a Bausch and Lomb binocular micro scope. All measurements were made at lOx with the aid of an ocular microm eter. Flies tested for mating propensity were placed in 8 oz wax-coated cartons with screen on both top and bottom, with 1 pair of flies per carton. Tests were initiated at 1 :30-2 :00 PM EST and pairs were observed for mating every 0.5 h thereafter until 4 :00 PM. Mating observed in field cages was highest at this time of day (Landolt and Hendrichs 1983). All combinations of ages, from 1 to 6 days old, were tested for mating propensity with 40 pairs (4 groups of 10 pairs) per age combination. Groups of flies in which mating occurred were not used in subsequent mating tests. To obtain an estimate of the normal maximum size of papaya fruit fly ovaries, 72 adult females were netted in a papaya grove during March 1982. These were dissected and the ovaries measured to determine the normal range of ovary size in a field population. Females collected while ovipositing (n=32) were considered mature and were evaluated separately to determine the size of ovaries in reproductively mature females.

RESULTS Emerging female papaya fruit flies had undeveloped ovaries without mature eggs. The average length and width of ovaries of 1-day-old flies were 1.25 and 0.44 mm respectively (Fig. 1), while the maximum egg length average 0.23 mm (Fig. 2). In most flies at this age the abdomen was filled with fat body as clusters of fat body cells. The length and width of ovaries and length of eggs increased nearly linearly for 6 days after emergence, decreasing thereafter. The average length and width of ovaries peaked at 3.14 and 0.75 mm respectively and the maximum length of eggs was 1.3 mm in females 6 days old. The fat body of 3 to 4-day-old flies was restricted to the posterior area of the abdomen but had disappeared in 6 to 7-day-old females in which the ovaries had reached maximum size. Ovipositing female papaya fruit flies collected in the commercial papaya grove had ovaries with an average length of 3.1 mm (SE = 0.3) and width of 0.9 mm (SE = 0.1), comparable to that of 5 to 6-day-old flies in the labo- 242 Florida Entomologist 67 (2) June, 1984

~ ~ f z LENGTH -w 2 N en- > 0: oct ~ 1 I I i WIDTH

1 2 3 4 5 6 7 8

NO. DAYS POST - EMERGENCE Fig. 1. Average length and width of ovaries of female papaya fruit flies at 1 to 8 days post-emergence. Each average calculated from measurements of ovaries of 15 flies. ratory and significantly larger than that of 1 to 4-day-old females (Student's t-test, p ~ 0.05) (Table 1). Assuming that ovipositing females were repro ductively mature, then females in the laboratory reached maturity in 5 to 6 days. Ovaries of the field-collected females that were not ovipositing when collected averaged 3.2 mm (SE = 0.3) in length and 0.9 mm (SE = 0.2) in width and were nearly identical to those of ovipositing females. Of the 72 females collected in the field, none had ovaries as small as 1 to 3-day-old females in the laboratory. This suggests that female papaya fruit flies do not search out host plants until reproductively mature. Mating rates for papaya fruit flies of various ages are presented in Table 1. Males of all ages tested (1 to 6 days old) mated, although rates were somewhat lower with males 1 day old. Generally, females did not mate unless 5 days old or older, although a few matings were noted at 3 and 4 days (Table 1). Mating rates of females at 5 days were significantly higher than at 4 days, with all ages of males (Student's t-test, p ~ 0.05).

DISCUSSION The data on ovary and egg development (Fig. 1 and 2) and mating propensity (Table 1) all indicate a pre-mating period in the female papaya fruit fly. Attainment of maximum size of eggs and ovaries was coincident with that of maximum mating propensity at 6 days post-emergence, and with Landolt: Toxotrypana curvicauda 243

T •T f T 1 1 I T • ::z:: t; T 1 Z • W ...I 1 T •1

1 2 3 4 5 6 DAYS AFTER EMERGENCE Fig. 2. Average maximum length of eggs in ovaries of female papaya fruit flies at 1 to 8 days post-emergence. Each average calculated from measurement of the largest eggs in each of 15 flies. the age at which females mated in field-cage observations (Landolt and Hendrichs 1983). Unlike other tephritids studied (Bateman 1972), papaya fruit fly re productive development did not require feeding on protein rich foods. All flies in these studies were given only sugar and water. Also, no difference was detected between maximum ovary length of females held 5 days without

TABLE 1. PERCENTAGE OF PAIRS OF PAPAYA FRUIT FLIES OF INDICATED AGES THAT MATED, 2:00 TO 4:00 PM, APRIL TO JULY 1982, MIAMI, FL. EACH PERCENTAGE CALCULATED FROM 40 PAIRS.

Age of Age of Females males 1 day 2 days 3 days 4 days 5 days 6 days

1 day 0 0 0 5.4 30 50 2 days 0 0 0 7.5 85 80 3 days 0 0 ,.,0 7.5 45 85 4 days 0 0 I 14 73 87 5 days 0 0 0 16 65 87.5 6 days 0 0 2.5 15 60 71 244 Florida. Entomologist 67 (2) June, 1984 a protein source (2.8 ± 0.6 mm, n=25) or with yeast hydrolysate paste (2.84 ± 0.4 mm, n=25). Apparently, enough reserves are stored during the larval stages to permit reproductive development without such feeding. This may be related to young larvae feeding on the more nutritious seed embryos within the fruit (Mason 1922) rather than on fruit pulp as do other frugiverous tephritid larvae. No premating period was evident in male papaya fruit flies. Toxotrypana curvicauda apparently is the first tephritid species for which such a sharp difference in reproductive maturation between the sexes has been demon strated. Maximum male mating propensity was reached by the second day after emergence, coinciding with the age at which my personal observations indicate that males begin puffing (indicative of pheromone release).

ACKNOWLEDGMENTS Technical assistance was provided by L. Lichtenstein and B. Brown. R. Nelson kindly allowed use of his papaya grove. Spanish translation of the abstract was provided by Alicia Arner and Maria Nieto.

REFERENCES CITED BATEMAN, M. A. 1972. The ecology of fruit flies. Ann. Rev. Ent. 17: 493-518. FLETCHER, B. S. 1969. The structure and function of the sex pheromone glands of the male Queensland fruit fly, Dacus tryoni. J. Insect Physiol. 15: 1309-22. LANDOLT, P. J., AND J. HENDRICHS. 1983. Behavior of the papaya fruit fly, Toxotrypana curvicauda Gerstaecker (Diptera: Tephritidae). Annals Ent. Soc. America 76: 413-7. MASON, S. C. 1922. Biology of the papaya fruit fly, Toxotrypana curvicauda, in Florida. USDA Bull. 1081: 1-9. MAZOMENOS, B., J. L. NATION, W. J. COLEMAN, K. C. DENNIS, AND R. ESPONDA. 1977. Reproduction in Caribbean fruit flies: Compari son between a laboratory strain and a wild strain. Florida Ent. 60: 139-44. MOURIKIS, P. A. 1965. Data concerning the development of the immature stages of the Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae) in different host fruits and on artificial media under laboratory conditions. Ann. Inst. Phytopathol., Benaki, 7 (N.S.) : 59-105. NATION, J. L. 1972. Courtship behavior and evidence for a sex attractant in the male Caribbean fruit fly, Anastrepha suspensa. Ann. Ent. Soc. America 1364-7. ---. 1974. The structure and development of two sex specific glands in male Caribbean fruit flies. Ann. Ent. Soc. America 67: 731-4. PROKOPY, R. J., E. W. BENNETT, AND G. L. BUSH. 1972. Mating behavior in Rhagoletis pomonella. II. Temporal organization. Canadian Ent. 104: 97-104. Lin & Harper: Entomophthora gammae Culture 245

ISOLATION AND CULTURE OF ENTOMOPHTHORA GAMMAE, A FUNGAL PARASITE OF NOCTUID LARVAE

JING-SHYAN LIN AND JAMES D. HARPER Department of Zoology-Entomology Alabama Agricultural Experiment Station Auburn University, AL 36849 USA

ABSTRACT Entomophthora gammae (Weiser) was isolated by allowing conidia from infected soybean looper, Pseudoplusia includens (Walker), larvae to shower onto egg yolk agar. Fungal and bacterial contaminants were always present, but at 20°C, E. gammae grew fast enough to allow the subculture of hyphal bodies to clean medium. Once isolated, the fungus was cultured on Czapek solution agar, potato dextrose agar, Sabouraud dextrose agar, and Sabouraud maltose agar, all containing 0.2% yeast extract, and on Grace's insect tissue culture medium. It did not grow on Wolf's medium, potato agar or potato dextrose agar without yeast extract. The fungus grew most rapidly in culture when maintained at 25°C and 100% RH. Conidia on glass slides held at 25°C germinated only at 100% RH. A gradual increase in conidium size occurred over a 10-month period of continuous culture. At tempts to infect larvae with the cultured fungus were unsuccessful.

RESUMEN Entomophthora gammae (Weiser) fue aislado permitiendo que las conidias provenientes de las larvas del medidor de la soya, Pseudoplusia includens (Walker) se dispersaran sobre agar de la yema de huevo de gallina. Hongos y bacterias contaminantes estuvieron siempre presentes, pero a 20°C, E. gammae creci6 10 suficientemente rapido para permitir la subcultura de cuerpos hifales para clarificar el medio. Una vez aislado, el hongo fue cultivado en una soluci6n de agar Czapek, papa dextrosa agar, Sabouraud dextrosa agar y Sabouraud maltosa agar, todos los medios con tenian 0.2% de extracto de levadura y sobre el medio Grace para cultivo de tejidos de insectos. No hubieron crecimientos sobre el medio Wolf, papa agar o papa dextrosa agar cuando no se agreg6 extracto de levadura. El creci miento maximo y la germinaci6n de las esporas ocurri6 a los 25 ° C y con un 100% de humedad relativa. Un incremento gradual del tamafio de las conidias ocurri6 durante los 10 meses de cultivos continuos. Intentos para infestar larvas con el hongo cultivado no tuvieron exito.

Several species of the noctuid subfamily Plusiinae are susceptible to infection by Entomophthora (Tarichium) gammae (Weiser). Harper and Carner (1973) reported this pathogen from soybean looper, Pseudoplusia includens (Walker), and cabbage looper, Trichoplusia ni (Hubner), and Weiser (1965) reported it from Plusia gamma (L.) in Europe. It is a wide spread pathogen of P. includens in the southeastern United States where this pest attacks soybean (Harper et al. 1983). Several studies on this pathogen have defined environmental conditions which are related to its ability to induce and maintain epizootics (Newman and Carner 1974, 1975a,b). This report presents results of work on the isolation and culture 246 Florida Entomologist 67 (2) June, 1984

of E. gammae, on defining conditions for in vitro growth of the fungus, and on laboratory infectivity studies.

METHODS AND MATERIALS

ISOLATION Fourth- and fifth-instar P. includens larvae were collected on soybeans in fields near Tallassee, Alabama. At the time of collection, an epizootic of E. gammae was in progress. Larvae were placed in individual 30-ml cups containing meridic diet (Shorey and Hale 1965) and returned to the lab oratory. Larvae infected with E. gammae typically died in late afternoon and rapidly turned an abnormal pale yellow color. Dead larvae were taped to the inside of the lids of 20 x 100 mm petri dishes containing either egg yolk agar (EYA), potato agar, potato dextrose agar (PDA), Sabouraud maltose agar, Sabouraud dextrose agar, or Wolf's medium (Wolf 1951). Plates of each medium were held at 20 and 25°C. By the following morning, conidiophores and conidia had been produced, the latter having showered onto the medium surface below. Once Entomophthora could be identified microscopically on the plates, it was transferred to fresh medium using a sterile inoculating needle. By transferring at this stage, contaminants, which were always numerous, were left behind. Several passages were sometimes necessary to completely eliminate all contaminants.

CULTURE Once isolated, small quantities of hyphae were transferred with sterile inoculating loops to the centers of plates containing various media or into tubes of liquid media and maintained at 20 or 25 °C. Transfers were made every 14 days to maintain actively growing cultures. Cultures were also held at 10°C and transferred every 2 months without loss of viability.

GROWrH AND GERMINATION Plates of EYA and PDA plus 0.2% yeast extract were inoculated with E. gammae as described above and incubated in temperature cabinets main tained at 5 ± 1 degree intervals from 5 to 35°C. Colony size was determined daily for 13 days by placing each dish over a cool light source and recording the average of two diameter measurements taken at right angles to each other. Seven petri dishes containing pure sporulating cultures of E. gammae were inverted over seven EYA plates and allowed to shower for 3-4 hours. Inoculated plates were covered and maintained at the temperatures used in the growth rate studies. Plates were examined daily under a microscope to determine the percent germination of 100 conidia. Humidity chambers were established using glass desiccator jars con taining saturated salt solutions to provide the following humidity levels (Solomon 1951); <1% (CaCI, without water), 30% (CaCI, • 6H,O), 50% [CO (NO,) , • 4H,O], 70% (NH.CI + KNO,) , 90% (ZnSO•• 7H,O), 100% (distilled water). Humidity in each chamber was checked with a humidity Lin & Harper: Entomophthora gammae Culture 247 gauge (Bacharach Industrial Instrument Co.), allowing sufficient time after closure for the humidity level to stabilize. Sporulating cultures of E. gammae were inverted over glass slides for 3-4 h, and the slides were introduced into the chambers. All chambers were held at room temperature (ca. 25°C) under fluorescent lighting. Slides were removed from the chambers at ca. 5-h intervals for 1 day and again at 57 h. Percent germination was determined at each interval from the first 100 conidia encountered while scanning the central area of the slide.

INFECTIVITY STUDIES

Field collected cadavers exhibiting symptoms of E. gammae infection were suspended over individual healthy, laboratory-reared P. includens larvae held inside petri dishes. Seven larvae were successfully exposed in this manner. Ten larvae were exposed to conidia showering from each of 3 separate, inverted culture plates which had been maintained in the labora tory through 17 transfers over a I-year period. After 4-6 hours exposure, all larvae were transferred to individual I-ounce cups containing diet and were held at either 90 or 100% RH until they pupated or died.

RESULTS

ISOLATION

Although several media supported growth of E. gammae, EYA held at 20°C was the only medium-temperature combination from which a pure culture could be initiated. The inoculation technique introduced many bac terial and fungal contaminants which at 25°C overgrew the plates before E. gammae colonies could become established. These contaminants grew more slowly at 20°C on EYA, and E. gammae colonies were able to develop vegetatively and could be transferred to clean plates before being over grown. After the transfer of hyphal bodies, the mycelium grew vegetatively for 2-3 days; then hyphal bodies produced conidia which were ejected. Those landing on fresh medium germinated, produced germ tubes which elongated, thickened and grew vegetatively, repeating the above cycle. This process continued as a series of expanding concentric rings of growth until the petri dish limited further growth.

CULTURE Hyphal bodies transferred to media other than EYA exhibited varying growth responses. Potato agar, Wolf's medium, and PDA did not support growth. A very weak growth was obtained in Czapek solution containing 0.2% yeast extract. PDA + 0.2% yeast extract supported good growth, while Sabouraud dextrose agar and Sabouraud maltose agar, both contain ing 0.2% yeast extract, supported even greater growth. Greatest growth occurred on EYA containing 0.2% yeast extract and on Grace's insect tissue culture medium. On EYA held at 25°C, E. gammae, following the growth cycle described above, covered a 10 cm plate within 1 week. On Grace's medium, the inoculum floated to the surface and grew as a white, compact mass. 248 Florida Entomologist 67 (2) June, 1984

GROWTH AND GERMINATION Growth on EYA was so rapid and diffuse that growth rates were difficult to determine. PDA + yeast extract supported slower growth with more clearly defined colony borders; growth on this medium at 5 to 35°C is pre sented in Table 1. Growth was clearly inhibited at the two temperature ex tremes. Most rapid growth occurred at 25°C, followed in order by 20 and 30°C. Growth was negligible at 10 and 15°C. Conidial germination and hyphal development occurred equally well at 20, 25, and 30°C, but no germination or growth was observed at 5, 10 or 35°C (Table 2). In humidity chambers, 20% of conidia on glass slides germinated after 6 hours at 100% RH and one conidium germinated after 12 hours at 90%; no germination occurred through 57 hours at any of the other RH levels tested. Distinct morphological changes occurred in E. gammae cultured con tinuously on EYA for 10 months. Naturally occurring conidia and those produced by initial isolations were the typical shape and size (18.80 ± 2.20 x 8.80 ± 1.60 ,urn) reported for the fungus (Harper and Carner 1973). After 10 months in culture, conidia were more oval, larger in size (28.77 ±

TABLE 1. GROWTH OF Entomophthora gammae AT DIFFERENT TEMPERA- TURES ON POTATO DEXTROSE AGAR CONTAINING YEAST EXTRACT.

Tempera- ture (OC) Colony diameter (mm), at day after inoculation (n = 3 plates) 1 1 2 3 4 5 6 7 8 9 10 11 12 13

10 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.5 0.8 0.8 1.2 15 0.3 0.3 0.3 0.3 0.3 0.3 0.6 1.2 1.4 1.9 2.9 3.1 4.6 20 0.3 0.3 ~.1 2.8 4.2 5.8 7.3 9.8 12.8 19.8 23.8 28.8 33.8 25 0.3 1.1 2.1 5.6 7.1 10.6 13.6 19.6 24.6 31.6 36.6 38.6 43.6 30 0.3 0.3 0.4 2.1 3.1 4.9 6.6 8.6 11.6 14.6 16.6 20.6 23.6

'No growth occurred at 5 or 35°C. All plates were inoculated with a 0.3 mm diameter mass of hyphae.

TABLE 2. CONIDIAL GERMINATION AND GROWTH OF Entomophthora gammae ON EGG YOLK AGAR AT DIFFERENT TEMPERATURES FOLLOWING INOCULATIONS

Temperature (OC) % Germination' Growth' 24 h 48h 72 h 96 h

10 0 a 0 0 15 a a a 0 20 30 80 + ++ 25 70 90 +++ +++ 30 60 95 ++ ++ 35 a 0 a a

'Based on examination of 100 conidia at each temperature. '0 = No growth, + = poor growth, + + = Good growth, + + + = Very good growth. Lin & Harper: Entomophthora gammae Culture 249

4.34 x 19.85 ± 4.65 /lm), and contained larger and more numerous oil globules. In culture, conidiophores were difficult to find. Conidia were normally formed and ejected from the tips of the stocky hyphal bodies. The membrane or gelatinous layer which surrounds the naturally occurring conidium is often quite thick in relation to conidium size (Harper and Carner 1973). After continuous culture, this layer became very thin. Repeated attempts to infect P. includens larvae with conidia from cul tures of E. gammae were unsuccessful, and only one of 7 larvae became in fected after exposure to freshly sporulating, field collected larvae. The in fected larva died after 7 days, became mumonified, retained its shape, and became glossy black in color. On dissection, the cadaver was found to be filled with typical E. gammae resting spores as described by Harper and Carner (1973).

DISCUSSION There is considerable information on optimal conditions for growth and development of many species in the Entomophthoraceae. Members of the Entomophthoraceae vary considerably in their nutritive requirements. Coagulated egg yolk or media containing egg yolk has proven satisfactory for culturing many species (see review by King and Humber, 1981). EYA was clearly the optimal solid growth medium for E. gammae in this study. Once isolated, the fungus grew on a number of standard mycological media if supplemented with yeast extract. While resting spores are commonly produced by E. gammae under field conditions (Harper and Carner 1973, Newman and Carner 1975b), they were never produced on any medium under any of the conditions utilized in this study. Temperature requirements for growth of the Entomophthoraceae vary considerably (Gustafsson 1965, Hall and Bell, 1960, 1961). In this study, the optimal temperature for growth of E. gammae in vitro was 25°C or slightly lower. Spore germination on medium also appeared to be optimal in the same temperature range, which agrees with an earlier report by New man and Carner (1975a), who found that sporulation by E. gammae on cadavers was maximal at 21 and 26.7°C. The procedures used to determine optimal humidity conditions for spore germination served only as an indication of this requirement, as the re sponse was very low. However, it is clear that germination occurred only at or near conditions of 100% humidity, which is also in agreement with re sults of Newman and Carner (1975a). They found no germination of conidia of this species at 90% RH, only 1% at 95% RH, and 95% at 100% RH. Lack of infectivity by the cultured fungus may have resulted from the environmental conditions or techniques employed, or from a loss of infec tiousness following repeated subculturing. The low level of infectivity using fresh cadavers as the inoculum source suggests that the inoculation tech nique was inadequate. However, the gradual change in spore morphology with continued culture indicates that a selection process was occurring, possibly for an organism that was better adapted for saprophytic growth at the expense of pathogenicity. This phenomenon would reinforce findings with other pathogens which have shown the importance of maintaining minimal passage stock cultures to insure maintenance of desireable char- 250 Florida Entomologist 67 (2) June, 1984

acteristics in cultures used for experimental or production purposes. Alabama Agricultural Station Journal No. 15-83449.

REFERENCES CITED GUSTAFSSON, M. 1965. On species of the genus Entomophthora Fres. in Sweden II. Cultivation and physiology. Lantbr. Hogsk. Annlr. 31: 405-57. HALL, I. M., AND J. V. BELL. 1960. The effect of temperature on some entomophthoraceous fungi. J. Insect Pathol. 2: 247-53. ---. 1961. Further studies on the effect of temperature on the growth of some entomophthoraceous fungi. J. Insect Pathol. 3: 289-96. HARPER, J. D., AND G. R. CARNER. 1973. Incidence of Entomophthora sp. and other natural control agents in populations of Pseudoplusia includens and Trichoplusia ni. J. Invertebr. Pathol. 22: 80-5. HARPER, J. D., M. SHEPARD, AND R. M. MCPHERSON. 1983. Geographical and seasonal occurrence of parasites, predators and entomopathogens. Pages 7-19 In "Natural Enemies of Arthropod Pests in Soybeans". H. R. Petri, ed. Southern Coop. Ser. Bull. 285. 90 p. KING, D. S., AND R. A. HUMBER. 1981. Identification of the Entomoph thorales. Pages 107-127 In "Microbial Control of Pest and Plant Diseases 1970-1980". H. D. Burges, ed. Academic Press, London. NEWMAN, G. G., AND G. R. CARNER. 1974. Diel periodicity of Entomophthora gammae in the soybean looper. Environ. Ent. 3: 888-90. ---. 1975a. Environmental factors affecting conidial sporulation and germination of Entomophthora gammae. Environ. Ent. 4: 615-8. ---. 1975b. Factors affecting the spore form of Entomophthora gammae. J. Invertebr. Pathol. 26: 29-34. SHOREY, H. H., AND R. L. HALE. 1965. Mass-rearing of the larvae of nine noctuid species on a simple artificial medium. J. Econ. Ent. 58: 522-4. SOLOMON, M. E. 1951. Control of humidity with potassium hydroxide, sul furic acid, or other solutions. Bull. Ent. Res. 42: 543-53. WEISER, J. 1965. Not~s on two new species of the genus Tarichium Cohn (Entomophthoraceae). Ceska Myko. 19: 201-4. WOLF, F. T. 1951. The cultivation of two species of Entomophthora on synthetic media. Bull. Torrey Bot. Club 78: 211-20. • • • • • • • • • • • . . .. LIRIOMYZA TRIFOLII: OVIPOSITION AND DEVELOPMENT IN FOLIAGE OF TOMATO AND COMMON WEED HOSTS T. G. ZOEBISCH, D. J. SCHUSTER AND J. P. GILREATH Agricultural Research & Education Center 5007-60th Street East Bradenton, FL 34203 USA

acteristics in cultures used for experimental or production purposes. Alabama Agricultural Station Journal No. 15-83449.

ABSTRACT Oviposition and development of Liriomyza trifolii (Burgess) were studied in the laboratory in foliage of tomato, Lycopersicon esculentum Mill. cv. 'Walter,' a common nightshade, Solanum nodifiorum Jacq., com mon beggar-tick, Bidens alba (L.) DC., and downy groundcherry, Physalis pubescens L. Liriomyza trifolii oviposited successfully in foliage of all plant Zoebisch et al.: Liriomyza trifolii 251 species. However, females exposed simultaneously to all plant species de posited more eggs per cm2 of tomato foliage than per cm2 of foliage of the other plant species. Differences among plant species in percent egg hatch, percent pupation or pupal weight of L. trifolii were not significant. Larvae pupated most rapidly from foliage of nightshade.

RESUMEN Se estudi6 en el laboratorio la oviposici6n y el desarrollo de Liriomyza trifolii (Burgess) en follaje de tomate, Lycopersicon esculentum Mill. variedad 'Walter,' Solanum nodifiorum Jacq., Bidens alba (L.) y Physalis pubescens L. Liriomyza t1'ifolii oviposit6 con buen exito en el follaje de todas las especies antes mencionadas. Sin embargo, las hembras expuestas simul taneamente a todas las especies de plantas depositaron mas huevecillos por cm2 en el follaje de tomate que en las demas especies de plantas. No se encontraron diferencias significativas entre las especies de plantas con respecto a la eclosi6n, porciento de pupaci6n y peso pupal de L. trifolii. Las larvas que provenian del follaje de Solanum nodifiorum puparon mas rapidamente.

Liriomyza spp. leafminers are considered important pests of tomato although evidence indicates they are secondary in nature (Oatman and Kennedy 1976, Johnson et al. 1980). Nevertheless, in Florida, if populations are not controlled, up to 70% by weight of the 5th harvest may be lost (Schuster, unpublished data). An understanding of the role of alternative host plants in the popula tion dynamics of the pest is important for the management of any pest. Alternative host plants may serve as sources of epidemic numbers of pests or as sources of low levels of pests which may initiate and maintain infesta tions in crop plants. The most abundant leafminer attacking tomato on the west coast of Florida at the present time is L. trifolii (Burgess) (Schuster, unpublished data). The role of alternative host plants in the population dynamics of this pest is not known. The host range of L. trifolii has been summarized by Stegmaier (1966), Spencer (1973), and Spencer and Stegmaier (1973). However, no records are available regarding the relative suitability of weed hosts for L. trifolii. As a first step in elucidating the role of weeds in L. trifolii population dynamics, we compared oviposition, egg hatch and larval development of L. trifolii in foliage of tomato, eggplant and selected weeds. Nightshade, downy groundcherry and common beggar-tick were selected for evaluation because 80-95% of the Liriomyza larvae observed on weeds on the perim eters of commercial tomato fields on the west coast of Florida during the spring of 1982 occurred on these plants (Schuster, unpublished data). Pig weed was included as a less favorable host (as determined by field observa tions) .

MATERIALS AND METHODS Experiments were conducted in a room at ca. 27°C and 12 h photoperiod. Liriomyza trifolii adults for all experiments were obtained from a colony that had been maintained for at least 3 generations on Phaseolus limensi.bean by techniques similar to those of Ketzler and Price (1982). Four to 6 week old seedlings of tomato (Lycopersicon esculentum Mill., cv. 'Walter'), a common nightshade (Solanum nodiflorum Jacq.), downy groundcherry (Physalis pubescens L.), common beggar-tick (Bidens alba (L.) DC), pigweed (Amaranthus viridis L.) and eggplant (S. melongena L. cv. 'Florida Market') in 10-cm-diam. pots were used to evaluate the ovipositional response of flies. The first 4 species were used in the experiments to evaluate development. Plants of sim ilar age were used in each replication. Percentage data were transformed by arcsine V%/100 prior to statistical analysis but are presented in the original scale. Two-way analyses of vari ance were performed on all data and, if significant F values (P = 0.05) for plant species effects were detected, Duncan's new multiple range test was used to compare means. To study ovipositional preference, 1 seedling of each of the 6 plant species was randomly assigned to a position in a circle within a 61 x 61 x 61 cm cage covered with nylon organdy in a greenhouse. About 100 L. trifolii adults were released into the cage for 24-48 h after which the numbers of eggs deposited were counted with the aid of a dissecting microscope. Leaf area for each plant was measured to the nearest 0.1 cm2 with a LI-COR® Model LI-3000 area meter (Lambda Instruments Corp., Lincoln NB). The adults were fed honey 24 h prior to testing and during the test were pro vided with a 9-cm diam. filter paper dipped in honey. The experiment was replicated 9 times. To evaluate the impact of plant host on oviposition (in the absence of alternative plant choices) and development of immature stages, clip cages were constructed (Beegle et al. 1981) so that 4.5 cm2 of leaf surface could be enclosed. One male and 2 female L. trifolii were confined 24 h on the upper surface of the leaflet with a clip cage supported so that the plane of the leaflet was horizontal. One clip cage was used on each of 2 leaflets of about the same age from the middle of each plant. The adults were fed with honey at least 24 h prior to testing to enhance oviposition. Two drops of honey were also placed on the inside of each cage. Eggs were counted on each leaflet and were marked as they hatched. To prevent overcrowding, the first 3 to 5 larvae per leaflet were allowed to survive while all others were killed with a dissecting needle. In order to obtain puparia, the leaflets were covered with polyethylene bags with small ventilation punctures. Puparia were weighed to the nearest 0.01 mg with a semi-microbalance. Twelve plants were used per plant species for the oviposition, egg hatch and pupal weight measurements and 6 were used to determine the time to pupation.

RESULTS AND DISCUSSION When L. trifolii females were permitted a choice among the plant spe des, significantly more eggs were deposited in foliage of tomato, nightshade and eggplant than in foliage of the other plant species (Table 1). No eggs were deposited in foliage of pigweed. Because the leaf sizes varied among the plant species, egg counts were converted to a per cm2 basis. On an area basis, significantly more eggs were observed in tomato foliage than in foliage of the other plant species. More eggs were deposited per cm2 of nightshade foliage than per cm2 of downy groundcherry or pigweed foliage. Zoebisch et al.: Liriomyza trifolii 253

TABLE 1. OVIPOSITION OF Liriamyza trifalii ON SELECTED HOST PLANTS WHEN ADULTS WERE EXPOSED SIMULTANEOUSLY TO ALL PLANT SPECIES (CHOICE) AND WHEN ADULTS WERE CONFINED ON FOLIAGE OF SINGLE PLANT SPECIES (NO CHOICE).

Choice No choice Plant Eggs/plant Eggs/cm2 Eggs/cm2

Tomato 120.2 a l 1.4 a 5.0 a Nightshade 119.9 a 0.8 b 6.3 a Eggplant 94.7 a 0.5 bc 5.5 a Downy groundcherry 45.7b 0.1 cd 4.9 a Common beggar-tick 31.6 bc 0.5 bc 4.3 a Pigweed 0.0 c 0.0 d 0.1 b

'Means within a column not followed by the same letter are significantly different at the P = 0.05 level, Duncan's new multiple range test. When L. trifalii females were confined on foliage of individual leaflets of the plant species and thus, were not given a choice of oviposition site, sim ilar numbers of eggs were deposited in foliage of all plant species except pigweed (Table 1). Larvae from foliage of nightshade pupated in significantly less time than larvae from foliage of the other plant species (Table 2). The percent egg hatch, percent pupation or pupal weights of L. trifolii were not significantly affected by the plant species evaluated. Nightshade, downy groundcherry and common beggar-tick were demon strated to be acceptable hosts of L. trifolii in the laboratory. All 3 species may support populations of L. trifolii and serve as sources of infestation for tomato in the field. Of the 3, nightshade may serve as a source of greater numbers of L. trifalii since, when females were given a choice, more eggs were deposited on a per plant basis in nightshade foliage than in downy groundcherry or common beggar-tick foliage. In addition, less time was re quired for larvae to complete development in nightshade foliage. Manage ment of these weeds, particularly nightshade, might aid in the management of L. trifalii on tomato. However, the role of these weeds as sources of para sites of Liriamyza must first be elucidated. It must also be confirmed that

TABLE 2. DEVELOPMENT OF Liriamyza trifolii IN FOLIAGE OF SELECTED HOST PLANTS.

% egg Days to % Pupal Plant hatch pupation pupation wt. (mg)

Downy groundcherry 96.7 N.S.l 4.8 a 2 74.6 N.S. 0.48 N.S. Tomato 95.7 4.8 a 83.3 0.45 Nightshade 91.9 4.1 b 88.2 0.48 Common beggar-tick 91.1 5.0 a 89.1 0.45

'N.S. indicates no significant differences among means at the P = 0.05 level, F tests in two-way analyses of variance. 2Means within a column followed by the same letter are not significantly different at the P = 0.05 level, Duncan's new multiple range test. 254 Florida Entomologist 67 (2) June, 1984

L. trifolii reared from foliage of the weed species are not host-conditioned and will oviposit and develop on tomato.

ACKNOWLEDGMENT We wish to thank Dr. David Han for identification of the weed species utilized in these experiments. Florida Agricultural Experiment Stations Journal Series No. 4675.

REFERENCES CITED BEEGLE, C. C., H. T. DULMAGE, D. A. WOLFENBARGER, AND E. MARTINEZ. 1981. Persistance of Bacillus thuringiensis Berliner insecticidal ac tivity on cotton foliage. Environ. Ent. 10: 400-1. JOHNSON, M. W., E. R. OATMAN, AND J. A. WYMAN. 1980. Effects of in secticides on populations of the vegetable leafminer and associated parasites on summer pole tomatoes. J. Econ. Ent. 73: 61-6. !{ETZLER, L. D., AND J. F. PRICE. 1982. Methods for growers to evaluate effects of their cultural practices on Liriomyzu, trifolii leafminers in a simple laboratory. Proc. Florida State Hort. Soc. 95: 162-4. OATMAN, E. R., AND G. G. KENNEDY. 1976. Methomyl induced outbreak of Liriomyza sativae on tomato. J. Econ. Ent. 69: 667-8. SPENCER, K. A. 1973. Agromyzidae (Diptera) of Economic Importance. W. Junk B.V., The Hague. 418 p. SPENCER, K. A., AND C. E. STEGMAIER, JR. 1973. Arthoropods of Florida and Neighboring Lands, Vol. 7. Agromyzidae of Florida with a sup plement on species from the Caribbean. Florida Dept. Agric. Con sumer Serv., Div. Plant Industry, Bureau Ent. Contribution No. In. 205 p. STEGMAIER, C. E., JR. 1966. Host plants and parasites of Liriomyza trifolii in Florida (Diptera: Agromyzidae) . Florida Ent. 49: 75-80. • • • ., . • ••• • • •• METHYL BROMIDE FUMIGATION AS A QUARANTINE TREATMENT FOR LATANIA SCALE, HEMIBERLESIA LATANIAE (HOMOPTERA: DIASPIDIDAE) PETER C. WITHERELL Miami Methods Development Station USDA, APHIS, PPQ Miami, FL 33158 USA

ABSTRACT Fumigation tests using methyl bromide at normal atmospheric pressure showed that 16 glm' for 2% h at 22.8°C or above, or 32 g/m3 for 2% hat 18.3 to 22.2°C was sufficient as a quarantine treatment against latania scale, Hemiberlesia lataniae on nursery stock. For use on avocado fruits, recom mendations are for 24 g/m3 for 2% h at 22.8°C or above, and 32 glm' for 3 h at 18.3 to 22.2°C. Microscopic examination of fumigated and control scales showed that certain body movements, particularly pharyngeal pulsa tion, is a much more reliable indicator of scale viability than the mere presence of body fluids. Following fumigation, body fluids required up to 31 days to dehydrate completely, under humid conditions. 254 Florida Entomologist 67 (2) June, 1984

L. tri/olii reared from foliage of the weed species are not host-conditioned and will oviposit and develop on tomato.

ACKNOWLEDGMENT We wish to thank Dr. David Han for identification of the weed species utilized in these experiments. Florida Agricultural Experiment Stations Journal Series No. 4675.

REFERENCES CITED BEEGLE, C. C., H. T. DULMAGE, D. A. WOLFENBARGER, AND E. MARTINEZ. 1981. Persistance of Bacillus thuringiensis Berliner insecticidal ac tivity on cotton foliage. Environ. Ent. 10: 400-1. JOHNSON, M. W., E. R. OATMAN, AND J. A. WYMAN. 1980. Effects of in secticides on populations of the vegetable leafminer and associated parasites on summer pole tomatoes. J. Econ. Ent. 73: 61-6. !{ETZLER, L. D., AND J. F. PRICE. 1982. Methods for growers to evaluate effects of their cultural practices on Liriomyzu, trifolii leafminers in a simple laboratory. Proc. Florida State Hort. Soc. 95: 162-4. OATMAN, E. R., AND G. G. KENNEDY. 1976. Methomyl induced outbreak of Liriomyza sativae on tomato. J. Econ. Ent. 69: 667-8. SPENCER, K. A. 1973. Agromyzidae (Diptera) of Economic Importance. W. Junk B.V., The Hague. 418 p. SPENCER, K. A., AND C. E. STEGMAIER, JR. 1973. Arthoropods of Florida and Neighboring Lands, Vol. 7. Agromyzidae of Florida with a sup plement on species from the Caribbean. Florida Dept. Agric. Con sumer Serv., Div. Plant Industry, Bureau Ent. Contribution No. In. 205 p. STEGMAIER, C. E., JR. 1966. Host plants and parasites of Liriomyza tri/olii in Florida (Diptera: Agromyzidae) . Florida Ent. 49: 75-80. • • • ., . • ••• • • •• METHYL BROMIDE FUMIGATION AS A QUARANTINE TREATMENT FOR LATANIA SCALE, HEMIBERLESIA LATANIAE (HOMOPTERA: DIASPIDIDAE) PETER C. WITHERELL Miami Methods Development Station USDA, APHIS, PPQ Miami, FL 33158 USA

RESUMEN Pruebas de fumigaci6n con bromura de metilo a una presion atmosferica normal demostraron que 16 gjm3 por 21;2 horas a 22.8°C 0 mas, 0 32 g/m'S por 2Y2 horas de 18.3 a 22.2°C fue suficiente como un tratamiento de cuarantena contTa la escama Iatania, Hemiberrlesia lalaniae en viveros de plantas. Para nsar en frutas de aguacate, las recomendaciones son 24 g{m3 por 2Y, horas a 22.8°C 0 mas, y 32 g{m3 por 3 horas de 18.3 a 22.2°C. Examenes microsc6picos de escamas fumigadas y el control demonstraroD que movimientos del cuerpo, especialmente pulso faringeo, eS un indicador mejor de Ia viabilidad de la escama que la mera presencia de los fhiidos corporales. La perdida total de fluidos corporales tomo hasta 31 dias post tratamiento en condiciones humedas despues de fumigarse.

Latania scale, Hemiberlesia lataniae (Signoret) is widespread through out tropical and subtropical areas, and is also found in greenhouses in temperate climates (Dekle 1976, Hamon 1979, Nakahara 1982). The species is polyphagous, and is known to infest more than 300 plant species in Flor ida (Dekle 1976). It is a serious pest of palms and other ornamentals, particularly ponytail palm or elephant-foot tree (Beaucarnea recurvata Lem.) , Australian pine (Casuarina ecraiseti/olia) , loquat (Eriobotrya japonica) , and rose (Rosa spp.). Leaves, stems, and fruits may all be damaged as a result of the insects sucking plant juices. H. lataniae was best described by Ferris (1938). The armor is usually round (female) or oval (male), but may be irregular in shape when sev eral scales overlap or crowd together. The armor is strongly convex and 1.5 to 2 mm in diameter when mature. Exuviae are large, pale brown, centrally located to sub-centrally, and are surrounded by a dirty white area. When the armor is carefully removed with a pin, the soft, lemon-yellow, flattened, sac-like body is revealed. The body is attached to the host by thread-like stylets that arise near the middle of the body, and extend as long or longer than the body. Legs are absent. in all secondary stages except 1st instar nymphs (crawlers). H. lataniae is quarantined in some foreign countries. This often presents a barrier to the free flow of international commerce. During the recent out~ breaks of Mediterranean fruit fly (Ceratitis capitata (Wied.» in Cali fornia, avocado fruits originating from outside regulated areas of that state were shipped to Japan and elsewhere. Shipments received a combined treat ment of methyl bromide (MB) plus refrigeration as a precaution (i.e., MB @ 32 g{m3 for 2'h h at 21.1°C or above, followed by 7 days of storage at 7.2°C or below).' Subsequently, inspectors of the Japanese Ministry of Agriculture, Fish eries and Forestry (MAFF), reported live latania scales in 3 out of 5 sea containers of California avocados, which had received the USDA's pre cautionary treatment against Mediterranean fruit fly. The Japanese au thorities re-treated these fruits with hydrogen cyanide (HON). The Cali fornia avocado industry then requested a fumigation schedule be developed which would allow acceptance of certified fruit. A predeparture treatment

lEquivalent to 32 oz./1000 ft.3 . This is Treatment Schedule T 102 (a) (1) of the USDA Plant Protection and Quarantine Programs Treatment Manual (Anon. 1981). 256 Florida Entomologist 67 (2) June, 1984 would eliminate the necessity and cost of an additional treatment upon ar rival in Japan, and would give the avocados a longer shelf-life. To determine scale viability Japanese inspectors puncture or crush the scale, and if body fluids are present, the scale is considered to be alive. (This is the so-called "squash test".) Determining scale mortality on com modities receiving quarantine fumigation treatment presents a special prob lem. Several factors are involved, including time elapse between treatment and inspection and temperature and RH during in-transit storage. More over, the waxy scale covering may prevent large female scales from desic cating, so that when the last scales have dehydrated, the commodity has exceeded its marketable shelf-life. A more time-consuming but more accurate method of assessing scale viability is the so-called "pump test"; an unpublished technique developed at the University of California, Riverside. Naked specimens are examined microscopically for vital signs, viz., slight body movements, particularly the pumping activity of the pharynx. The experiment reported here was de signed to clarify the question of treatment efficacy by microscopically ex amining fumigated scales.

METHODS AND MATERIALS Attempts to obtain scale-infested avocado fruits from packing houses in Florida and California were unsuccessful. Moreover, since avocados are not a favored host of this insect (D. Fiskaali, California Dept. of Food & Agric., pers. comm.), securing sufficient numbers of infested fruits presented a formidable problem. Therefore, a more favored host was obtained, viz., ponytail palm or elephant-foot tree, Beaucarnea recurvata Lem. (Liliaceae). In addition, unpublished results of MB fumigations of latania scale, by the California Dept. of Food & Agriculture in 1936 through 1948, were ob tained and analysed. Twenty moderate-to-heavily infested potted ponytail palms (ca. 46 em tall) were purchased locally, a,nd scales were microscopically examined to confirm that some scales were alive. Plants were stored over night at the fumigation temperatures, in the room equipped with fumigation chambers. The following morning, plants were divided into treatment groups (lor 2 plants each). There were approximately equal numbers of scale insects in each treatment group. Fumigation trials were conducted, using the follow ing test schedules (MB at normal atmospheric pressure) : (A) 16 g/m3 for 2% h; 22.8-23.9 and 18.3°C temperatures (B) 32 g/m3 for 2% h; 22.8-23.9 and 18.3°C temperatures (C) 48 g/m3 for 2% h; 22.8-23.9 and 18.3°C temperatures (D) 32 g/m3 for 4 h; 24.4-25 and 20°C temperatures2 (E) Combination of MB fumigation plus cold treatment: 32 g/m3 for 2% h; 24.4-25 and 20°C temperatures, followed by 7 days in storage at 4.4-7.2°C3 (F) Controls (no treatment)

2This is Treatment Schedule T 105 (a) (l) of the USDA Plant Protection and Quarantine Programs Treatment Manual (Anon. 1981), except that temperature during treatment must' be 21.1°0 or above. (Recommended for avocados.) "This is Treatment Schedule T 102 (a) (1) of the USDA Plant Protection and Quarantine Programs Treatment Manual (Anon. 1981), except that the temperature during treatment must be 21.100 or above; during storage, 7.200 or below. (Recommended for avocados.) Witherell: Fumigation for Latania Scale 257

ME gas concentrations within the fumigation chambers were monitored during the course of the treatments by using a Gow-Mac gas analyser.4 Following treatment, plants were isolated from each other and placed in a greenhouse, except for plants receiving Treatment E. The latter were placed in a walk-in cold room (4.4-7.2°C, lights on). RH in the two loca tions ranged from 75 to 90%. Plants were watered (roots only) twice weekly, immediately following examination of scales. A representative sample of 100 scales per treatment was microscopically examined twice a week post-treatment over a 31-day period to document the rate of desiccation and mortality. All stages were examined, except eggs. During inspection, one or more leaves were removed from the plant, and examined under a binoccular, dissecting microscope (15X), using a 30 W spotlight. Plants receiving Treatment E (fumigation plus refrigeration) were exposed to room temperature before they were inspected. High magni fication (45 or 90X) was used to give a more detailed view. The scale cover (armor) was teased off with a pin to reveal the yellow, sac-like body be neath. The large body of mature females often had a cluster of yellow eggs andlor crawlers immediately adjacent to it. If alive, the female would con tract, expand, undulate, or twist when g~ntly prodded with a blunt object. If no movement was detected, the insect was teased from the substrate, and turned upsidedown. The triangular-shaped base of the mouthparts was ex amined for pulsating or pumping action in the pharynx of live specimens. Often, also the threadlike stylet tube moved. Scales examined were classified into 3 categories: (D) Dead (dry, with no body juices) (L) Live (showing pulsation or other body movements) (M) Moribund (body juices still intact, at least partially, but body show ing no movement when prodded) The latter category also included scales in various stages of decomposition, as well as those which were parasitized.

RESULTS AND DISCUSSION Early unpublished work by the California Dept. of Food & Agriculture on ME fumigation for quarantine control of latania scale is summarized in Table 1. All treatments were reported as 100% effective. Steinweden (1948), however, reported poor control with ME at 32 g/m3 at 15.5°C against this scale on avocados (2-h fumigation). In tests conducted in Miami, FL, the ponytail plants tolerated the treat ments well, except for Treatment D at 24.4-25°C, which resulted in ex cessive browning of the outer whorls of leaves ("tip burn"). Latania scales showed a definite preference for the upper leaf surface of ponytail palm, and congregated in the leafaxils, where the tissue is succulent. Thoroughness of treatment, as indicated by periodic monitoring of gas concentration during the course of fumigation, can be expressed in terms of a "CT product" (mean concentration of ME gas (g/m3 ) X time in hours). The CT products of Treatments A through F were calculated as follows: tests at 22.8-25°C, 38.1, 71.9, 105, 113.2, 81.9, and 0, respectively, and tests at 18.3-20°C, 35, 73.8, 106.9, 126.7, 83.1, and 0, respectively.

4Manufactured by Gow-Mac Instrument Co., Bridgewater, NJ (Model No. 20-350). t-:l 01 00

TABLE 1. METHYL BROMIDE FUMIGATION TRIALS REPORTED EFFECTIVE AGAINST LATANIA SCALE IN CALIFORNIA, 1936-1948.a

Fumigation MB dosage Exposure Post-fumigation Day of ~ Host temp. (OC) -(g/m3 ) b storage conditions mortality counts c time (h) -;] ~ Areca lutescens 26.7 40· 1.5 Not stated 7, 14, 15, & 28 t::j (palm: leaves) 26.7 32 2 1/ 21 " ;2...... Persea americana 25.6-26.7 32 2 " " 42 c (avocado: fruits) 22.8-25.6 & 26.7 40 2 4.4-7.2°C " ~ " 21.1-23.9 48 2 1/ 1/ C 1/" 1/ 1/ 1/ 1/ C 64 2 CCl- 1/ 1/ 1/ 21.1-22.2 72 2 " ~""'. Solanum tuberosumd 21.1 32 2 Rm. temp. for 35 days, 1/ ...... 0':> (potato: tubers) then 5°C for 7 days -::J 1/ 26.7 32 2 Rm. temp. for 32 days, 39 " t-:l then 5°C for 7 days '-'--- 1/ " " 40 2 Rm. temperature 42

aThese data are summarized from unpublished reports of the California Dept. of Food & Agriculture (D. A. Fiskaali, pers. comm.). The avocado data were summarized by Steinweden (1948). bg/m' is equivalent to oz./1000 ft.' CFumigated under 50.8 em of vacuum. All other tests were conducted at normal atmospheric pressure. dPotatoes are not a normal host for latania scale, and were used only for insectary rearing. ~ s:: :::s ~(t)

~ ~ 00 ~ Witherell: Fumigation for Latania Scale 259

The efficacy of Treatments A through F is shown in Table 2. At the dosages used, ME fumigation was 100% effective at 22.8-25°C. Good results even at the low dosage of 16 g/m3 for 2lh h (Treatment A) confirms the validity of earlier data from California (Table 1). At cooler temperatures (18.3-20°C), fumigation was still effective, except in Treatment A (16 g/m3 for 2lh h), where 4-25% (x = 13.1%) survival occurred. Similar scale survival was recorded in the controls (Treatment F), i.e., 1-26% (x = 12.4% survival). Eody decomposition and moisture loss occurred over an extended period of time under the humid conditions prevalent in Florida, and at the high RH at which fresh fruits and vegetables are best stored. At 31 days post treatment a few insects retained some body moisture. Among the controls examination revealed 32 to 99% (x = 73.5%) of the scales were dead (D); 1 to 62% (x = 22.4%) were alive (L); and 0 to 12% (x = 3.5%) were moribund (M). For all practical purposes scales classified as "moribund" were dead. There was no "delayed-action mortality," when lack of move ment rather than lack of moisture was used as the criterion of death. A few fumigated scales recorded as "moribund" contained a wasp larva, which appeared to be dead. As the scales decomposed and lost moisture, their bodies changed from a bright lemon-yellow color, bloated and flexible, to an amber, moist, to straw colored and dehydrated. Immature scales, especially, desiccated quickly, became dry, brittle, and dark brown in color. Fully mature female scales dehydrated more slowly than others. Occasionally the armor of dead scales sloughed off prior to inspection. Also, some of the dead scales ex amined contained an opening made by thrips or emerging wasps. In con trols, live thrips were occasionally found within hollowed-out scales. Whether these thrips were predators or merely scavengers is unknown. Color and consistency provide good evidence of mortality, but should not be the only consideration when examining freshly fumigated scales. The progression of color changes was somewhat slower among scales which had received Treatment E. This is probably a result of the preservative effect of refrigeration. Inasmuch as these scales retained their lemon-yellow color longer, warming them and checking for slight movement was deemed especially important for this group. No movement, however, was ever ob served.

CONCLUSIONS AND RECOMMENDATIONS The results of this study lead to the following conclusions and recom mendations: 1.) The "squash test" for determining viability of armored scales in the field is not an adequate test for fumigated specimens. An accurate assess ment of treatment efficacy can be made only by microscopic examination to detect slight body movements and pharyngeal pulsations. (2) During fumigation, the dosage, length of exposure, and temperature of the commodity are factors critical to the success of treatment. Higher dosages of a fumigant are required at lower temperatures, in order to effect a complete kill, because of reduced metabolism in the insects (Monro 1969, Anon. 1981). Data from California (D. Fiskaali, unpubl, Steinweden 1948) illustrated that ME at a dosage of 32 g/m3 for 2 h was completely effective t\J O':l 0

TABLE 2. EFFICACY OF METHYL BROMIDE FUMIGATION ON LATANIA SCALE, Hemibe1'Zesia Zataniae (SIG.) INFESTING PONYTAIL PALM, Beaucarnea recurvata. (MIAMI, FL, OCT.-DEC., 1982).

Viability of scales after treatmenta ':l:j 3 7 10 14-15 17 21 24 28 31 ...... Fumigation C Treatmentb temp. (OC) D L M D L M D LM D L M D L M D L M D L M D LM D L M ;;] ~ ~ A 22.8-23.9 81 o 19 91 0 9 98 0 2 98 o 2 99 o 1 99 o 1 100 o 0 99 o 1 100 o 0 18.3 91 5 4 67 23 10 70 19 11 95 4 1 94 4 2 84 11 5 77 23 0 96 4 0 74 25 1 ~ <:"i- B 22.8-23.9 99 0 1 98 0 2 97 0 3 100 0 0 99 0 1 100 0 0 100 0 0 100 o 0 99 0 1 C 18.3 93 0 7 97 0 3 99 0 1 98 o 2 99 o 1 99 o 1 100 0 0 100 o 0 100 0 0 ~ C C 22.8-23.9 80 o 20 99 0 1 100 0 0 99 0 1 100 0 0 99 0 1 100 0 0 100 0 0 100 0 0 C' 18.3 80 o 20 98 0 2 98 0 2 98 o 2 97 o 3 99 o 1 100 o 0 97 0 3 99 0 1 ~.,.,. ~ D 24.4-25 95 0 5 99 0 1 100 0 0 100 o 0 100 0 0 100 0 0 100 o 0 100 o 0 100 o 0 <:"i- 20 84 o 16 91 0 9 95 0 5 96 o 4 99 0 1 98 o 2 97 o 3 *** * * * O':l -:J E 24.4-25 - -- 90 o 10 100 0 0 100 0 0 98 o 2 100 0 0 100 o 0 100 0 0 100 0 0 20 - -- 85 0 15 93 0 7 95 0 5 99 0 1 * ** * * * ** * * * * ---t\J F (control) 24.4-25 66 24 10 90 6 4 38 50 12 47 44 9 55 40 5 32 62 6 67 21 2 69 27 4 80 18 2 -- 18.3 73 26 1 79 20 1 93 5 2 86 13 1 73 23 4 91 9 0 99 1 0 96 4 0 89 11 0

aNumbers in heading across top of table indicate the days on which scale viability was evaluated. Abbreviations: D = % dead; L = % live; M = % moribund. Percentages are based up 100 scales. "Treatments were as follows: A. B, and C: 16, 32, and 48 g/m" respectivdy, for 2% h; D. 32 g/m" for 4 h; E, 32 g/m" for 2 1h h, followed by refrigeration for 7 days at 4.4-7.2o C; F, controls (no treatment). g/m" is equivalent to oz./1000 ft." *No data, because scale supply was exhausted. <:..j -No data, because plants were undergoing 7 days of refrigeration. ~ ::l ,!i>

~ e.o 00 Il::o- Witherell: Fumigation for Latania Scale 261 at 21.1°C, but killed only 28.3% of the latania scales at 15.6°C. Based on the results of the present study, the following are the minimum schedules that can be recommended as effective against latania scale on ornamentals (MB at normal atmospheric pressure) : 16 g/m3 for 2lh h; 22.8°C or above. 32 g/m3 for 2lh h; 18.3 to 22.2°C. Air circulation should be provided so that temperatures throughout the chamber load are relatively uniform, and never below 18.3°C. (3) If the scale-infested commodity being fumigated is avocado, one should bear in mind that these fruits, like many others, tend to absorb MB gas (P. C. Witherell, unpub. data). Gas loss is even greater if the fruits are packed in corrugated fiberboard cartons or under tarpaulin. Accordingly, fumigation schedules of slightly higher dosage or longer exposure time than aforementioned would be advisable. The following schedules, therefore, are recommended for use against latania scales on avocado (MB at normal atmospheric pressure) : 24 g/m3 for 2lh h; 22.8°C or above. 32 g/m3 for 3 hj 18.3 to 22.2°C. Fruit load in the chamber or under tarpaulin should be 70% or less, cal culated by the height of the chamber or enclosure. Treatments D and E, at 21.1°C or above-both of which are prescribed for use on avocados against eggs and larvae of fruit flies (Anon. 1981)-are also 100% effective against latania scale. These schedules, however, may cause some phytotoxic damage to the fruit. Tolerance of avocado to fumigation depends, to some extent, on the variety of fruit. Varieties exhibit a wide range of tolerance to MB (Witherell et al. 1982). Normal fruit rots, particularly anthracnose, de velop faster after MB fumigation, since the ripening process is accelerated by 2 to 4 days. Vigorous brushing of fruits to remove scales is also possible, but probably impractical for large shipments.

ACKNOWLEDGMENTS I wish to thank Mr. Kenneth D. Havel (USDA, APHIS, PPQ, Hoboken, NJ), Miss Alicia Arner (USDA, ARS, Miami, FL), and Mr. Don A. Fiskaali (California Dept. of Food & Agric.) for their help with this manuscript. I also thank Mr. David Lowe (formerly with Florida DACS, DPI) for pro viding a source of infested material.

POST SCRIPT: In a recent meeting between USDA and Japanese MAFF officials in Tokyo, the data in this paper were presented. Agreement was reached that Japanese inspectors will use microscopic examination to de termine scale insect mortality in fumigated commodities.

REFERENCES CITED ANONYMOUS. 1981. Plant Protection and Quarantine Programs Treatment Manual (as revised). Sect. VI, Series T-lOO, pp. 1-29. USDA, APHIS, PPQ, Hyattsville, MD. DEKLE, G. W. 1976. Page 71 In: Florida Armored Scale Insects. Vol. 3 of: Arthropods of Florida and Neighboring Land Areas. Florida DACS, DPI, Gainesville, FL. 262 Florida Entomologist 67 (2) June, 1984

FERRIS, F. G. 1938. Atlas of Acale Insects of North America. Vol. 2, Ser. Nos. 137-268. California: Stanford Univ. Press. HAMON, A. B. 1979. Latania scale, Hemiberlesia lataniae (Signoret) (Homoptera: Coccoidea: Diaspididae. Ent. Circ. No. 208. Florida DACS, DPI, Gainesville, FL, 2 p. MONRO, H. A. U. 1969. Manual of Fumigation for Insect Control. F AO Agric. Stud., No. 79. Rome: FAO, UN, 381 p. NAKAHARA, S. 1982. Checklist of the Armored Scales (Homoptera: Diaspididae) of the Conterminous United States. USDA, APHIS, PPQ, Beltsville,tMD, 110 p. (p. 42). STEINWEDEN, J. L. 1948. Commodity treatments. Pages 236-41 In: 29th Annu. Rep., period ending Dec. 31, 1948. California Dep. Agric. Bull. 37. WITHERELL, P. C., D. H. SPALDING, AND C. A. BENSCHOTER. 1982. Tolerance of Florida avocado cultivars to methyl bromide fumigation treatments effective against fruit flies. Proc. Florida State Hort. Soc. 95: 227-9. • •• •• • • • • • • • • RESPONSE OF TRICHOGRAMMA PRETIOSUM AND T. EVANESCENSl TO WHITELIGHT, BLACKLIGHT OR NO-LIGHT SUCTION TRAPS J. H. BROWER AND L. D. CLINE Stored-Product Insects Research and Development Laboratory Agricultural Research Service, USDA Savannah, GA 31403 USA

ABSTRACT The response of Trichogramma pretiosum Riley and T. evanescens Westwood to blacklight, whitelight, or no-light suction traps in empty 44.7 m 3 rooms showed that significantly more (P0.05) was found in the re sponse of T. pretiosum to whitelight and no-light traps, while significantly more (P<0.05) T. evanescens were caught in the whitelight trap than in the no-light trap. Females, especially those of T. evanescens, appeared to be more responsive than males as more females than males of both species were usually caught. The implications of these findings should be considered in planning releases of Trichogramma for control of storage moths in warehouses.

RESUMEN EI comportamiento de Trichograma pretiosum Riley and T. evanescens Westwood hacia trampas de succion de luz negra, luz blanca, 0 trampas sin luz en cuartos vacios de 44.7 m 3 , indico que significativamente (P < 0.01) ambas especies respondieron mas a las trampas de luces negras que a las trampas de luces blancas 0 a las de sin luces. Alrededor de 40% y 60% de T. evanescens y T. pretiosum respectivamente fueron atrapadas en la trampa de luz negra. No se encontro ninguna diferencia significativa (P >

1 Hymenoptera, Trichogrammatidae 262 Florida Entomologist 67 (2) June, 1984

1 Hymenoptera, Trichogrammatidae Brower & Cline: Light Trapping Trichogramma 263

0.05) en el comportamiento de T. pretiosum hacia trampas con luz blanca 0 sin luz, mientras que significativamente mas (P < 0.05) T. evanescens fueron atrapadas en trampas de luz blanca que en trampas sin luces. Hembras, especialmente aquellas de T. evanescens, parecen responder mas que los machos, mientras que mas hembras que machos de ambas especies fueron usualmente atrapadas. Las implicaciones de este descubrimiento debe de considerarse cuando se planee soltar Trichograma para controlar polillas en almacenes.

Traditionally pests of stored-food commodities have been controlled by chemical means, but recently the potential of biological control methods has received more attention. Some studies have suggested that various species of parasitic Hymenoptera might be useful as control agents for stored product Lepidoptera (Corbet and Rotheram 1965, Reinert and King 1971, Takahashi 1973, Press et al. 1977), and it is well known that eggs of stored-product moths can serve as laboratory hosts for the rearing of Trichogramma (Alden and Farlinger 1931, Morrison and King 1977). In fact, eggs of the Angoumois grain moth, Sitotroga cerealella (Olivier) 2, are usually used to produce the large numbers of Trichogramma needed for releases to control field crop pests (Morrison et al. 1976). Thus, Brower (1983) suggested that large numbers of Trichogramma might be released periodically into warehouses to reduce or eliminate populations of stored product moth pests. However, the effectiveness of parasites in indoor situa tions has not been studied extensively and many questions need to be answered regarding warehouse releases. Many parasites and predators seem to be lured to and can often be col lected at warehouse windows and around lights (Ghani and Sweetman 1955), and Trichogramma spp. are known to be positively phototactic (Morrison et al. 1976, Martin 1969). To determine the strength of this at tractancy to light we tested the response of 2 species of Trichogramma to whitelight, blacklight, and no-light suction traps.

MATERIAL AND METHODS

EXPOSURE ROOMS

Tests were conducted in two adjacent rooms that were each ca. 44.7 m 3 • Several modifications had to be made to the rooms to make them tight enough to confine the small test insects: (1) clear plexiglass shields were built around the light fixtures and electrical outlets, (2) 295 p.-mesh Nitex® screens were placed over the heating and return air ducts, (3) special doors were constructed that could be closed tightly, (4) an anteroom connecting the rooms was built to reduce the chances of any accidental introduction of unwanted insects, and (5) all cracks and crevices were caulked and re caulked as needed between replicates. The test conditions of the rooms were kept as constant as was practical. The minimum temperature was 27* °C; the maximum temperature was not controlled, but it never exceeded 31 % 0 C. The fluorescent ceiling lights automatically turned on at 0700 h and off at 1900 h which coincided with the

"Lepidoptera. Gelechiidae 264 Florida Entomologist 67 (2) June, 1984

light-dark cycle at which the Trichogramma were reared. The relative humidity in the rooms fluctuated between 40-60%.

LIGHT TRAPS The New Jersey mosquito light traps3 used in these tests were identical except for their light sources. The whitelight (WL) trap had a 25-watt clear incandescent bulb; the blacklight (BL) trap had a 32-watt, ca. 0.3 m circline fluorescent blacklight lamp (GE FC12TI0-BL Rapid Start). The standard wire screen funnel was lined with 295 I'-mesh Nitex® screen and the collection jars contained a small amount of deobase to retain and kill the trapped insects. Both traps were modified so the light source could be turned off with the fan left running, thus creating a no-light, suction trap. One light trap was suspended from the ceiling of each room in the corner furthest from the door, so that the center of the light source was ca. 0.5 m away from the walls and the ceiling and ca. 2.1 m from the floor. The traps ran continuously during the tests and the collection jars were changed daily.

TEST INSECTS

Two species of Trichogramma were used in this test, T. pretiosum Riley and T. evanescens Westwood. T. pretio8um cultures were obtained from the Cotton Insects Research Laboratory, College Station, TX; and they origi nated from a field collection in Arkansas. T. evanescens cultures were ob tained from the Southern Grain Insects Research Laboratory, Tifton, GA; and they had originated in Poland. Both species Were reared for many generations at the Savannah Laboratory on the eggs of the almond moth, Cadra cautella (Walker)', in a controlled environment at 27±214 °C and 60±5% RH with a 1~L:12D light cycle. Large numbers of almond moth eggs were collected by placing 0- to 2-day-old adults in inverted 8.8-liter jars with screen bottoms. Eggs were removed after 24 h, sifted to remove debris, and placed in open plastic petri dishes for exposure to adult Trichogramma. The 0- to 24-h-old almond moth eggs were exposed by placing the open petri dish in a closed 8.8-liter jar along with an open petri dish containing large numbers of parasitized almond moth eggs that contained Trichogramma 0 to 24-h before emergence. The almond moth eggs were exposed to Tri chogramma adults for 8 h and then removed from the exposure chambers. One day before adult Trichogramma emergence, parasitized almond moth eggs were counted into groups of 200, and each group was glued to a labeled 9 cm filter paper disk and placed in a petri dish. Dishes containing each species were then placed in the center of either the WL or BL room, or were reserved as controls. The parasitized eggs were placed in the rooms at ca. 1400 h. Emergence of control insects held at the aforementioned con trolled conditions usually started by 0800 h the following morning. Trap catches were removed each morning for 6 days and the specimens were separated by species, sexed, counted; and recorded. The 2 species were

3Manufaetured by Hausherr's Machine Work, R. D. 1, Old Freehold !Wad, Toms River, NJ 08753. ~ Lepidoptera, Pyralidae. Brower & Cline: Light Trapping Trichogramma 265 tested simultaneously and 4 replicates were run for each light trap both with and without the trap lights on. After 6 days, the petri dishes were removed from the light-trap room and from the control cabinet, and the eggs were examined microscopically to determine percentage of Trichogramma emergence. Control adults were counted and random samples of 50 adults from each replication were sexed to determine sex ratio for each of the 2 species. Significance of the data was determined by comparing sample means with Student's t-test of paired observations.

RESULTS Parasitized control eggs averaged 97.4% emergence and a male :female ratio of 1:1.31 for T. pretiosum and 97.9% emergence and a sex ratio of 1 :1.02 for T. evanescens. Over 99% of the eggs that had emergence holes produced only a single Trichogramma adult. The average percentage of emergence in the WL and BL rooms, respectively, was 97.7 and 97.4 for T. pretiosum and 98.1 and 97.9 for T. evanescens. The sex ratios or num ber of emerged adults could not be determined in the experimental rooms. BL traps were significantly more attractive (P0.05) between the 2 types of traps for either species (Table 1). However, there was a consistent trend for the WL type of trap to catch more insects than the BL type of trap. This non-significant difference may have been caused by the greater restric-

TABLE 1. TOTAL NUMERICAL RESPONSE OF MALES AND FEMALES OF Tri chogramma pretiosum AND T. evanescens TO WHITELIGHT, BLACK LIGHT OR NO-LIGHT SUCTION TRAPS OVER A 6 DAY TRAPPING PERIOD. (MEAN OF 4 REPS)'

Response by sex Trap type Species response T. T. and light --'1'. T. pretiosum evanescens sources pretiosum evanescens (; <;> (; <;>

Lighted WL 41.8 * 25.5 16.0 * 25.8 7.5 ns 18.0 ** ** ** ** ** ** Lighted BL 123.8 * 81.1 42.0 ** 81.8 29.3 * 51.8 No-light WL 36.8 ** 13.8 19.0 ns 17.8 2.3 * 11.5 ns ns ns ns ns ns No-light BL 25.1 ** 9.3 10.8 ns 14.3 2.0 * 7.3

'Significance between pairs of values in rows or in columns is indicated by: ns = none. * = P

DISCUSSION General observations of Trichogramma spp. usually show a movement toward or aggregation near light sources (Martin 1969, Morrison et aI. 1976). However, to our knowledge no controlled studies of this phenomenon have been published until now. This study showed that T. pretiosum and T. evanescens moved several meters to a BL source, but that an incandescent WL source was only minimally attractive. More than 41 % and 63% of the emerged T. evanescens and T. pretiosum adults, respectively, were caught by the BL traps, but less than 22% of either species was caught by the WL traps. This observation supports the supposition of Costas (1941) that WL only increases the activity and dispersion of Trichogramma, but not ori ented movement. Previous observations of positive phototactic behavior of Trichogramma may have resulted from increased movement or attraction to the UV portion of most WL sources. Females of T. pretiosum were caught significantly more often than males by both BL and WL traps, but the interpretation of this finding must be tempered by the fact that the sex ratio of this species was skewed in favor of females. Equal numbers of females and males of T. pretiosum were caught in the no-light traps. Fe males of T. evanescens were caught significantly more often than males in

TABLE 2. PERCENTAGES OF Trichogramma pretiosum AND T. evanescens ADULTS CAUGHT IN WHITE LIGHT, BLACK LIGHT AND NO-LIGHT SUCTION TRAPS DURING 5 DAYS OF OPERATION.

T. pretiosum T. evanescens Lighted Lighted Combined Lighted Lighted Combined Day WL BL no light WL BL no light

1 28.7 22.2 16.2 51.0 34.3 38.3 2 62.3 67.7 48.6 23.2 55.5 51.4 3 8.4 8.5 32.0 9.8 8.3 8.7 4 0.6 1.6 3.2 1.0 1.9 1.6 5 0 0 ° 0 0 ° Brower & Cline: Light Trapping Trichogramma 267 the no-light traps. Females are apparently more active fliers than males of this species since the sex ratio was near unity in the controls. Females of other species of parasitic wasps have also been shown to be very attracted to BL sources (Hagstrum and Sharp 1975, Cline et al. 1983). A comparison of the response of T. pretiosum with that of T. evanescens showed some species specific differences. A significant difference (P<0.05), was found between the number of T. evanescens caught by the WL trap with the light on and the number caught with the light off. In contrast, there was no significant difference (P>0.05) between the numbers of T. pretiosum caught by the WL trap with the light on or off. Of course, be cause of the strong response to BL, both species were caught significantly more often by the BL trap than by the no-light trap. The release of large numbers of insectary-reared Trichogramma into commodity storages for control of moth populations has been proposed (Brower 1983). Recently, 2 species of Trichogramma have been found naturally parasitizing almond moth eggs in peanut storages in Georgia (Brower, in press). In order to maximize the chances of success, the find ings reported here indicate that BL traps should not be used concurrently with Trichogramma releases. If monitoring of moth populations is es sential then pheromone traps or perhaps WL traps could be used. It might also be desirable to eliminate or reduce UV sources such as open windows, doors, skylights, etc. to reduce Trichogramma loss to the outside. However, the warehouse should probably not be kept in total darkness because Tri chogramma may not find or parasitize their hosts in total darkness (Costas 1941, Orphanides and Gonzalez 1970, Ashley et al. 1973). If light traps are to be used to sample for the presence of or to monitor populations of Tri chogramma, then BL traps should be selected rather than WL traps be cause of the greater response to BL.

ACKNOWLEDGEMENTS The authors are grateful for the help of R. K. Morrison, Cotton Insects Research Laboratory, USDA, ARS, P. O. Box DG, College Station, TX 77841, who kindly supplied cultures of T. pretiosum and to D. A. Nordlund, Southern Grain Insect Research Laboratory, USDA, ARS, Georgia Coastal Plains Experiment Station, Tifton, GA 31793, who supplied the cultures of T. evanescens. Mention of a commercial or proprietary product does not constitute an endorsement by the USDA. Revised for publication 15 February 1984.

REFERENCES CITED ALDEN, C. H., AND D. F. FARLINGER. 1931. The artificial rearing and colonization of Trichogramma minutum. J. Econ. Ent. 24: 480-83. ASHLEY, T. R., D. GONZALEZ, AND T. F. LEIGH. 1973. Reduction in effective ness of laboratory-reared Trichogramma. Environ. Ent. 2: 1069-73. BROWER, J. H. 1983. Utilization of stored-product Lepidoptera eggs as hosts by Trichogramma pretiosum Riley (Hymenoptera: Trichogram matidae). J. Kansas Ent. Soc. 56: 50-4. BROWER, J. H. 1984. The natural occurrence of the egg parasite, T1'i chogramma, on almond moth eggs in peanut storages in Georgia. J. Georgia Ent. Soc. (In Press). 268 Florida Entomologist 67 (2) June, 1984

CLINE, L. D., B. R. FLAHERTY, AND J. W. PRESS. 1983. Response of selected parasitoids and predators of stored-product insects to whitelight or blacklight traps. J. Econ. Ent. 76: 2"98-301. CORBET, S. A., AND S. ROTHERAM. 1965. The life history of the ichneumonid Nemeritis (Devorgilla) canescens (Gravenhorst) as a parasite of the Mediterranean flour moth, Ephestia (Anagasta) kuehniella Zeller, under laboratory conditions. Proc. R. Ent., Soc. London 40: 67-72. COSTAS, L. A. 1941. The effect of varying conditions on oviposition by Tri chogramma on eggs of Angoumois grain moths. J. Econ. Ent. 34: 57-8. GHANI, M. A., AND H. L. SWEETMAN. 1955. Ecological studies on the granary weevil parasite, Aplastomorpha calandrae (Howard). Biologia 1: 118-39. HAGSTRUM, D. W., AND J. E. SHARP. 1975. Population studies on Cadra cautella in a citrus pulp warehouse with particular reference to diapause. J. Econ. Ent. 68: 11-4. MARTIN, F. J. 1969. Searching success of predators in artificial leaf litter. American Midland Natur. 81: 218-27. MORRISON, R. K., AND E. G. KING. 1977. Mass production of natural enemies. Pages 183-217. In: Ridgway, R. L., and S. B. Vinson (eds.). Biological control by augmentation of natural enemies. Plenum Press, New York. 480 p. ---, R. E. STINNER, AND R. L. RIDGWAY. 1976. Mass production of Tri chogramma pretiosum on eggs of the Angoumois grain moth. South western Ent. 1: 74-80. ORPHANIDES, F. M., AND D. GONZALEZ. 1970. Importance of light in the biol ogy of Trichogramma pretiosum. Ann. Ent. Soc. America 63: 1734-40. PRESS, J. W., B. R. FLAHERTY, AND R. T. ARBOGAST. 1977. Interactions among N emeritis canescens (Hymenoptera: Ichneumonidae), Bracon hebetor (Hymenoptera: Braconidae), and Ephestia cautella (Lepi doptera: Pyralidae). J. Kansas Ent. Soc. 50: 259-62. REINERT, J. A., AND E. W. KING. 1971. Action of Bracon hebetor Say as a parasite of Plodia interpunctella at controlled densities. Ann. Ent. Soc. America 64:' 1335-40. TAKAHASHI, F. 1973. An experimental study on the suppression and regula tion of the population of Cadra cautella (Walker) (Lepidoptera; Pyralidae) by the action of a parasitic wasp, Nemeritis canescens Gravenhorst (Hymenoptera; Ichneumonidae). Mem. CoIl. Agric., Kyoto Univ. 104: 1-12. Elvin & Sloderbeck: Key to Nabid Nymphs 269

A KEY TO NYMPHS OF SELECTED SPECIES OF NABIDAE (HEMIPTERA) IN THE SOUTHEASTERN USA

M. K. ELVINI AND P. E. SLODERBECK2 Department of Entomology and Nematology University of Florida Gainesville, FL 32611 USA

ABSTRACT A key is presented for distinguishing nymphs of the 4 most commonly recorded species of nabids (Hemiptera: Nabidae) found in row crops in the southeastern United States: Nabis roseipennis Reuter, Nabis americoferus Carayon, Tropiconabis capsiformis (Germar), and Hoplistoscelis deceptivus (Harris). Characters used to separate the species include body shape and patterns of pigmentation on the head, antennae, thorax, and legs.

RESUMEN Una clave es presentada para distinguir las ninfas de las 4 especies registradas mas comunes de nabidos (Hemiptera: Nabidae) encontrada en cultivos de surcos en el sur-este de los Estados Unidos: Nabis roseipennis Reuter, Nabis americoferus Carayon, Tropiconabis capsiformis (Germar), and Hoplistoscelis deceptivus (Harris). Los caracteres usados para separar las especies incluyen la configuraci6n del cuerpo y la forma de la pigmen taci6n en la cabeza, antena, t6rax, y las patas.

Nabids (Hemiptera: Nabidae) belonging to the tribe Nabini are some of the most common predators found in row crops in the southeastern United States (Dinkins et al. 1970, Shepard et al. 1974, Deitz et al. 1976, and Pitre et al. 1978). The 4 most commonly recorded species are Nabis roseipennis Reuter, N. americoferus Carayon, Tropiconabis capsiformis (Germar), and Hoplistoscelis deceptivus (Harris) (Dinkins et al. 1970, Neal 1974, Deitz et al. 1976, and Pitre et al. 1978). The adults of these 4 species of nabids may be identified by using keys found in a monograph on North American nabids (Harris 1928) or one of several keys to selected species (Benedict and Cothran 1975, Deitz et al. 1976, Hormchan et al. 1976, and Irwin and Shepard 1980). Unfortunately, there are no keys available for distinguishing to species the nymphs of these common preda tors. To amend this situation, we present a key to the nymphs of the 4 most common species of nabids in row crops in the southeastern United States.

FAMILY: NABIDAE Immature nabids have a long, slender, 4 segmented labium arising from the anterior part of the head. The labium is not located in a rostral groove in the gular region. They also have long, slender antennae, apical tarsal

1 Present address: Dept. of Entomology, University of Arkansas, Fayetteville, AR 72701 USA. 'Southwest Area Extension Office, Kansas State University, Garden City, KS 67846 USA. 270 Florida Entomologist 67 (2) June, 1984 claws, and usually 3-4 pair of scent glands present on the abdomen (De Coursey 1971).

SEPARATION OF NYMPHAL INSTARS The same general characteristics may be used to distinguish to species individuals in each nymphal instar. However, for some nymphal instars, there are unique characteristics that are used to separate the 4 species. Such characteristics are noted in the key where appropriate. The nymphal instar of a particular immature nabid can be determined by assessing the amount of wing pad development. Individuals in the 1st nymphal instar show no wing pad development (Fig. la, 2a, 3a, and 4a). By the 2nd nymphal instar, the fore wing pads appear as slightly rounded, lateral extensions of the dorsum of the mesothorax (Fig. 2b, 3b, and 4b). Nymphs in the 3rd instar have obvious wing pads on the dorsum of both the meso- and metathorax (Fig. 2c, 3c, and 4c). Individuals in the 4th nymphal instar have hind wing pads that cover at least the dorsum of the first ab dominal segment and extend over the dorsum of the second abdominal seg ment (Fig. 2d, 3d, and 4d). By the 5th instar, the hindwing pads extend over 1/3-1/2 of the dorsum of the abdomen (Fig. 2e, 3e, and 4e). Some immatures of H. deceptivus show little or no wing pad development; how ever, they can be distinguished from immatures of the other 3 species by their distinctive body shape (Fig. 1). The following key is intended as a field key. Most of the characters used for separating the species can be seen with the use of a hand lens; however, some characters may require finer optics. Be aware that this key may not account for the polymorphism that occurs within species, and that the drawings are not to scale.

KEY 1. Body pear shaped with posterior part of abdomen 1-1/2-3 times as wide as head a)ld thoracic regions. Body color usually dark brown to black. Fig 1. _ Hoplistoscelis deceptivu8 1'. Body elongate oval. Posterior part of abdomen usually never

a b d Fig. 1. H oplistoscelis deceptivus: 1st-4th nymphal instars, a-d. Elvin & Sloderbeck: Key to Nabid Nymphs 271

a b c d e

Fig. 2. Nabis roseipennis: 1st-5th nymphal instars, a-e.

a b c d e Fig. 3. Tropiconabis capsiformis: 1st-5th nymphal instars, a-e. The dotted lines represent color patterns of stored specimens.

more than 1-1/2 times as wide as head and thoracic regions. Body color variable, but usually light brown to orange in early instars and light brown to gray in later instars. 2 2 (1'). Distal end of hind femur with a dark pigmented band. 1st and 2nd antennal segments with pigmented bands. In 4th and 5th instars, pigmented bands may be faint or lacking on the 1st antennal segment. Fig. 2. Nabis roseipennis 2'. Distal end of hind femur and 1st and 2nd antennal segments without pigmented bands. 3 3(2'). Freshly killed specimens, all instars: Vertex of head (top of head between eyes) with distinct markings that either run transversely between the eyes or are V-shaped. In the 1st in star, vertex of head is darkened. Stored specimens: In the 1st instar, frontal sutures prominent between the eyes, frons with out pigmented, longitudinal lines (Fig. 3a). For the 2nd and 3rd instars, prothorax with considerable markings, and the 272 FlorilUJ, Entomologist 67 (2) June, 1984

a b c d • Fig. 4. Nabis amerieoferus: 1st-5th nymphal instars, a-e.

abdomen is narrow and sides parallel (Fig. 3b-3e). In the 4th and 5th instars, base of spines on front femur not darkened (Fig. 3d-3e). TropieonabUi eapsiformUi 3'. Freshly killed specimens, all instars: Vertex of head without markings or markings do not run transversely between eyes (Fig. 4a-4e). Stored specimens: In the 1st instar, frontal su tures not prominent, but frons with 2 dark longitudinal lines which may extend onto the tylus (Fig. 4a). In the 2nd and 3rd instars, prothorax with few markings, and abdomen is somewhat elongate oval (Fig. 4b-4c). For the 4th and 5th in stars, base of spines on front femur darkened (Fig. 4d-4e). ______NabUi amerieoferus

ACKNOWLEDGMENTS We acknowledge Drs. K. Yeargan, D. Habeck, and R. Sailer for critically reviewing an earlier draft of this manuscript. We also acknowl edge Ms. K. Braman for providing specimens and reviewing an earlier draft of this manuscript, and Dr. R. Hemenway, for providing specimens. We also thank Mr. S. Gross for the drawings used in the figures. Florida Agriculture Experiment Station Journal No. 4825.

REFERENCES CITED BENEDICT, J. H., AND W. R. COTHRAN. 1975. Identification of the damsel bugs, Nabis alternatu8 Parshley and N. americoferus Carayon (Heteroptera: Nabidae). Pan-Pacific Ent. 51: 170-1. DECOURSEY, R. M. 1971. Keys to the families and subfamilies of the nymphs of North American IIemiptera-Heteroptera. Proc. Ent. Soc. Wash ington 73: 413-28. DEITZ, L. L., J. W. VAN DUYN, J. R. BRADLEY, JR., R. L. RABB, W. M. BROOKS, AND R. E. STINNER. 1976. A guide to the identification and biology of soybean arthropods in North Carolina. North Carolina Agric_ Expt. Stn. Tech_ BulL 238: 1-264. DINKINS, R. L., J. R. BRAZZEL, AND C. A. WILSON. 1970. Species and rela- Elvin & Sloderbeck: Key to Nabid Nymphs 273

tive abundance of Chrysopa, Geocoris, and Nabis in Mississippi cotton fields. J. Econ. Ent. 63: 660-661. HARRIS, H. M. 1928. A monographic study of the Hemipterous family Nabidae as it occurs in North America. Ent. Amer. 9: 1-97. HORMCHAN, P., L. W. HEPNER, AND M. F. SCHUSTER. 1976. Predacious damsel bugs: Identification and distribution of the subfamily Nabinae in Mississippi. Mississippi Agric. For. Expt. Stn. Tech. Bull. 76: 1-4. IRWIN, M. E., AND M. SHEPARD. 1980. Sampling predaceous Hemiptera on soybeans. Pages 503-531 In M. Kogan, and D. C. Herzog, eds. Sampling methods in soybean entomology. Springer-Verlag, New York. 587 p. NEAL, T. M. 1974. Predaceous arthropods in Florida soybean agroeco system. M.S. Thesis. University of Florida, Gainesville, Florida. 194 p. PITRE, H. N., T. L. HILLHOUSE, M. C. DONAHUE, AND H. C. KINARD. 1978. Beneficial arthropods on soybeans and cotton in different ecosystems in Mississippi. Mississippi Agric. For. Expt. Stn. Tech. Bull. 90: 1-9. SHEPARD, M., G. R. CARNER, AND S. G. TURNIPSEED. 1974. Seasonal abun dance of predaceous arthropods in soybeans. Environ. Ent. 3: 985-8. •• • •• • • ••• • • • DAMAGE OF SUNFLOWER BY THE SOUTHERN ARMYWORM (LEPIDOPTERA: NOCTUIDAE) EVERETT R. MITCHELL Insect Attractants, Behavior, and Basic Biology Research Laboratory, Agricultural Research Service, USDA, Gainesville, FL 32604 USA

ABSTRACT Spodoptera eridania (Cramer) occurred in economic infestations in ca. 50% of the 567 ha of sunflower planted in Alachua County, Florida in the fall of 1982. Outbreaks occurred in sunflower having dense stands of pig weed (Amaranthus sp.), a primary host. After consuming pigweed, the southern armyworm larvae moved en masse to the sunflower. Differences in seed yields from sunflower heads harvested from plants with 0, 50, and 100% defoliation caused by larval feeding were highly significant. Seed yields from plants showing 50 and 100% defoliation were reduced 45 and 98%, respectively, compared to plants showing little or no defoliation. The impact of native parasites on S. eridania populations in sunflower was negligible. Parasites recovered included Opion sp., Chelonus insularis Cresson, Meteorus autographae Muesebeck, and tachinids.

RESUMEN Infestaciones de Spodoptera eridania (Cramer) causando danos eco nomicos, ocurrio en aproximadamente et 50% de las 567 ha de girasoles sembrados en el Condado de Alachue de la Florida en el otono de 1982. Brotes de infestacion ocurrioen girasoles teniendo un denso establecimiento de A maranthus sp., que es un hospedero principal. Despues de comerse las plantas de Amaranthus sp., las larvas de S. eridania se movieron en masa hacia el girasol. Diferencias en el rendimiento de semillas de girasoles cosechadas de plantas con un 0, 50, y 100% de defoliacion causado por las larvas, fueron altamente significativas. Los rendimientos de semillas de plantas con un 50 y 100% de defoliacion fueron reducidos un 45 y 98% respectivamente cuando se compararon con plantas con poca 0 sin defoliacion.

Elvin & Sloderbeck: Key to Nabid Nymphs 273

EI impacto de parasitos nativos en poblaciones de S. eridania fue neglegible. Los parasitos encontrados incluyeron Opium sp., Chelonus insularis Cresson, Meteorus autographae Muesebeck, y tachinedos.

Increasing production costs, high land prices, and low commodity prices are causing farmers to seek more effective means of using machinery and land resources in the production of crops such as corn, soybean, and sorghum. Double cropping is gaining favor as one way of spreading fixed costs over a longer growing season. Double cropping also provides op portunities for more efficient utilization of fertilizer where a second crop uses the residue from the first crop. Production of more than one crop on the same land each year also reduces the financial risks for growers who may experience losses in one crop because of inclement weather conditions, insects, or other natural disasters. Growers in north-central Florida are experimenting with sunflower, Helianthus annuus L., as an alternate crop in double cropping systems. Sunflower is resistant to frost, and yields are high when planted the first half of February in the Gainesville, Florida area. When sunflower is planted in early February in northern Florida, the crop escapes severe losses by diseases and insects. Sunflower planted in August also shows promise as a late-season crop in north-central Florida because late plantings usually are not seriously damaged by mildew or rust, although the crop may have heavy infestations of the corn earworm, Heliothis zea (Boddie), and the sunflower moth, Homoeosoma electellum (Hulst) (Bailey et al. 1978). This paper reports outbreaks of the southern armyworm, S. eridania (Cramer), in sunflower in Alachua County, Florida, in October 1982. This is the first documented report of this species as a pest of commercial sun flower. The impact of plant defoliation by southern armyworm larvae on sunflower yields also w~s evaluated.

MATE~IALSAND METHODS Approximately 567 ha of sunflower was grown in Alachua County in 1982, 50% of which were infested with southern armyworm larvae. Two fields infested with southern armyworm were selected as study sites. Each field was ca. 16 ha and was adjacent to cultivated sunflower on at least one side. Study fields were separated by ca. 5 km, and were planted to sunflower (cv. 'Dalgren 164') between 25 August (field A) and 28 August (field B). Initial observations were made on 20 October when plants were ca. 1.75 m in height and in full bloom (i.e., ray petals fully exposed). The plants were already infested with larvae when first examined. Approximately 450 larvae were collected from each field, classified as small (..e:::0.8 cm), medium (0.9-3.4 cm), or large (>3.5 cm), and placed individually in a 30 ml size plastic cup containing sufficient diet (Leppla et al. 1979) for complete larval development. Cups were checked several times a week, and the fate of each larva recorded-larva died before pupation, pupa died, moth emerged, or parasite emerged. Degrees of plant defoliation by southern armyworm larvae on sunflower head size (1 head/plant on 'Dalgren 164') was determined by measuring the width of 10 consecutive mature heads at 10 randomly selected sites in dif- Mitchell: Southern Armyworm on Sunflower 275

ferent areas of the field. Estimates of defoliation were made within 7 days after most (>98%) of the larvae had matul'ed and dropped from the plants to the soil to pupate. Three defoliation levels were estimated: 0-5%-little or no feeding damage on any leaves; 50%-nearly all leaves showing signs of heavy feeding with many large holes and some leaves almost totally eonsumed; 100%-all green leaf tissue consumed with only major leaf veins and the stalk remaining (Fig. 1). Groups of 10 consecutive plants that most nearly fit the leaf damage criteria for each of the 3 categories were selected and marked for evaluation of head size and yield when the plants matured. A total of 100 heads (10 replicates of 10 heads each) were measured for f:lach of the 3 defoliation levels. The effect of plant defoliation on seed yield was determined by harvesting 10 consecutive mature heads from the same areas from which the head measurements were taken. Heads were air dried, hand threshed, and weighed. The data were analyzed using Student's t-test.

RESULTS AND DISCUSSION Most of the larvae collected from fields A and B on 20 October were medium-sized. Only a few were classified as small or large (ca. 10% for each class). Examination of other infested fields in Alachua County revealed a similar larval age, indicating that infestations apparently resulted from a general flight of adults that occurred simultaneously over the entire area. The southern armyworm is a polyphagous species that attacks a wide variety of weeds and cultivated crops (tomatoes, potatoes, celery, sweet potato, and various flowering ornamentals) (Kimball 1965). Pigweed, Amaranthus sp. (Amaranthaceae), is a favored host of the southern army worm (Tingle et al. 1978). The severe armyworm infestations observed in study fields A and B, as well as elsewhere in the area, were directly asso ciated with pigweed infestations of cultivated fields. In those areas of the fields where armyworm densities were greatest,

Fig. 1. Defoliation levels used for assessing impact of southern army worm feeding on sunflower (left to right) : 0-5% damage, 50% damage, and 100% damage. 276 Florida Entomologist 67 (2) June, 1984 larvae had consumed all but the stems of pigweed before moving to the surrounding sunflower. Ten to 25 medium and large larvae/sunflower were common throughout the infested areas when first observed. Virtually all sunflower in field A, and an estimated 70% in field B, had a high level of defoliation (50-100%). After larvae consumed the sunflower foliage, they moved to the head. Most head feeding was confined to the soft tissue on the backside and periphery, but larvae chewing through the backside into the head was rare. Heads were generally not totally destroyed by feeding larvae except when the initial attack occurred during the late budding stage. Sunflower defoliation had a significant effect on head size and on weight of the seed produced (Table 1). Severe defoliation affected seed weight much more than head size. Reduction in yield at the 50% defoliation level reflected a reduced capacity on the part of the sunflower plant to furnish the nutrients required to fill its seed, as was evident by the large number of fully formed but "light" seed. Plants classified as 100% defoliated produced only small heads with few or no seed. The impact of native parasites on southern armyworm densities in sun flower was negligible, as is often the case with insect pests occurring in out breaks. Of the 996 larvae collected and held in the laboratory, 426 produced adults (ca. an equal number of males and females), 159 produced parasites, and 411 died from unknown causes. Parasites recovered included Ophion sp. (16), Chelonus insularis Cresson (6), Meteorus autographae Muesebeck (75), and tachinids (62). Forty-seven larvae yielded 1 fly each; 2 or more flies were recovered from 15 larvae. Counting only live larvae that produced adults or parasites, ca. 27% of the larvae were parasitized. In conclusion, the southern armyworm can be a serious pest of sunflower in fields where pigweed is allowed to grow in the crop. Egg masses or newly hatched larvae were never found on sunflower in any field in this study,. which indicated that. the southern armyworm probably did not oviposit directly on the sunflower. Therefore, the most obvious control measure for

TABLE 1. EFFECTS OF DEFOLIATION BY SOUTHERN ARMYWORM LARVAE ON HEAD SIZE AND YIELD OF SUNFLOWER (CV. 'DALGREN 164'). ALACHUA COUNTY, FLORIDA. OCTOBER-NoVEMBER, 1982.

Defoliation X Headdiam. X Seed wt/head level (%) (mm ± S.E.)l (g ± S.E.)l

Field A 0-5 50 149.5 ± 3.2 a 31.8 ± 2.0 a 100 117.5±3.7b 8.1 ± 1.1 b Field B 0-5 131.7 ± 3.5 a 74.6 ± 4.1 a 50 123.5 ± 4.0 a 40.7 ± 3.6 b 100 84.1 ± 3.1 b 1.2 ± 0.3 c

'Means followed by different letters in the same column differ significantly at the 1% level, Student's t-test. Mitchell: Southern Armyworm on Sunflower 277 this sunflower pest involves early destruction of pigweed in and around fields.

ACKNOWLEDGMENT I thank Bill Copeland and Ron Hines for their assistance in locating infested fields, collecting and rearing larvae, cleaning and weighing sun flower seed, and summarizing data. Revised manuscript received for publica tion 27 January 1984.

REFERENCES CITED BAILEY, B. A., E. B. WHITTY, AND V. E. GREEN, JR. 1978. Agronomy facts: Sunflower production in Florida. Florida Coop. Ext. Servo Publ. No. 74, University of Florida, Gainesville. 4 p. KIMBALL, C. P. 1965. Lepidoptera of Florida. Div. of Plant Industry, Florida Dept. of Agric., Gainesville, FL. 363 p. LEPPLA, N. C., P. V. VAIL, AND J. R. RYE. 1979. Mass rearing and handling techniques for the cabbage looper. Pages 57-75 In Proc. Radioisotopes and Radiation in Ent., Training course, FAO/IAEA. TINGLE, F. C., T. R. ASHLEY, AND E. R. MITCHELL. 1978. Parasites of Spodoptera exigua, S. eridania [Lep.: Noctuidae] and Herpetogramma bipunetalis [Lep.: Pyralidae] collected from A maranthus hybridus in field corn. Entomophaga 23: 343-7. • • • •• • • • • • •• • POPULATION FLUCTUATIONS OF THE SCHAUS SWALLOWTAIL (LEPIDOPTERA: PAPILIONIDAE) ON THE ISLANDS OF BISCAYNE BAY, FLORIDA, WITH COMMENTS ON THE BAHAMAN SWALLOWTAIL WILLIAM F. LOFTUS AND JAMES A. KUSHLAN National Park Service, South Florida Research Center P.O. Box 279, Homestead, FL 33030 USA

ABSTRACT Censuses of the Schaus swallowtail butterfly (Heraelides (= Papilio) aristodemus poneeanus (Schaus», a federally listed threatened species, were conducted from 1979 to 1982 on islands in Biscayne Bay, southern Florida. The population size was small during 1979 and 1980, but adults were widely distributed in suitable habitat. Emergence season was similar in both years, from late April to late June. Numbers of the Schaus swallow tail were lower in 1981, and the emergence period seemed to be retarded, probably due to the very dry winter and spring. Following the wet winter of 1981-1982, the largest numbers of adult Schaus swallowtails since 1972 were recorded. It appears that the severity of the winter dry season may determine the population size of this species in southern Florida. In combination with results of previous censuses, these data provide a continuous eleven-year record of Schaus swallowtail population fluctuations on the Biscayne Bay islands. Human disturbances, especially habitat de struction, have reduced the numbers and range of the Schaus swallowtail, so the butterflies on these islands presently form the nucleus of its population

Mitchell: Southern Armyworm on Sunflower 277 this sunflower pest involves early destruction of pigweed in and around fields.

RESUMEN Desde 1979 al 1982, en las islas de Biscayne Bay en el sur de la Florida se hizo un censo de la mariposa, Heraclides ( = Papilio) OIf'istodemus poneeanus, la cua! el gobierno federal la ha declarado como una especie amenazada de extincion. EI numoro de la poblacion durante 1979 y 1980 fue minima, pero los adultos estaban bien distribuidos en habitaci6n ade cuada. La estaci6n de emersi6n fue muy similar durante los dos alios, desde finales de Abril al final de Junio. Los numeros de la mariposa Schaus fueron mas bajos en 1981 y el perfodo de emersion parece haber sida atrasado, posiblemente debido al invierno y la primavera seea. Luego del invierno humedo de 1981-1982, se registr6 el mayor numero de la mariposa Schaus desde el 1972. Asi que parece que la severidad de la estaci6n seea del invierno puede determinar el tamano de la poblacion de esta especie en nuestra area de estudio. Junto a otros censos estos datos proveen un record continuo de once anos de las fluctuaci6nes en Ia poblaci6n de la mariposa Schaus en los cayos de la Bahia de Biscayne. Los disturbios humanos1 especialmente la destruc cion del ambiente han reducido los numeros y el area geografica de la mariposa Schaus y por 10 tanto las mariposas en estas islas forman el Dueleo de la poblaci6n de este especie en los Estado Unidos. La mariposa Heraclides (=Papilio) andraemon bonhoteino Se observ6 durante los censos, y dudamos que est. establecida alii.

The Schaus swallowtail (Heraelides (~ Papilio) aristodernus poneeanus (Schaus» was first described at the turn of this century from the Miami area of Dade County in southern Florida (Schaus 1911). It was eliminated from that site as the 'city developed but was rediscovered further south on the Florida Keys, Monroe County. The species received much attention fol lowing the 1935 hurricane because of a report of its extirpation on the keys by the storm (Grimshawe 1940). Henderson (1945a, b) later documented the continued existence of the Schaus swallowtail in the Florida Keys. By this time, the butterfly had become a glamour species in great demand by collectors, and, during the next two decades, both Klots (1951) and Kimball (1965) noted that over-collection of this rare species was possible. In 1972, Covell and Rawson (1973) and Brown (1973a, b) independently surveyed the upper Florida Keys, including the Elliott Key group in Biscayne Bay, Dade Gounty, to the north of the main-line Florida Keys. They found that the Schaus swallowtail was well-established on these northern islands where Brown (1973a) reported seeing 100 adults in a day. From 1973 to 1976, Covell (1977) continued to observe this species on the islands of Biscayne Bay but found that it had decreased in numbers from the levels of 1972. He speculated that the decrease may have been the result of winter drought conditions that inhibited new growth of the larval food plants during those years. Limited surveys were also conducted during 1977 and 1978, with few adults being observed on the keys (J. Tilmant, pers. comm.). The Bahaman swallowtail (fle1"aclideR (= Papilio) o:ndraemon bonhote:i (Sharpe» has been recorded from southeastern Florida and the Florida Loftus & Kushlan: Schaus Swallowtail on Biscayne Bay 279

Keys on several occasions during this century. Only on a few occasions has the existence of a breeding population been suggested. The collection of a fresh adult on Long Key in Monroe County, Florida was used to support such a contention (Kimball 1965), as was a more recent report by Brown (1973a, b) of large numbers of adults, including ovipositing females, on the islands in Biscayne Bay. The two swallowtail butterflies became the first invertebrates to be included on the U. S. List of Endangered and Threatened Species. The overlapping range of these species in the United States was thought to be limited to tropical hardwood hammocks of the upper Florida Keys, includ ing those in southern Biscayne Bay, where environmental conditions are suitable for hammock growth. Such hammocks are normally found on the highest land in southern Florida and the Florida Keys, usually near the coast where rapid urban and commercial development has taken place. Habitat destruction has been devastating to the butterflies, especially con sidering that their population sizes and ranges in Florida appear to have been small under natural conditions. Little biological data existed for these butterflies at the start of our study. Our purpose was to determine the status and distribution of these species on the relatively undisturbed islands in Biscayne Bay, to document the season and the duration of adult emergence, to confirm the larval food plants, and to define those factors that affect the population status of the butterflies. Preliminary results from 1979-1981 censuses and suggested management options for the Schaus swallowtail were presented by Loftus and Kushlan (1982). A companion study by Gerold Morrison (1981) ex amined food plants and larval ecology. Both species of butterflies, originally placed in the genus Papilio, have recently been reassigned to the genus H eraclides (Miller and Brown 1981) .

STUDY AREA We studied two islands in Biscayne Bay, Elliott Key and Old Rhodes Key, because of their mature hardwood hammocks in which many swallow tails had been sighted in recent years. On Elliott Key, we established an approximately 1-km transect south of Billy's Point, near Petrel Point (Fig. 1). This route crossed the island from the ocean to the bay through mature hammock, included a section of the cleared central portion of the island, and looped southward into a hammock on the bay side. The transect included large stands of the host plants for Schaus swallowtail larvae, torchwood (Amyris elemifera) and wild lime (Zanthoxylu1'n fagara) (Rutkowski 1971), and an old grove of key limes (Citrus aurantifolia) , the reported larval host plant of the Bahaman swallowtail (Klots 1951). A similar transect on Old Rhodes Key was an approximately 1-km loop through a mature hammock on the southern tip of the island (Fig. 1). This hammock had an unbroken canopy, except where trees had fallen and hosted large stands of torchwood and wild lime trees. On the surveys in 1980, additional monthly visits were also made to an abandoned key lime grove at the northern tip of Totten Key (Fig. 1). The lime trees were mixed with encroaching hammock species, resulting in a more open canopy than in the mature hammock. Rainfall on the keys follows a pattern similar to that of the mainland, 280 Florida Entomologist 67 (2) June, 1984

TRANSECT ROUTES •

ELLIOTT KEY MARINA N t BISCAYNE BAY

D"'[Sf ARSI!~tCkER

CAESAR CREEK ~LON' "SEN'eK,.

~ FAn OARS'ENICl(ER OCEAN

o I ~ 3 KM

Fig. 1. Map of the major keys in Biscayne National Park showing the locations of the survey transects. with most precipitation occurring during summer and fall (Fig. 2). Winter and spring are normally times of low rainfall, although this pattern can vary considerably from year to year.

METHODS Monthly censuses of the transects were conducted from late March to late September in 1979 and 1980, except for March 1979 when boat trouble Loftus & Kushlan: Schaus Swallowtail on Biscayne Bay 281

320 385.1 300 ~ 280 260 240 220t E 200 195.6 E 180 ...J ...J 160 ~ 147.3 z 140

RESULTS The peak of the flight season of the Schaus swallowtail was mid-to-Iate May (Table 1), although adults were observed from late April (G. Morri son, pers. comm.) to mid-July (D. Peters, pers, comm.). Adults emerged in early May; all individuals present on the transects at that time looked very fresh and untattered. By late May and June, nearly all adults were tattered 282 Florida Entomologist 67 (2) June, 1984 and dull. The exception to this pattern was the sighting of three freshly emerged adults on 23 June, 1980 on Totten Key. In 1979 and 1982, adults were found on the transects only during May (Table 1). Limited data from the 1983 season showed a similarily restricted emergence period, with 8 adults located on 8 May and 15 adults seen on 15 May (C. Hauke, pers, comm.). The flight seasons in 1980 and 1981 were more protracted, extending into June and July respectively (Table 1). The number of adults seen on the transects was lowest in 1981 and the emer gence period appeared to be retarded. In 1982, the highest numbers of Schaus swallowtails during the four years of censusing were recorded (Table 1). The limited data from 1983 suggest that this was also a year of high population levels.

TABLE 1. DATES OF CENSUSES AND NUMBERS OF OBSERVATIONS OF INDIVIDUAL SCHAUS SWALLOWTAIL ADULTS ON TWO BISCAYNE BAY ISLAND TRANSECTS FROM 1979-1982.

Elliott Key Old Rhodes Key

1979 19 March 0 12 April 0 0 24 May 5 7 19 June 0 0 25 July 0 0 30 August 0 0 27 September 0 0 1980 31 March 0 0 22 April 0 0 5 May 2 2 20 May 2 4 23 June 2(1 0 17 July 0 0 25 August 0 0 29 September 0 0 1981 -- 24 April 0 0 13 May 2 0 28 May () 0 10 June 1 0 25 June ;{ 0 8 July 1 1 1982 -- 20 April 0 0 7 May 2 9 15 May 25b 10.June 0 0 7 July 0 0

"3 additional adults seen on Totten Key. hl1 additional adults seen elsewhere on Elliott Key. Loftus & Kushlan: Schaus Swallowtail on Biscayne Bay 283

Surveys of the keys in Biscayne National Park during peak emergence times revealed that the Schaus swallowtail is widespread. It appears to be limited to hammock and hammock edge habitats. We observed specimens on Elliott, Old Rhodes, Adams, and Totten Keys and assume that this species is present in suitable habitat throughout the Park. We rarely observed mating or oviposition, although courtship activity was common. On 24 May 1979 we watched a Schaus swallowtail oviposit on wild lime (Zanthoxylum). Efforts to locate Schaus swallowtail larvae on new torchwood leaves were usually unsuccessful, although fresh feeding signs were often visible. We were able to find and photograph only a few larvae during this study. Fresh feeding sign was most apparent in May and early June, with a few weeks of adult emergence. The few larvae discovered along the transects were also located during May. We never found the Bahaman swallowtail on the keys in Biscayne Bay, in spite of attempts to locate this species. We planned our census times to include the times of Brown's (1973a, b) sightings. The transects included old key lime plantings, and if this butterfly was present, we would have ex pected to find it there. Dennis Leston (pers. comm.) and Gerold Morrison (pers. comm.) were studying butterflies on the keys at the times of our censuses and did not find this species. Bahaman swallowtails were not ob served in 1981 or 1982 surveys on these keys (D. Peters, pers. comm.; C. V. Covell, Jr., pers. comm.).

DISCUSSION

BAHAMAN SWALLOWTAIL The specimen history of the Bahaman swallowtail in Florida has been sporadic. Some records appear dubious and few provide any evidence that the butterfly has been established (Klots 1951, Kimball 1965, Florida Game and Fresh Water Fish Commission 1982). Records of H. andraemon in Florida appear to be the result of sporadic colonizations during which short term breeding colonies may form, only to eventually die out (Miller 1975, Morrison 1981). Morrison (1981) pointed out that current taxonomic in terpretation of H. andraemon does not recognize a distinct Florida subspecies. This would indicate that if establishment had ever taken place, it has not been of sufficient duration for morphological differentiation to have de veloped. The sightings by Brown (1973a) and his students provided the impetus to list the Bahaman swallowtail as a threatened species. They reported the presence of numerous adults on Elliott Key in 1972, including females that were ovipositing on citrus trees. Covell (1976) was collecting on Elliott Key in 1972, and, although he collected Schaus swallowtails, he did not observe the Bahaman swallowtails. Covell (1977) continued to collect in the park area from 1973-1976 but never observed H. andraemon bonhotei. This species was not observed during surveys of Elliott Key in 1977-1978 (J. Tilmant, pers. comm.) or during our censuses. We also examined a number of key lime trees for larvae during each month of the 1979-1980 censuses, but we never found H. andraemon larvae. We believe that the specimens reported by Brown (1973a, b) failed to establish a breeding colony in Florida. There appears to be an inconsistency in identification in Brown's papers 284 Florida Entomologist 67 (2) June, 1984

(Brown 1973a: 139, and Brown 1974: 11). Recounting what seems to be the same experience, Brown (1973a) stated "on one occasion a P. aristodemus ponceanus eluded our pursuit and flew safely from the south end of one key and entered the jungle on the northeast side of an adjacent key in a route that covered over one-half mile," while in Brown (1974), he stated in identical wording that H. andraemon bonhotei eluded capture. Both Brown (1973a, 1974) and Covell (1977) use this incident to show that H. aristodemus ponceanus can fly between islands; however, it would seem that the identity of the species is unclear. Based upon the results of our work and of others, we do not believe that sufficient evidence exists to establish the permanent residency of the Bahaman swallowtail in the United States. The Bahaman swallowtail does not appear to be in danger of extirpation in its native Bahaman Islands, and it has become numerous in Jamaica where it was recently introduced (Tyler 1975). Under the 1978 amendments to the Endangered Species Act regarding invertebrates, this subspecies should not be listed unless it can be shown to be in danger throughout a significant portion of its range. Until the existence of established breeding populations in this country can be verified, it must be assumed that the occurrence of this butterfly in the United States is sporadic.

SCHAUS SWALLOWTAIL The Schaus swallowtail is adapted to life within the shady tropical ham mocks of the Biscayne Bay keys and upper Florida Keys. Its normal popu lation size appears to be low in all life stages, although the numbers of adults may follow a cyclical pattern of unknown periodicity. Large numbers of adults were present throughout its known range on the Biscayne Bay keys from 1969-1972 (Brown 1973, Covell and Rawson 1973). The popula tion levels were low during the next nine years but rebounded in 1982 and 1983. In both 1979 and 1980, adults occurred in small numbers along the transects. The slightly lower numbers observed in 1980 do not necessarily represent a decline but may be an artifact of the census method. The emergence season in both years extended from late April to June and cor responded well with the times of emergence reported by Covell and Rawson (1973) and Brown (1973). In 1981, the number of Schaus swallowtails recorded along the transects was lower than in 1979 and 1980 (Table 1). This decrease may have been the result of high larval mortality in 1980 (Morrison 1981) or could have been related to the severe spring dry season of 1981. The 1981 data indicate that spring droughts may affect the butterfly and its larval food plant by delaying emergence and new leaf production, respectively. On 13 May 1981, torchwood trees were just beginning to produce new leaves and many trees showed no new leaves at all, while in previous years, most torchwood trees were fully flushed with new growth by late April. A coincident delay in the emergence of Schaus swallowtail adults occurred in 1981. The first adults were recorded on 13 May 1981 and not on earlier censuses in late April and early May. The last sighting of adults occurred in mid-July 1981. The emer gence period of the Schaus swallowtail appears to have been delayed by several weeks in 1981, coinciding with the delayed flush of new leaves by Loftus & Kushlan: Schaus Swallowtail on Biscayne Bay 285 torchwood. Sightings in July 1981, several weeks later than the final sight ings in 1979 and 1980, suggest that the entire emergence season was re tarded in 1981. Other lepidopteran species along the transects showed sim ilar delays in emergence in 1981 when compared with the previous two years (D. Peters, pers. comm.). It appears that the severe spring dry season may have been responsible for the retardance of emergence in 1981, but this relationship requires much more study. The number of adult swallowtails increased greatly in 1982 (Table 1). RaiJ1lfall during the winter and spring of 1982 was relatively high (Fig. 2) which may have contributed to the greater survival of individuals. Adult emergence began in early May, but did not extend into June, resulting in a very short flight season. Numbers of other butterfly species on the islands was also high during 1982 (D. Peters, pers. comm.). The factors leading to changes in abundance are not well understood, but Covell (1976) suggested that severe droughts may have deleterious effects on the timing of emergence and numbers of adult Schaus swallowtails. There is evidence that the pupae can overwinter for two years during a severe drought (Covell 1976), a capability which may have been an important evolutionary factor for this species (Morrison 1981). Droughts occur naturally in southern Florida and, even during relatively wet years, there is a period in winter and early spring when rainfall is low (Fig. 1). It appears that H. aristodemus ponceanus emergence coincides with the beginning of leaf flush in the larval host plant, torchwood. Our census re sults indicate that rainfall during the months immediately preceding the usual emergence time may affect both the timing and duration of emergence and the number of adults that will emerge. This relationship had been previously suggested by Covell (1976), Miller (1975), and was discussed by Morrison (1981). In 1981, total rainfall during the period of February through April (80.7 mm) was the lowest recorded for the four survey years (Fig. 2), and the number of Schaus swallowtail adults was correspondingly low on the transects. In addition, the timing of emergence was delayed and the entire flight season was retarded. Rainfall from February through April of 1982 (337.6 mm) exceeded that of any similar period during the four survey years (Fig. 2), and butterfly numbers were also the highest re corded. Similarly, in February through April in 1983, when rainfall totaled 396.5 mm, limited censusing data showed high numbers of adults present in May. We usually observed fresh herbivory on torchwood leaves within one or two weeks after the emergence of adult Schaus swallowtails. The adults presumably mate and lay eggs soon after emerging. From our lack of re sults in locating larvae, it appears that few larvae are either produced or survive, and those that do survive through several instars often disappear before pupating (Morrison 1981). Much of the life cycle of the Schaus swallowtail is spent in the pupal stage, in which it can remain in diapause for up to two years (Grimshawe 1940). Covell (1977) discussed the possibility of a second emergence of Schaus swallowtail adults in late summer. The latest sighting of adults during four years of censusing occurred in mid-July 1981, even though censuses were continued into August or September. Food for early instar larvae produced by a second emergence would probably be scarce because they feed on new 286 Florida Entomologist 67 (2) June, 1984

leaves which are primarily produced in spring (Morrison 1981). This source would be less available later in the year. It seems more likely that the late season records of adults represent sporadic emergences, possibly resulting from asynchronous termination of diapause. Several natural and human factors, such as freezes, insecticides, and collecting, may affect Schaus swallowtail populations (Covell 1976). Such factors are more serious on the upper Florida Keys than on the islands we studied. Collecting, destruction of habitat, and insecticide spraying are minimal on the islands in Biscayne National Park; however, good habitat outside the park, especially on Key Largo, is constantly being degraded by aerial mosquito spraying or lost through destruction of hammock for de velopment. The amount of collecting outside of the park is unknown, but poaching inside the park is discouraged by the isolation of the keys and the patrolling by park rangers during the emergence period. Freezes do not seem to be as damaging on the islands as they are on the mainland. During the hard freeze of 1977, no adverse effects on vegetation or wildlife occurred on the islands in Biscayne Bay because of the moderat ing effect of the surrounding waters (J. Tilmant, pers. comm.). Hurricanes can severely affect vegetation and insect populations in south Florida (Grimshawe 1940; Craighead 1971). A direct pass of a hurricane over the Biscayne Bay keys could devastate the remaining population of the Schaus swallowtail through wind damage, flooding, and reduced availability of the food plant for several seasons. Such an event could extirpate the Florida subspecies because its geographical range is now limited, and because popu lations on the upper Florida Keys from which recolonization might occur have been severely reduced. The butterflies on the islands of Biscayne Bay, located in Biscayne National Park, appear to form the nucleus of the Schaus swallowtail popu lation in Florida at this time. Little evidence of breeding on the major Florida Keys outside the park was found in 1980 (Morrison 1981), and recent development pressures on northern Key Largo will result in the further diminution of suitable habitat. The islands in the park therefore represent the last undisturbed and unthreatened Florida habitat for this insect. ACKNOWLEDGEMENTS This survey was sponsored by The National Park Service, South Florida Research Center. Biscayne National Park personnel provided logistical sup port. We are very grateful to James Tilmant, Daniel Peters, Darcy Bailey, and Craig Hauke of Biscayne National Park for their cooperation in this project. Mr. Tilmant shared information on butterfly distribution and abun dance on the keys in 1977 and 1978. Mr. Peters, Ms. Bailey, and Mr. Hauke accompanied us on several surveys, provided data from 1981-1983 surveys, and supplied rainfall data. Dr. Charles V. Covell and others kindly offered information on their observations of Schaus swallowtails on the keys. We especially thank Gerold Morrison for sharing his data on distribution and larval mortality during 1980, and Oron Bass for reviewing this paper.

REFERENCES CITED BROWN, L. N. 1973a. Populations of Papilio andraemon bonhotei Sharpe Loftus & Kushlan: Schaus Swallowtail on Biscayne Bay 287

and Papilio aristodemus ponceanus Schaus (Papilionidae) in Bis cayne National Monument, Florida. J. Lepid. Soc. 27: 136-140. ---,. 1973b. Populations of a new swallowtail butterfly found in the Florida Keys. Florida Nat. 46(2): 25. ---,. 1974. Haven for rare butterflies. National Parks and Conserv. Mag. July 1974: 10-13. COVELL, C. V., JR. 1976. The Schaus swallowtail: a threatened subspecies? Insect World Dig. 3: 21-26. ---. 1977. Project Ponceanus and the status of the Schaus swallowtail (Papilio aristodemus ponceanus) in the Florida Keys. Atala 5: 4-6. ---, AND G. W. RAWSON. 1973. Project Ponceanus: A report on first efforts to survey and preserve the Schaus swallowtail (Papilionidae) in southern Florida. J. Lepid. Soc. 27: 206-210. CRAIGHEAD, F. C., SR. 1971. The trees of south Florida. Vol. 1. University of Miami Press, Coral Gables, Florida. FLORIDA GAME AND FRESH WATER FISH COMMISSION. 1982. Draft recovery plan for the Schaus Swallowtail Butterfly (Heraclides (Papilio) aristodemus ponceanus (Schaus» including recommendations for Bahaman Swallowtail Butterfly (Heraclides (Papilio) andraemon bonhotei (Sharpe». Report to U.S. Fish and Wildlife Service, At lanta, Georgia. 46 p. GRIMSHAWE, F. M. 1940. Place of sorrow: the world's rarest butterfly and Matecumbe Key. Nature Mag. 33: 565-567,611. HENDERSON, W. F. 1945a. Papilio aristodemus ponceana Schaus (Lepidop tera: Papilionidae). Ent. News 56: 29-32. ---. 1945b. Additional notes on Papilio aristodemus ponceana Schaus (Lepidoptera: Papilionidae). Ent. News. 56: 187-188. KIMBALL, C. P. 1965. Lepidoptera of Florida. Arthropods of Florida and neighboring land areas. Vol. 1. Fla. Div. Plant Indust., Gainesville, Florida. 363 p. KLOTS, A. B. 1951. Field guide to the butterflies. Houghton-Mifflin, Boston, Massachusetts. 349 p. LOFTUS, W. F. AND J. A. KUSHLAN. 1982. The status of the Schaus swal lowtail and the Bahama swallowtail butterflies in Biscayne National Park. National Park Service, South Florida Research Center Report M-649, Homestead, Florida. 18 p. MILLER, L. D. 1975. Threatened status for two butterflies? Field Museum Nat. Hist. Bull: 15-18,22. ---, AND F. M. BROWN. 1981. A catalogue/checklist of the butterflies of America north of Mexico. Lep. Soc. Mem. 2. 280 p. MORRISON, G. 1981. Draft recovery plan for the Schaus Swallowtail (Papilio aristodemus ponceanus Schaus), with recommendations con cerning the Bahaman Swallowtail (Papilio andraemon bonhotei Sharpe). Final Rept. to Florida Game and Fresh Water Fish Comm., Tallahassee, .Florida. 23 p. RUTKOWSKI, F. 1971. Observations on Papilio aristodemus ponceanus (Papilionidae). J. Lepid. Soc. 25: 126-136. SCHAUS, W. 1911. A new Papilio from Florida, and one from Mexico (Lepid). Ent. News. 22: 438-439. TYLER, H. A. 1975. The swallowtail butterflies of North America. Nature graph Pub. Inc., Healdsbury, California. 192 p. 288 Florida Entomologist 67 (2) June, 1984

BUPRESTIDAE (COLEOPTERA) OF THE VIRGIN ISLANDS

MICHAEL A. IVIE AND RICHARD S. MILLER Department of Entomology The Ohio State University Columbus, OH 43210, USA

ABSTRACT Nine species of Buprestidae are recorded from the Virgin Islands. Acmaeodera gundlachi Fisher, Trigonogya uniformis (Waterhouse), and Chrysobothris wolcotti Fisher are recorded for the first time from the Virgin Islands. All species are keyed, diagnosed and notes on synonomy, distribu tion, taxonomic history, and biology are provided. A generic key for larvae is provided. Polycesta thomae Chevrolat is returned to synonymy with Polycesta porcata (Fabricius).

RESUMEN Nueve especies de Buprestidae de las Islas Virgenes fueron registradas. Acmaedora gundlachi Fisher, Trigonogya uniformis (Waterhouse), y Chrysobothris wolcotti Fisher fueron registradas por primera vez en las Islas Virgenes. A todas las especies se les hicieron claves, se diagnosticaron y se proveeron notas de sinonimos, distribucion, historia taxonomica y biologia. Una clave generica para las larvas es dada. Polycesta thomae Chevolat es devuelta como sinonimo de Polycesta porcata (Fabricius).

The West Indian Buprestidae were last studied as a whole by Fisher in 1925; they are poorly known due in large part to a scattered literature and a general lack of museum specimens. Concerning the Virgin Islands, mis identifications and lack of collecting have led to confusion about faunal composition. This study of the Virgin Island Buprestidae was undertaken to report recently assembled material, clear up misidentifications, and pro vide a sound taxonomic framework for field biologists, taxonomists and other interested workers. Knowledge of larval habits and other biological aspects of the Virgin Island species is largely incomplete. Larvae.of only 2 species included here are known. Although most buprestids are wood borers in the larval stage, at least one Virgin Island species is suspected of being a leaf miner. More field work is needed to complete distributional and life history information. Discussions of the origin and zoogeography of this group would be premature, as distributions and phylogeny are poorly known. Heatwole and Levins (1972: 115) gave conclusive evidence that some buprestid larvae can survive for long periods in marine flotsam around the Puerto Rican Bank. Thus, larval transport by rafting may play an important role in maintain ing the integrity of the more widespread species. Specimens cited are deposited in various collections, indicated with the following acronyms: AMNH-American Museum of Natural History, New York, NY. BMNH-British Museum (Natural History), London, England. Ivie & Miller: Buprestidae of the Virgin Islands 289

CNCI-Canadian National Collection of Insects, Ottawa, Canada. ICCM-Carnegie Museum, Pittsburgh, PA. MAIC-Michael A. Ivie, private collection, Columbus, OH. NMPC-Narodni Museum, Prague, Czekoslovakia. OSUC-Ohio State University Collection of Insects and Spiders, Colum- bus,OH. PMNH-Peabody Museum of Natural History, New Haven, CT. RSMC-Richard S. Miller, private collection, Columbus, OH. TAMU-Department of Entomology, Texas A & M University, College Station, TX. UCDC-Department of Entomology, University of California, Davis, CA. UZMD-Zoologisk Museum, Copenhagen, Denmark. VIER-Virgin Islands Ecological Research Station, St. John, V.I. VIES-Virgin Islands Cooperative Extension Service, St. Croix, V.1. VIFW-Virgin Islands Division of Fish and Wildlife, St. Croix, V.1.

KEY TO SPECIES (ADULTS) 1. Lateral edges of first visible abdominal tergite produced for ward to contact metepimeron; metacoxal plate expanded medially 2 1'. Hind coxa extending laterally between first abdominal tergite and metepimeron 6 2 (1). Tooth on exterior surface of profemur absent, obtuse, or rounded; Length greater than 8 mm __ 3 2'. Tooth on exterior surface of profemur acute; length less than 7 mm 4 3 (2). Pronotum with 3 to 5 distinct depressions on each side, lateral margins obtusely angulate; eyes separated above by approxi mately the diameter of antennal scape; shining black; elytra costate Chrysobothris tranquebarica (Gmelin) 3'. Pronotum with an indistinct depression on each side behind anterior margin and a distinct one in front of anteriormost point of elytron, lateral margins evenly sinuate; eyes sep arated above by approximately the length of antennal scape; bright metallic green-blue with purplish maculations on elytra; elytra without costae Chrysobothris fisheriana Obenberger 4 (2'). Disk of pronotum with simple punctation; pronotum bright red Chrysobothris thoracica (Fabricius) 4'. Disk of pronotum transversly rugose-punctate; pronotum purple to black 5 5 (4'). Apex of last visible sternite notched on each side [slight in some males (Fig. 2) J; pronotum purplish ______Chrysobothris wolcotti Fisher 5'. Apex of last visible sternite not notched, truncate to sub- truncate; pronotum greenish black ______Chrysobothris thomae Kerremans 6 (1'). Elytra with coarse punctures arranged in distinct rows; length greater than 4 mm 7 6'. Elytra with fine punctures scattered on surface; length less than 4 mm ------____ --- 8 290 Florida Entomologist 67 (2) June, 1984

a q 8

Fig. 1-9. 1) Last visible sternite of Chrysobothris fisheriana Obenberger (sex unknown). 2) Same of C. wolcotti Fisher (female). Same of C. tran quebarica Gmelin 3) male, 4) female. 5) Profemur of C. tranquebarica. 6) Same of C. thoracica (Fabricius). 7) Same of C. fisheriana. 8) Same of C. wolcotti. 9) Brachys thomae Fisher, habitus. Ivie & Miller: Buprestidae of the Virgin Islands 291

7(6). Scutellum not visible; tarsal claws dentate; elytra marked with yellow (Fig. 11) Acmaeodera gundlachi Fisher 7'. Scutellum visible; tarsal claws simple; black with metallic reflections ,------Polycesta porcata (Fabricius) 8(6'). Head strongly narrowed by antennal cavities; elytron with longitudinal carina extending from prominent humerus to near apex (Fig. 9) ; flattened dorsally Brachys thomae Fisher 8'. Head not narrowed by antennal cavities; elytra without carinae (Fig. 10); convex dorsally ______. Trigonogya uniformis Waterhouse

PROVISIONAL KEY TO GENERA (LARVAE) EXCEPT Trigonogya* (Adapted from Burke 1917) 1. Leaf miners, body wedge-shaped, gradually narrowing from first segment to twelfth _ ------.______Brachys 1'. Twig, bark or wood borers; with wide, flattened head; abdomen distinctly narrower than the thorax . .______2 2 (1'). Dorsal plate of prothorax marked with an inverted Y, V, or U . . Chrysobothris 2'. Dorsal plate of prothorax bearing only a single median impression or groove _...... 3 3(2'). Head with a dark, sunken spot laterad of each mandible; ventral plate laterally with longitudinal corrugations ._. Polycesta 3'. Head not as above, ventral plate not corrugated ____ Acmaeodera

Acmaeodera gundlachi Fisher (Fig. 11) Acmaeodera gundlachi Fisher 1925: 45. Wolcott 1936: 214; 1941: 88; 1951: 277. Blackwelder 1944: 307. Ramos 1946: 34. Martorell 1976: 221, 228. This species is variable in size and color pattern as noted by Fisher (1925: 45). The Anegada specimens are within the range of variation of the type series. DIAGNOSIS: This species can be separated from all others of the family known from the Virgin Islands by the absence of a visible scutellum (Fig. 11). From other West Indian Acmaeodera it can be recognized by its slender form, the 4th antennal segment not serrate, and the pronotum widest at base and no wider than the elytra. BIOLOGY: Specimens have been taken on low vegetation and feeding on the large yellow flowers of Urechites lutea and on flowers of Lantana involvucrata L. Wolcott [1936: 214, 1951: 227 (summarized by Martorell 1976: 221, 228)] noted collections from the flowers of red mangrove, Rhizo phora mangle L. at Martin Pena and on sugarcane, Saccharum ofjicinarum L. at Gminica, Puerto Rico. The larval habits of this species are -unknown. DISTRIBUTION: Puerto Rico, Mona, Anegada. TYPE LOCALITY: Gminica, Puerto Rico, holotype in NMNH. MATERIAL EXAMINED: (The number of specimens in each series is in italics.)

*The larvae of Trigonogya are unknown. 292 Florida Entomologist 67 (2) June, 1984

Fig. 10-11. 10) Trigonogya uniformis (Waterhouse), habitus. 11) Acmaeodera gundlachi Fisher, habitus.

Puerto Rico: Holotype and paratypes from Gminica, Anasco and other localities (NMNH). Anegada: 20-23-VIII-1980, M. A. Ivie, on flowers of Urechites lutea (L.) Britt. (Apocynaceae) (15 MAIC, 4 UCDC, 1 VIES). 1-Setting Point, 09-VIII-21-IX-1975, R. K. Butlin (specimen reported in unpublished "Re port of the Cambridge Anegada Expedition", R. P. Dunne and B. E. Brown, 1975) (BMNH). 20-Setting Point, 20-30-VIII-1982, on flowers of Lantana involucrata L. (Verbenaceae), R. S. Miller; 1-ibid. on flowers of Urechites lutea (L.) Britt. (Apocynaceae) (16 RSMC, 5 aSUC).

Brachys thomae Fisher (Fig. 9)

Brachys thomae Fisher 1925: 180. Blackwelder 1944: 333. Ivie & Miller: Buprestidae of the Virgin Islwruls 293

The inclusion of this species in the Virgin Island fauna is questionable. Fisher copied the label data of the unique holotype (La Belle, St. Thomas) and assumed it referred to St. Thomas, Virgin Islands. No La Belle was encountered by M. A. Ivie in 2 years of residence on St. Thomas, and no such locality is listed in the U. S. Board of Geographic Names Gazetteer, 1958, for the Virgin Islands. Gazetteers for other areas do show that La BeLIe or Belle is a common name in the West Indies, occurring on Jamaica, Barbados and Hispaniola. It also occurs in Guyana, Trinidad, and Florida. On both Jamaica and Barbados, a Belle occurs in St. Thomas Parish. Fisher (1924: 4) noted that the buprestid material he borrowed from the Carnegie Museum was from the Isle of Pines and Jamaica. The only other record of Brachys from the West Indies is that of B. aerQsa Melsheimer from Jamaica (also common in the United States) (Blackwelder 1944: 332). Until more material of B. thomae becomes avail able, its true distribution is in doubt. DIAGNOSIS: This species is distinguished by its small size (3 mm) in combination with its deeply grooved prosternal process and the lateral carinae on the pronotum and elytra (Fig. 9). BIOLOGY: Unknown. Other members of this genus are leaf miners in the larval stage. DISTRIBUTION: St. Thomas? TYPE LOCALITY: La Belle, St. Thomas (unique holotype in ICCM). MATERIAL EXAMINED: St. Thomas: La Belle (holotype, ICCM).

Chrysobothris fisheriana Obenberger (Fig. 1,7)

Chrysobothris fisheriana Obenberger 1928: 299 rJisheri pI. 6., Fig. 20 (lapsus calami). Hoscheck 1931: 158. Obenberger 1934: 663. Blackwelder 1944: 317. Obenberger's original description carried the name jisheriana in the text and fisheri on the plate. The label on the type reads "typus, Chrysobothris Fisheri m. Type = Fisheriana m. n. n., det. Dr. Obenberger." Thus, it ap~ pears that Obenberger originally intended that fisheri be the name for this species. Under article 32b of the International Code of Zoological Nomencla ture it is up to the first revisor to choose between multiple spellings in an original description. This species has not been included in a revision and this treatment is certainly not one. However, Chrysobothris fisheri is preoc cupied by a TMry (1927) species from Cuba. To choose fisheri for the Obenberger species would unnecessarily complicate the nomenclature. There fore, we have followed previous authors (including Obenberger 1934) in using fisheria1UL for this species. The unique type, represents a species quite unlike any other we have seen from the West Indies. Fearing to destroy information for a future revisor, we did not dissect the specimen, and its sex is unknown. DIAGNOSIS: The shape of the last visible sternite (Fig. 1) will separate this species from all others in the Virgin Islands, and possibily the West Indies. The shape of the profemur (Fig. 7) and the brilliant green pronotum further distinguish it. BIOLOGY: Unknown. 294 Florida Entomologist 67 (2) June, 1984

DISTRIBUTION: St. Thomas. TYPE LOCALITY: St. Thomas, holotype in NMPC. MATERIAL EXAMINED: St. Thomas: holotype (NMPC).

Chrysobothris thomae Kerremans Chrysobothris thomae Kerremans 1899: 337. Fisher 1925: 118. Blackwelder 1944: 318. DIAGNOSIS: In addition to the characters in the key, this species can be distinguished by the presence of significant areas of true red on the elytra. BIOLOGY: Unknown. DISTRIBUTION: St. Thomas. TYPE LOCALITY: St. Thomas, 2 syntypes in BMNH. MATERIAL EXAMINED: No specimens of this species were seen. The key and diagnosis were based on a series of SEM photos and color transparencies of the syntypes.

Chrysobothris thoracica (Fabricius) (Fig. 6)

Buprestis thoracica Fabricius 1798: 138. Chrysobothrus thoracica Castelnau and Gory 1837: 96. Fisher 1925: 123. Blackwelder 1944: 318. The Virgin Island material reported here was compared with a specimen in the NMNH from Gmlnica, Puerto Rico, determined by W. S. Fisher. Fisher (1925: 125) reported a specimen in the BMNH from St. Thomas. For a complete synonymy of the early literature on this species see Fisher (1925: 125). DIAGNOSIS: This species can be recognized by an acute spine on the profemur (Fig. 6) in combination with a red pronotum. BIOLOGY: The Anegada specimens were beaten from Croton sp. and white mangrove [Laguncularia racemosa (L.) Gaertn]. DISTRIBUTION: Puerto Rico, St. Thomas, St. John, Anegada, St. Croix. TYPE LOCALITIES: Of thoracica, St. Thomas, 3 cotypes in the UZMD, (Zimsen 1964: 157). MATERIAL EXAMINED: St. Thomas: I-Mus. Weterm. (UZMD). St. John: 4-June 1826, Stamm. colI. (UZMD). Anegada: I-Flamingo pond, 22-VIII-1980, M. A. Iviej 4-23-VIII-1980, M. A. Ivie (2 MAIC, 1 RSMC, 1 VIES). St. Croix: 1-Frederiksted, 25-1-1977, J. A. Yntema (NMNH). No specific island: I-Vest Indien, Mus. Hauschild; I-Ind. orient, Mus. Drews (UZMC).

Chrysobothris tranquebarica (Gmelin) (Fig. 3, 4, 5)

Buprestis impressa Fabricius 1787: 182 (not Fabricius 1775). Buprestis tranquebarica Gmelin 1788: 1932 (replacement name). Chrysobothris tranquebarica, Kerremans 1892: 214. Fisher 1918: 173, 1925: 96, 1942: 198. Snyder 1919: 155. Blackwelder 1944: 318. Paulian 1947: Ivie & Miller: Buprestidae of the Virgin Islands 295

147. Benoit 1966: 330. Miskimen and Bond 1970: 86. Martorell 1976: 31, 48, 110, 142, 221. This economically important species in one of the most common and widespread buprestids in the West Indies. This has led to a large body of literature and several complicated synomymies and homonymies in the early literature (for an account see Fisher 1918: 176 and 1925: 96). Since 1925, the name tranquebarica has been uniformly applied to this species. Therefore, the citations included here are limited to those pre-1925 refer ences of greatest importance and those since 1925 pertinent to the Virgin Island fauna. The citation of Tranquebar for the type locality of this West Indian species is probably due to mislabelling. Fabricius received considerable ma terial through members of the Danish Civil Service and Military (Zimsen 1964: 11). Tranquebar, India, was a Danish colony in Fabricius' time, and it is possible that the type reached him through a contact that had served both in India and the Danish West Indies (now the U. S. Virgin Islands). Such material could easily have been mislabelled, or simply labelled with the port from which it was shipped. In any case, the species is restricted to the West Indies and south Florida. This species was recorded from St. Thomas by Fisher (1925: 99) based on a specimen in the BMNH. Miskimen and Bond (1970: 86) reported it from St. Croix. The Virgin Island specimens reported here were compared with 2 specimens from Puerto Rico (NMNH), determined by G. B. Vogt. DIAGNOSIS: This species can be recognized by its large size (10 mm), the rounded or obtuse tooth on the profemur (Fig. 5), the angulate pronotal margins, by usually having 8-10 depressions on the blackish pronotum, and the shape of the last visible sternite (Fig. 3, 4). BIOLOGY: The life history was studied by Snyder (1919), who illustrated all life stages of this species [Benoit (1966: 330) further described the larva]. Snyder reported it as an economic pest of the introduced Australian pine (Casuarina equisetifolia Forst.) and of the red mangrove (Rhizophora mangle L.). Chamberlin (1926: 174) recorded it from buttonwood (Cono carpus erectus L.), a record verified by material from Key Largo, Florida (R. L. Westcott, pers. com.). Martorell (1976: 31, 48, 110, 142, 221) sum marized Puerto Rican host records, adding gree-gree (Bucida buceras L.), Eucalyptus sp. and Inga vera Willd. to the above. Miskimen and Bond (1970: 86) reported it from St. Croix in year-round, sub-economic levels on Australian pine. M. A. Ivie reared one specimen from red mangrove and a series from a limb of a fallen tree on St. Thomas (see material examined). Four tachinid flies [Zelia (Minthozelia) sp., det. D. M. Wood, specimens in the CNCI] were reared from this limb, presumably they were parasitic on C. tranquebarica. DISTRIBUTION: Florida, Bahamas, Cuba, Hispaniola, Puerto Rico, St. Thomas, St. John, St. Croix, Guadeloupe, Grenada [South American local ities for this species (Blackwelder 1944: 318) belong to Halecia impressa (Fabricius, 1775) see Fisher 1918: 175J. TYPE LOCALITY: "Tranquebariae" [India]. MATERIAL EXAMINED: St. Thomas: 2-VI-[1826], Stamm. colI. (UZMD). 1-Est. Nazareth, National Park HQ, 27-VII-1980, emerged 21-VIII-1980 from red mangrove, 296 Florida Entomologist 67 (2) June, 1984

M. A. lvie. (MAIC). 49-Magens Bay Arboretum, collected as larvae 30 VII-1980, emerged 08-20-VIII-1980, M. A. Ivie. (22 MAIC, 10 asuc, 10 UCDC, 5 RSMC, 2 VIES). LARVAE: 9-ibid. (MAIC). PUPAE: 1-ibid. (MAIC). St. John: 1-VIII-1968, A. E. D[ameon] (VIER). Chrysobothris wolcotti Fisher (Fig. 2, 8)

Chrysobothris wolcotti Fisher 1925: 119. Wolcott 1941: 89, 1951: 279. Black welder 1944: 319. Martorell 1976: 153. The St. Thomas specimen reported below was compared with a Puerto Rican specimen from Mayaguez (AMNH). The shape of the last visible sternite and the color of the elytra match quite well, although the St. Thomas specimen is significantly smaller than the Puerto Rican one. The Tortola specimen is more within the normal size range. DIAGNOSIS: This species can be distinquished from the other Virgin Island species by the notched last abdominal sternite (Fig. 2) ; the purplish, transversly rugose pronotum, and the toothed profemur (Fig. 8). BIOLOGY: Wolcott (1941: 89) reported all life stages in crepe myrtle (Lagerstroemia indica L.) in Puerto Rico. The St. Thomas specimen re ported above was found dead in an emergence tunnel in a limb of a dead shrub ca. 2 m tall. DISTRIBUTION: Puerto Rico, St. Thomas, Tortola. TYPE LOCALITY: Puerto Rico, holotype in NMNH. MATERIAL EXAMINED: St. Thomas: 1-Smith Bay, 08-VIII-1980, ex gallery in dead branch, M. A. Ivie (fragmented) (MAIC). Tortola: 1-Green Bank, 900 ft., 21-III-1974, C. L. Remington (PMNH) .

. Polycesta porcata (Fabricius) Buprestis porcata Fabricius 1775: 219. Polycesta porcata, Kerremans" 1906 : 516. Fisher 1925: 19. Wolcott 1941: 88; 1951: 275, Fig. p. 276. Blackwelder 1944: 306. Martorell 1948: 316, 1976: 251. Cobos 1981: 71, Fig. 82, 94. Polycesta thomae Chevrolat 1838: 55. Fisher 1925: 28. Blackwelder 1944: 306. Beatty 1944: 137. Ramos 1946: 34. Wolcott 1951: 276. Miskimen and Bond 1970: 87. Martorell 1976: 48. Cobos 1981: 71. RETURNED TO SYNONOMY. Polycesta nr. depressa, Miskimen and Bond 1970: 87 (not Linnaeus 1771). For references prior to 1925 see Fisher (1925). The nomenclatural status of West Indian Polycesta is unsettled (Cobos 1981: 68) and will remain uncertain until more material is assembled and a complete revision of the genus undertaken. The modern taxonomy of this genus requires males for authoritative determination; males can be dis tinguished from females in this species by a densely setose patch on the 1st visible abdominal sternum. Unfortunately the male genitalia of many of the nominate West Indian taxa remain unstudied. The first Virgin Island record of Polycesta dates to the description of P. thomae Chevrolat from a female specimen in 1838. The St. Thomas records for P. porcata were based on the synonymy of P. thomae with P. Ivie & Miller: Buprestidae of the Virgin Islands 297 porcata by Kerremans (1906: 516). Fisher (1925: 27) reinstated thornae as a valid species, and reported (1925: 28) 2 additional St. Thomas speci mens, both females, under that name, one in the BMNH, the other in the AMNH. Beatty (1944: 137) recorded this species from St. Croix (as thornae), as did Miskimen and Bond (1970: 87, as P. nr. depressa, voucher a female, det. "D.M.W." in NMNH). Cobos (1981: 71) expressed uncer tainty of the validity of P. thornae, but lacked males to settle the issue. The material assembled for this study includes several males. The genitalia of the Anegadan male matches Cobos' illustration (1981: Fig. 94) of that of typical porcata. This specimen also has the scutellar costae well developed, as used by Fisher to distinguish porcata from thornae. However, 2 females collected from the same shrub by R. S. Miller lack this character, though there can be little doubt of their conspecificity. Most other Virgin Island specimens lack scutellar costae. Males exhibit considerable variation in the form of the base of the lateral lobes of the genitalia, though the St. Croix male is somewhat intermediate between the well developed Anegadan specimen and the lesser so St. Thomas, Saba, and St. John specimens. The characters used by Fisher to reinstate thornae: costae, punctation, body width and color are all variable in our series. Barr (1949) notes variation in all these characters in the North American species. He also notes that the genitalia are slightly variable in Polycesta, as is apparent in our ma teriaL DIAGNOSIS: This species can be recognized from all others in the Virgin Islands by the characters in the key. From other Polycesta, it is best sep arated by the form of the median lobe of the male genitalia (Cobos 1981: Fig. 94). Habitus illustrations of this species are given by Wolcott (1951: 275) and Cobos (1981: Fig. 82). BIOLOGY: Wolcott (1941: 88) reported a specimen reared from Domini can mahogany [Swietenia rnahagony (L.) Jacq. (Meliacae)]. Ramos (1946: 34) collected larvae, pupae and adults from dead stems of Australian pine (Casuarina equisetifolia). Miskimen and Bond (1970: 87) report this spe cies on sea grape Coccoloba uvifera (L.) L. Whether the specimen was reared from the plant or the adult was collected on it is unmentioned, the label data is simply "attacking sea grape". A labelled pin in the NMNH states "P. thornae in Elaeodendrurn xylocarpurn, 'nothing nut', Sandy Cay Beach, British Virgin Islands, VIII 1972, Th. FlavelL" The larvae and pupae reported were found in the bole of a dead shrub, just below the first fork. One was reared to adulthood for confirmation of the identification. DISTRIBUTION: Hispaniola, Mona, Puerto Rico, Vieques, St. Thomas, St. John, Little Saba (V. S. Virgin Islands), Grass Cay (V.S.V.I.), Norman Is. (British Virgin Islands), Sandy Is. ? (B.V.I.), Anegada, St. Croix. TYPE LOCALITY: Of porcata, "Oriente Forskahl" (mislabeled ?), 2 syntypes in VZMD Copenhagen (Zimsen 1964: 153). Of thornae, St. Thomas, holotype in BMNH (Fisher 1925: 28). MATERIAL EXAMINED: St. Thornas: 6 (1 t )-R. H. Stamm. colI.; 1-0rsted; 1-1854, H. Krebs (VZMD). 1-VII-1915 [C. Shoemaker] (AMNH). 1-"46-84",. received from BMNH 1923; 1-24-VII-1965, R. Delgado; 1-20-III-1970, R. Cole (NMNH). 1 (t )-Frenchmans Bay Estate, 01-III-1978, M. A. Ivie; 3 (2 t )-Smith Bay, 04-VIII-1980, M. A. Ivie, reared from larvae (1 pair 298 Florida Entomologist 67 (2) June, 1984

MAIC,l (; OSUC). LARVAE: 2-ibid. (MAIC). PUPA: 1-ibid. (MAIC). Little Saba: 1 «; )-24-III-1979, M. A. Ivie; 3-05-VI-1980, C. A. Jennings (MAIC). Grass Cay: 1-19-IV-1971, R. Wharton (TAMU). St. John: 1-V. 1. National Park, Lameshur Bay, 17-VI-1968, R. M. (VIER); 1 «(; )-V. 1. National Park, Cruz Bay, 1970, A. E. D[ameon] (NMNH). Norman Is.: 1-03-VI-1979, M. A. and L. L. Ivie (MAIC) (fragmented). A negada: 3 (1 d' )-Setting Point, 21-24-IV-1983, R. S. Miller (1 pair RSMC, 1 MAIC). St. Croix: 2-Mus. Westerm. (UZMD). 1-Christianstead, rec. 19-XI 1941, H. A. Beatty; 1-Cane Bay, No. 1508, 23-VII-1960, attacking sea grape, Cocolobus sp. [sic]; 1 (d' )-Estate Canaan, elv. 120 m., Ol-VI 1980, W. 1. Knaussenberger (NMNH). 1-Golden Grove, CVI, 14-XI-1979, W. 1. Knaussenberger (VIES). No specific island: 1-Ins. Americ., Mus. Drews; 1-Vestindien, F. Riise; 3-Ins. Americ., Mus. S[ehested and] T[lmder] L[und] (UZMD). Trigonogya uniformis (Waterhouse) (Fig. 10) Mastogenius uniformis Waterhouse 1896: 105. Fisher 1925: 203. Black welder 1944: 34l. Trigonogya uniformis, Fisher 1949: 35l. We were able to compare the single Anegada specimen with the unique holotype from Grenada. The Virgin Island specimen differs in the pronotum being broader, with the lateral margin extending to the anterior margin, the hypomeron more convex, the transverse depression at the base of the elytra less deep, and in the shape of the clypeus and labrum. With only 2 specimens, we were unable to evaluate the possibility of an undescribed species vs. intraspecific variation. The sex of both specimens is unknown. Due to the rarity of specimens, it was considered undesirable to dissect them, to avoid the possibility of destroying data for a future revisor. Until more material becomes available from both localities, as well as the inter vening 800 km, it seems best to treat them as conspecific. DIAGNOSIS: This species can easily be separated from other Virgin Island species by its size (2 mm) and its lack of elytral costae (Fig. 10). BIOLOGY: Unknown. The Al1egada specimen was collected by sweeping low vegetation. The related genus Mastogenius is usually collected by beat ing various trees and shrubs, usually on portions containing dead twigs (R. L. Wescott, pers. com.). DISTRIBUTION: Grenada, Anegada. TYPE LOCALITY: Mount Gay Estate, leeward side, Grenada, West Indies. Unique holotype in the BMNH. MATERIAL EXAMINED: Anegada: 1-23-VIII-1980, M. A. Ivie (MAIC). Grenada: holotype (BMNH).

ACKNOWLEDGEMENTS We would like to thank the following people for the loan of material: L. Herman (AMNH), C. M. F. von Hayek (BMNH), G. Ekis (ICCM), Ivie & Miller: Buprestidae of the Virgin Islands 299

S. BUy (NMPC), J. M. Kingsolver (NMNH), D. G. Furth (PMNH), S. J. Merritt (TAMU), O. Martin (UZMD), W. I. Knaussenberger (VIES), J. Ynetma (VIFW), J. Miller, Virgin Islands National Park, St. Thomas (for material in VIER). C. A. Jennings, Starkville, Mississippi, provided valuable material. Special thanks to Mrs. E. Peacock (BMNH) for provid ing an excellent series of SEM photos and color transparencies of type material. We are grateful to D. M. Wood (CNCI) for the identification of tachinid parasitoids, and to Dr. W. G. D'Arcy, Missouri Botanical Garden, St. Louis, for identifying photos of floral hosts of Acmaeodera gundlachi from Anegada. To Dr. C. A. Triplehorn, Ohio State University, and Dr. R. L. Westcott, Oregon Department of Agriculture, Salem, go our thanks for their helpful reviews of an earlier draft of the manuscript.

REFERENCES CITED BARR, W. F. 1949. A revision of the species of the genus Polycesta oc curring in the United States (Coleoptera, Buprestidae). American Mus. Novit. (1432): 1-42. BEATTY, H. A. 1944. Fauna of St. Croix, Virgin Islands. J. Agric. Univ. Puerto Rico 28: 103-85. BENOIT, P. 1966. Descriptions of Chrysobothris larvae (Coleoptera: Buprestidae) occurring in the United States and Mexico. Canadian Ent. 98: 324-30. BLACKWELDER, R. E. 1944. Checklist of the coleopterous insects of Mexico, Central America, the West Indies, and South America. Part 2. United States Nat. Mus. Bull. 185: 189-341. BURKE, H. E. 1917. Flat-headed borers affecting forest trees in the United States. United States Dept. Agric. Bull. 437: 1-8. CASTELNAU, F. L., AND H. L. GORY. 1837. Historie naturelle et iconographie des insectes coleopteres, publier par monographies, separees, vol. 1 (various pagination). Paris. CHAMBERLIN, W. J. 1926. The Buprestidae of North America exclusive of Mexico. A Catalog. 289 p. + 1 p. index. Corvallis, OR. CHEVROLAT, L. A. 1838. Descriptions de trois buprestides et dun superbe Cyphus nouveaux. Silbermann's Rev. Ent. 5: 54-5. COBOS, A. 1981. Ensayo sobre los generos de la subfamilia Polycestinae (Coleoptera, Bl1nrestidae), Parte 2. Eos, Revista Espanola de En tomolgia [1979] 55-56: 23-94. FABRICIUS, J. C. 1775. Systema Entomologiae. 30 + 832 p. Lipsiae. ---. 1787. Mantissa Insectorum 1: 1-348. Hafinae. ---. 1798. Supplementum Entomologiae Systematicae. 572 p. Hafniae. FISHER, W. S. 1918. Chrysobothris tranquebarica Gmel. versus impressa Fabr. (Coleoptera, Buprestidae). Proc. Ent. Soc. Washington 20: 173-7. ---. 1925. A revision of the West Indian Coleoptera of the family Buprestidae. United States Nat. Mus. Proc. 65: 1-207. ---. 1942. A revision of the North American species of buprestid beetles belonging to the tribe Chrysobothrini. United States Dep. Agric. Misc. Publ. 470: 1-275. ---. 1949. New buprestid beetles from Mexico, Central and South America, and the West Indies. Proc. United States Nat. Mus. 99 (3240): 327-51. GMELIN, J. F. 1788 [1790]. Linne's Systema Naturae... 13 ed. 1: 1517-2224. HEATWOLE, H., AND R. LEVINS. 1972. Biogeography of the Puerto Rican Bank: Flotsam transport of terrestrial animals. Ecology 53: 112-7. 300 Florida Entomologist 67 (2) June, 1984

HOSCHECK, A. B. 1931. Beitrage zur kenntnis der Buprestiden (Col.) 3. Mitt. Zoo!' Mus. Berlin 17: 133-64. KERREMANS, C. 1892. Catalogus synonymicus des buprestides descrits de 1785 a 1890. Mem. Soc. Entomol. Belgique 1: 1-304. ---. 1899. Contribution it l'etude de la faune intertropicale americaine. Buprestides. Ann. Soc. Ent. Belgique 43: 329-67. ---,. 1906. Monographie des buprestides. 1. 534 p. Paris. LINNAEUS, C. 1771. Pages 521-52 In: Mantissa Plantarum II. Holmiae. MARTORELL, L. F. 1948 [1945]. A survey of the forest insects of Puerto Rico. Part 1. J. Agric. Univ. Puerto Rico 29: 69-354. ---. 1976. An Annotated Food Plant Catolog of the Insects of Puerto Rico. Agricultural Experiment Station, University of Puerto Rico. 303 p. MISKIMEN, G. W., AND R. M. BOND. 1970. The insect fauna of St. Croix, U. S. Virgin Islands. New York Acad. Sci. Scientific Survey of Porto Rico and the Virgin Islands 13: 1-117. OBENBERGER, J. 1928. Opuscula buprestologica 1. Arch. Naturgesch. 92: 225-350. ---,. 1934. Buprestidae 3. Coleopterum Catalogus, Pars 132. 12: 569-781. PAULIAN, R. 1947. Famille des Buprestidae. Pages 140-53 In Fleutiaux et al. Coleopteres des Antilles. Faune de I'Empire Fran~ais 7. RAMOS, J. A. 1946. The insects of Mona Island. J. Agric. Univ. Puerto Rico 30: 1-74. SNYDER, T. E. 1919. Injury to casuarina trees in Southern Florida by the mangrove borer. J. Agric. Research 16: 155-63. THERY, A. 1927. Buprestides nouveaux des Antilles. Ann. Soc. Ent. France 96: 31-6. WATERHOUSE, C. O. 1896. Observations on some Buprestidae from the West Indies and other localities. Ann. Mag. Nat. Hist. Series 6. 18: 104-7. WOLCOTT, G. N. 1936. Insectae Borinquenses: a revised annotated check list of the insects of Puerto Rico. J. Agric. Univ. Puerto Rico 20: 1-600. ---. 1941. Supplement to Insectae Boringquenses. J. Agric. Univ. Puerto Rico 25: 33-158. ---,. 1951 [1948]. The insects of Puerto Rico. Coleoptera. J. Agric. Univ. Puerto Rico 32: 225-416. ZIMSEN, E. 1964. The Type Material of 1. C. Fabricius. 656 p. Munksgaard. .. . . •• • • • • • • •• STRUCTURAL DETAILS OF THE SCALES OF 28 SPECIES OF MOSQUITOES KAMEL T. KHALAF AND RAMIZ K. KHALAF Loyola University New Orleans, LA 70118 USA

ABSTRACT Light photographs of scales on the wing and body of 28 species of mosquitoes, which belong to 11 genera, were taken from impressions made on triafol BN. The shape and striae of the scales were clearly discernible and therefore could be utilized in routine taxonomy. The type of scales found on the wing, as well as on the scutum, varied among the different taxa. Those on the abdominal terga were obovate, or intermixed in some

300 Florida Entomologist 67 (2) June, 1984

RESUMEN Con el uso de impresi6nes simples en triafol BN junto con microscopia de luz, micrografias de buena resolucion fueron conseguidas de las escamas del ala y del cuerpo de 28 especies de mosquitos, que pertenecen a 11 generos. Las formas de las escamas y sus estrias eran claramente discernibles con esta tecnica y por consiguiente podran ser utilizados en taxonomia rutinaria. Los tipos de escamas descubiertos en el ala, asi como en el escuto, variaron en differentes taxa. Esas en las tergas abdominales eran obovados, 0 en tremezclados en algunos generos con menor numero de escamas mas grandes. La cantidad de estrias en cualquier tipo de escama fue bastante fijo en cada especie.

A simple technique for preparing transparent impressions was used to study the cuticle surface (Khalaf 1980). The technique gave greater resolu tion than that obtainable by direct stereoscopic microscope examination. This method was used in the present investigations of the surface of the integument of mosquitoes. Theobald (1901, 1905) recognized the taxonomic value of scales of various bodily parts of mosquitoes, and described and named many of their shapes. He concluded that wing scales were important in the classification of genera but such taxa are now only ranked as subgenera or mere species groups. Apparently hampered by limited resolution, Theobald's descriptions were often general and repetitive. Subsequent authors sparingly used scales to differentiate taxa (Mattingly 1957-1965, Foote and Cook 1959). Christo phers (1960) described the scales of Aedes aegypti (Linnaeus). Because specimen preparations for light and scanning electron microscopy are lengthy, the structural details of the scales have not been effectively applied in the taxonomy of Culicidae. Only generalized features of the scales, such as color, width, and length, have been used. Using our method, details of scales in various regions of the body and wing can be easily distinguished, and hence can readi~y be used in systematic analysis.

MATERIALS AND METHODS Details of the method were described previously (Khalaf 1980). Im pressions of scales were made on triafol BN (Mobay Chemical Corporation, Plastic and Coatings Division, Pittsburgh, PA) rectangles which were moistened by a drop of acetone. After removing the specimen, the impression was examined and photographed under high magnification (X600). In the present investigation, scales on the upper surfaces of the wing, scutum, and abdominal terga of females of 28 species were studied (Table 1). Such scales from males of 9 species, as marked in Table 1, were similarly studied. Impressions were taken from 1 or 2 specimens of each species. Only striae of scales that produced an entire and flattened impression were counted. The scales varied in the different regions of the wing. This is also true 302 Florida Entomologist 67 (2) June, 1984

TABLE 1. NUMBER OF STRIAE IN SCALES OF FEMALES OF VARIOUS SPECIES OF MOSQUITOES.

Number of Striae in Scales on

Species R g R 4+5 Scutum Abd. Terga

Anopheles crucians Wiedemann 8-9 7 A. punctipennis (Say) ,;, 7-8 7 A. quad1"imaculaius Say* 6 6 Aedes atlanticus Dyar & Knab 4 7-8 5 12-14 A. sticticus (Meigen) 4 7-8 4 14 A. tcwniorhynchus (Wiedemann) 4 7 4 14 A. aegypti (Linnaeus) 3 7 6 12 A. vexans (Meigen) 4 7-8 5 13 . Psorophora (G.) columbiae (Dyar & Knab) * 6 12 4 12-14 P. (G.) confinnis (Lynch Arribalzaga) 6 10-12 5 12 P. (J.) ferox (Von Humboldt) 4 6 6 9 Culex nigripalpus Theobald 5 7 4 13 C. p. quinquefasciatus Say 3-4 5 [) 12 C. restuans Theobald 4 7 4-5 11 C. salinarius Coquillett* 4 7 5 12 Deinocerites cancer Theobald* 5-6 7 5-6 8-10 Culiseta inornata (Williston)'~ 4-5 6-7 4-5 12 Coquillettidia perturbans (Walker) 10 11-12 8 >20 Mansonia indubitans Dyar & Shannon 18 >18 6-7 11 (&20) M. titillans (Walker) 18 18 5 12 (&18) lVyeomyia mitchellii (Theobald) 10 lV.sp. 4 6-7 7 12 Vranotaenia lowii Theobald 6-7 6-7 10 V.sp. 12 12 11 Toxorhynchites (T.) amboinensis (Doleschall) * 9 9 12 11-1.2 T. (T.) brevipalpis Theobald 9 9 7 12 T. (L.) rutilus (Coquillett)* 9-10 9-10 12 10-11 T. (L.) theobaldi (Dyar & Knab) * 8-9 8-9 12 9-12

*Males possessed similar scales. for other parts of the body. For this reason their location has to be ac curately specified when referring to them. This study describes the lateral wing scales found on veins R g (second) and R 4+5 (third). In naming the type or shape of the scales in the present investigation (Fig. 20), an attempt was made to employ the terms that were available in the literature, especially those of Theobald (1901, 1905) and Harbach and Knight (1980). However, it was necessary to supplement this nomenclature with botanical terms used in describing leaf morphology.

RESULTS The shape and number of striae per scale were quite consistent for a given species (Table 1). The various shapes of the scales that are discussed below are illustrated in Fig. 20. Khalaf & Khalaf: Scales on Mosquitoes 303

+ • Wing: several kinds of scales were found on veins Rs and R 4 5 1-0blanceolate (widest subapically, and with short, pointed apex): Anopheles (Fig. 1). 2-Spatulate (rounded apically): Deinocerites (Fig. 4), Uranotaenia (Fig. 5) (approaching weak asymmetrical), Toxorhynchites (Lyn chiella) (Fig. 9). In Toxorhynchites (Toxorhynchites) the scales were shorter (obovate) (Fig. 8). 3-Asymmetrical (one side of the apex more or less produced): Man soniini (Fig. 6, 7).

Fig. 1-5. Light micrographs of replicas of the wings of female mos quitoes: 1, Anopheles punctipennis. 2, Aedes taeniorhynchus, vein Rs (left) and R,j,+5 (right). 3, Psorophora confinnis, vein Rs left. 4, Deinocerites cancer, vein Rs left. 5, Uranotaenia sp., veing Rs' All X 600. 304 Florida Entomologist 67 (2) June, 1984

Fig. 6-9. Light micrographs of replicas of the wings of female mos quitoes: 6, Coquillettidia pwrturbans. 7, Mansonia titillans, vein R3 above. 8, • Toxorhynchites amboinensis, vein R 3 right. 9, T. theobaldi, vein R 4+5 All X 600.

4-R3 scales were linear (long, slender, and rounded apically) while those of R4+5 were spatulate (Fig. 2). Aedes, Psorophora (Jan thinosoma), Culex, Culiseta, Wyeomyia. In Psorophora (Grabhamia) + the scales were shorter (R3 long spatulate, R 4 5 obovate or short spatulate) (Fig. 3). Scutum: Anopheles exhibited the most primitive condition-having no scales; however, the scutum of females and males of A. punctipennis and of females of A. crucians possessed some long, striated setae (primitive, pili form scales) (Fig. 10). These setae seem to represent a link in the evolution Khalaf & Khalaf: Scales on Mosquitoes 305

Fig. 10-13. Light micrographs of replicas of the scutum of female mos quitoes: 10, Anopheles punctipennis. 11, Culiseta inornata. 12, Aedes aegypti. 13, Psorophora ferox. All X 600. of true scales. Most of the genera studied exhibited falcate scales (~res centric, sickle-shaped) (Fig. 11, 12). Based solely on comparative morphol ogy, it seems that such falcate scales had their origin in striated setae, such as those illustrated for Anopheles (compare Fig. 10 and 11). On the other hand, other scale types were present in some primitive genera such as Toxorhynchites and Wyeomyia. Other types of scales in this region included: 1-Fusiform (spindle-shaped) : Psorophora (Janthinosoma) (Fig. 13). 2-0bovate-Spatulate: Wyeomyia (Fig. 14). Some larger scales were intermixed with the dominant ones. The obovate scales are wider than the short spatulate type. 3-0blanceolate: Toxorhynchites (Fig. 15). Scales of varying sizes were intermixed. Some of the scales of T. brevipalpis were fusiform to slightly falcate (Fig. 16). 306 Florida Entomologist 67 (2) June, 1984

Fig 14-16. Light micrographs of replicas of the scutum of mosquitoes: 14, Wyeomyia mitchelii female. 15, Toxorhynchites theobaldi male. 16, T. brevipalpis female. All X 600. Khalaf & Khalaf: Scales on Mosquitoes 307

19 "

Fig. 17-19. Light micrographs of replicas of the abdominal terga of mos quitoes: 17, Culex salinarius female. 18, Toxorhynchites rutilus male. 19, Coquillettidia perturbans female. All X 600. 308 Florida Entomologist 67 (2) June, 1984

Abdominal Terga: In this region, as in the scutum, there were no scales in Anopheles. In the rest of the genera studied, the scales were obovate (Fig. 17). The most common numbers of striae were 11-14. Fewer striae were found in Psorophora (Janthinosoma) and Deinocerites. More numerous striae were found in Coquillettidia. The dominant scales were intermixed with fewer larger scales in Deinocerites and Toxorhynchites (Fig. 18), and with fewer wider scales in the tribe Mansoniini (Fig. 19). In future comparative studies of mosquitoes, the scales of other regions, e.g. those of the posterior margin of the wing or other veins, etc. may serve

A B c

o E F G Fig. 20. Types of mosquito scales: A, oblanceolate. B, spatulate. C, asym metrical. D, linear. E, obovate. F, fusiform. G, falcate. Khalaf & Khalaf: Scales on Mosquitoes 309 taxonomy effectively. Keys of adult mosquitoes still involve clusters of spe cies, e.g. Anopheles crucians Wied., A. georgianus King, and A. bradleyi King, that cannot be differentiated from each other. Species of the group can be reared from determined larvae; a comparative survey of scales of all the areas and veins of these reared adults can then be done which might prove rewarding in differentiating members of the complex.

ACKNOWLEDGMENT The mosquitoes were received through the courtesy of Dr. Joseph C. Cooney and Mr. Eugene Pickard of Tennessee Valley Authority, Fisheries and Aquatic Ecology Branch; Mr. George Carmichael, Dr. Michael K. Carroll, and Mr. Stephen R. Sackett of New Orleans Mosquito Control; Mr. Leonard W. Trager, Jr. of Jefferson Parish Mosquito Control; and Florida Medical Entomology Laboratory at Vero Beach. This investigation received support from the Academic Grant Fund of Loyola University.

REFERENCES CITED CHRISTOPHERS, S. R. 1960. Aedes aegypti (L.) the yellow fever mosquito., University Press, Cambridge. 738 p. FOOTE, R. H., AND D. R. COOK. 1959. Mosquitoes of medical importance. Agriculture Handbook No. 152. ARS, USDA, Washington. 158 p. HARBACH, R. E., AND K. L. KNIGHT. 1980. Taxonomists' glossary of mos quito anatomy. Plexus Publishing, Inc. 415 p. KHALAF, K. T. 1980. Micromorphology of beetle elytra, using simple rep licas. Florida Ent. 63 (3): 307-40. MATTINGLY, P. F. 1957-1965. The culicine mosquitoes of the Indomalayan area. British Museum (Natural History), London. THEOBALD, F. V. 1901. The classification of mosquitoes. J. Trop. Med. 4: 229-35. ---. 1905. Genera Insectorum, Diptera, Fam. Culicidae. Fas. 26. Bel gium. 50 p.

310 Florida Entomologist 67 (2) June, 1984

SCIENTIFIC NOTES

STRUCTURAL ABNORMALITIES IN TWO WORKER YELLOWJACKETS (HYMENOPTERA: VESPIDAE)

KENNETH G. ROSSl Department of Entomology Cornell University Ithaca, NY 14853 USA

Developmental abnormalities among social wasps of the subfamily Vespinae have been infrequently reported (Edwards 1980). These most commonly involve irregularities in segmentation of the gaster (tergum splitting or fusion) (Chapman 1917, Duncan 1939, Janet 1903, Maneval 1932), but irregular segmentation of an antenna has also been recorded (Crevecoeur and De Walsche 1929). While collecting workers from 35 Vespula colonies during the summer and fall of 1982 in central New York state, I di£covered 2 workers with external morphological abnormalities. The first wasp was taken from a colony of V. fiavopilosa Jacobson on 17 August. The head of this worker was deformed on one side and the gaster was irregularly segmented (Fig. 1a & b). The left side of the head was compressed dorso-ventrally in the region of the left compound eye which was expanded laterally. Gastral terga I and II were apparently normal, while tergum III was split at the midline. Terga IV and V were incomplete and partially fused; tergum VI appeared normal. The ventral segmentation of the gaster and the sting apparatus appeared normal so that this wasp was unlikely to be gynandromorphic. This worker was collected from within the colony and the ~ings showed no sign of wear. Ninety other workers collected from this colony and 320 workers collected from 3 other conspecific colonies appeared normal in al! respects. A worker of V. vulgaris (L.) with an abnormal third gastral tergum was collected on 17 September (Fig. 1c). The sclerite was split medially and the right portion was partially fused to tergum IV. The wasp was normal in all other respects. This wasp was captured in a foraging trap placed over the colony entrance (Schmidt et aI. 1978), thus she was capable of flying and may have been an active forager. None of 39 other foragers examined from this colony, or 657 workers from 7 other conspecific colonies exhibited any obvious morphological abnormalities. Given that the 35 Vespula colonies collected during this survey constitute a representative sample, the discovery of markedly aberrant individuals in 2 of the colonies suggests that structural abnormalities in these wasps may be more common than previously suspected. Extensive surveys of workers within colonies have not often been undertaken, and deformed workers may be confined to the nest interior because of diminished flight or orientation capabilities. The V. vulgaris worker from this report and a foraging worker of V. germanica (F.) with abnormal segmentation of the gaster (Chapman 1917) are conspicuous exceptions to this expectation.

'Present address: Department of Entomology, University of Georgia, Athens, GA 30602. Scientific Notes 311

Fig. 1. a) Deformed head of worker of V. jlavopilosa. The left antenna has been removed (X 39, bar = 1.0 mm). b) Abnormal segmentation of gaster of V. jlavopilosa worker from Fig. 1a (X 28, bar = 1.0 mm). c) Ab normal segmentation of gaster of V. vulgaris worker (X 28, bar = 1.0 mm). The cause of deformity in the 2 wasps I describe is not clear. A greater number of similarly affected individuals would be expected in each colony if the abnormalities were due to the expression of recessive deleterious alleles or to teratogenic colony-wide environmental stress. Fusion of some segments would seem unlikely if the wasps were injured in the immature stages dur ing brood provisioning or maintenance by workers. Thus, isolated irregular ities in the developmental programs of the affected wasps due to unknown causes must presently suffice as an explanation. I thank Dr. Roger A. Morse and Cornell University Department of Entomology for funding support during the period of this study.

REFERENCES CITED CHAPMAN, -. 1917. An aberrant wasp. Proc. Ent. Soc. London 50: 49-50. CREVECOEUR, A., AND J. DE WALSCHE. 1929. Description de trois Hy menopteres gynandromorphes ou monstrueux. Bull. AnnIs. Soc. R. Ent. Belgium 69: 37-9. 312 Florida Entomologist 67 (2) June, 1984

DUNCAN, C. D. 1939. A contribution to the biology of North American vespine wasps. Stanford Univ. PubIs. BioI. Sci. 8: 1-272. EDWARDS, R. 1980. Social Wasps, Their Biology and Control. Rentokil, East Grinstead, Britain. 398 p. JANET, C. 1903. Observations sur les Guepes. Naud, Paris. 85 p. MANEVAL, H. 1932. Notes recueillies sur les Hymenopteres. Ann. Soc. Ent. France 101: 85-112. SCHMIDT, J. 0., M. S. BLUM, AND R. W. MATTHEWS. 1978. Mass capture of live yellowjackets: application and results of a capture-survey tech nique. J. Georgia Ent. Soc. 13: 163-8. • • ••• • • • • • • •• A NEW TRAP FOR CAPTURING DIAPREPES ABBREVIATUS (COLEOPTERA: CURCULIONIDAE) W. J. SCHROEDER U.S. Department of Agriculture-ARS Orlando, FL 32803 USA AND 1. F. JONES Tri-State University Angola, IN 46703 USA Diaprepes abbreviatus L. is an important weevil pest of sugarcane, citrus, and many other commercial crops in the Caribbean Basin. It was first reported in Florida by Woodruff (1964). During 1968, 2,630 ha in the infested area were quarantined. Infestations are now distributed through out 29,300 ha of commercial citrus in central Florida and 2,428 ha of out door nurseries and environs in south Florida. Because there has been no efficient method of detecting D. abbreviatus adult populations, recent re search efforts have been directed toward developing a trap and trapping methods. Previous trapping studies on this insect in citrus were conducted with light traps (Beavers et al. 1979) and sticky traps (Beavers, et al. 1982). Light traps were ineffective. Sticky traps are difficult to work with and cannot easily be placed in citrus trees where adult weevil populations are most abundant. Described here is a durable, lightweight trap without sticky surfaces that is more useful for detection and population survey of D. abbreviatus. Initially, 2 trap designs (A and B) were studied. The traps were similar in that each was a screen (0.32 cm mesh) trap consisting of a 15-cm-diam x 16-cm-high cylindrical holding chamber (cricket cage) atop a 22-cm-diam x 20-cm-high screen COIW with openings of 3.6 cm2 and 380 cm2• The two de signs were different in that they had modified surfaces under the cone (Fig. 1). Trap A had wire cloth vanes 38 cm long x 12.5 cm wide attached under the cone. A 2.7-cm-OD PVC pipe seated in the ground was used to support the trap so that the bottom of the vanes touched foliage in the upper canopy of the tree.

312 Florida Entomologist 67 (2) June, 1984

A B

Fig. 1. Diaprepes abbreviatus trap designs: (A) 4 vanes, and B) plastic pot.

Trap B had a tapered (15 ~ 10 cm diam) 16-cm-high plastic pot at tached beneath the cone. The pot was attached with wire to the cone so that a 2-cm space between cone and pot allowed weevils to enter the cage from outside the pot and through 2-cm2 holes in the bottom of the pot. Traps were supported and placed as in trap A. This design was conceived as a way to incorporate trap color into trap design, but data on color effect are not re ported here. A chloroform extract of 20 g of frass from a mixed laboratory colony (1:1,

(20 g frass in 225 ml CHClg ) were soaked into a cotton wick suspended in the trap. Twenty each of traps A and B were then placed (1 trap/tree) in selected citrus trees (2 m high). All traps were re-treated daily with ex tract. Low population levels of adults weevils in June necessitated a test period of 14 days to determine that trap A captured significantly more weevils (x = 2.2/trap) than trap B (x = 0.5/trap) (Student's t test, P = 0.05). A total of 43 (26

REFERENCES CITED BEAVERS, J. B., T. P. MCGOVERN, AND V. E. ADLER. 1982. Diaprepes ab breviatus: laboratory and field behavioral and attraetancy studies. Environ. Ent. 11: 436-9. ---, J. M. STANLEY, H. R. AGEE, AND S. A. LOVESTRAND. 1979. Diaprepes abbreviatus response to light traps in field and cage tests. Florida Ent. 62: 136-9. WOODRUFF, R. E. 1964. A Perto Rican weevil new to the United States (Coleoptera: Curculionidae). Florida Dept. Agric. Div. Plant Ind. Ent. Circ. 30: 1-20. • • • • • • • • • • CORRELATIONS OF BURROW CHARACTERISTICS AND BODY SIZE IN BURROWING WOLF SPIDERS (ARANEAE: LYCOSIDAE) GARY L. MILLER AND PATRICIA R. MILLER Department of Entomology Mississippi Entomological Museum, Drawer EM Mississippi State University, Mississippi State, MS 39762 USA The wolf spiders Lycosa carolinensis Walckenaer, L. georgicola Walckenaer and members of the genus Geolycosa (Araneae: Lycosidae) are similar in their use of constructed burrows for retreats (Wallace 1942). Some of the characteristics of the burrows constructed by G. domifex (Han cock) and G. godeffroyi (L. Koch) are positively correlated with the size of the spider inhabiting the burrow (McQueen 1978, Humphreys 1976, re spectively). However, similar relationships are not known for L. caro linensis or L. georgicola. Although these 2 wolf spiders construct burrows, they retain many of the typical lycosid behavioral characteristics, notably the tendency to wander in search of prey (Kuenzler 1958, pers. observ.). Also only certain instars of L. carolinensis burrow (Shook 1979). Because of these life history differences we predicted that burrow characteristics, such as opening diameter, would not be as strongly correlated with body size for the 2 Lycosa species as in the Geolycosa. We attempted to verify this from field collections in Florida and Mississippi and from laboratory observations. We collected specimens of Lycosa carolinensis (Levy county), Geolycosa escambiensis Wallace (Holmes County), G. micanopy Wallace (Alachua, Levy, Marion and Putnam counties), G. ornatipes (Bryant) (Levy and Putnam counties), G. patellonigra Wallace (Clay, Levy, Marion and Put nam counties) and G. turricola (Treat) (Alachua and Santa Rosa counties) in Florida between 5 and 10 December 1983. Specimens of L. georgicola were taken in Oktibbeha county near Starkville, Mississippi during the summer of 1983. All specimens were taken from burrows, and for each burrow, except those of L. georgicola and the laboratory populations, we measured the depth, the largest diameter of the burrow opening, the height of the turret when present (measured from ground level) and the distance across the burrow entrance between the 2 extreme points of the turret (when the turret was present) which we termed the "total burrow diameter". During Scientific Notes 315 our collecting in Florida, not all excavated burrows were measured. Spiders from these burrows were returned to the lab and allowed to construct new burrows in plastic cups containing sand. Inside burrow diameter and carapace width were measured for these and the data are presented sep arately (identified as "lab"). We measured only inside burrow diameter for L. georgicola. The parametric correlation between the natural log of the carapace width, which we consider a measure of relative age (Hagstrum 1971), and burrow characteristics and the corresponding t-value were de termined. The correlation between carapace width and inside burrow diameter for L. georgicola was not significant (Table 1). There was a moderate positive correlation between burrow diameter and carapace width for L. carolinensis, but the correlations were substantially stronger for the Geolycosa (Table 1). The average difference between carapace width (untransformed) and inside burrow diameter was 10.6, 14.8 and 4.7 mm respectively for L. caro ilnensis, L. georgicola and the combined Geolycosa. Only G. escambiensis and G. ornatipes had significant correlation between carapace width and the height of the turret. For all of the Geolycosa, there were significant correla tions between both burrow depth and total burrow diameter and carapace width (Table 1). The lab populations of G. micanopy and G. patellonigra had significant correlation between carapace width and inside burrow diam eter. The higher correlations observed in the Geolycosa are a reflection of the greater importance of the burrow in the life history of the group. The im portance of the burrow to the Geolycosa is well known. Humphreys (1975) described how G. godeffroyi used burrows to reduce mortality resulting from climatic conditions and McQueen (1978) reported that G. domifex used burrows as protection against predation. Shook (1978) indicated that mem bers of his desert population of L. carolinensis also derived some protective and thermoregulatory advantage from the burrow. However, his data indi cate that the members of the species may not need the protection of the burrow at all instars. Our observations on several populations of L. georgicola near Starkville, Mississippi indicate that the species is not de pendent on the burrow during all instars and does not use the burrow for prey capture. Juvenile L. georgicola do not build burrows and adults forage substantial distances from their burrows. It is not known whether the bur row provided a thermoregulatory advantage. The small average difference between the carapace width and burrow diameter in the Geolycosa is another indication of the restrictive lifestyle of the group. Geolycosa do not generally occupy burrows constructed by other spiders, whereas small L. carolinensis and L. georgicola are often found hidding in burrows which were presumably constructed by larger members of the species (pers. observ.). We appreciate the comments of G. Baker, G. Uetz, and M. Greenstone on an early draft of this note. We thank W. H. Cross for his continued support of our research. ....CI:i 0')

TABLE 1. CORRELATIONS OF CARAPACE WIDTH AND BURROW CHARACTERISTICS FOR SEVERAL SPECIES OF BURROWING WOLF SPIDERS .,...~ FROM FLORIDA AND MISSISSIPPI. NUMBEltS IN PARENTHESES ARE RESPECTIVELY THE T-VALUES AND PROBABILITY RANGE c ~ ASSOCIATED WITH EACH CORRELATION COEFFICIENT. R: ~ Correlation with LN (carapace width) ~ Species N Depth Inside diameter Total diameter Turret height ~ ~ c Lycosa carolinensis 15 0.26 (0.97,0.4-0.2) 0.57 (2.50,0.02-0.05) 0.32 (1.22, 0.4-0.2) 0.25 (0.93,0.4-0.2) 0 ~ L. georgicola 15 - 0.41 (1.62, 0.2-0.1) ~. G. escambiensis 10 0.92 (6.64, <0.001) 0.74 (3.11,0.01-0.02) 0.77 (3.41,0.01-0.001) 0.76 (3.31, 0.02-0.01) .,..,. G. micanopy 44 0.71 (6.53, <0.001) 0.77 (7.82, <0.001) 0.70(6.35, <0.001) 0.37 (2.58,0.9-0.5) 0') G. micanopy (Lab) 27 - 0.80 (6.67, <0.001) .--::J G.ornatipes 19 0.79 (5.31, <0.001) 0.73 (4.40, <0.001) 0.82 (5.91, <0.001) 0.71(4.16, <0.001) ~ G. patellonigra 23 0.75(5.19, <0.001) 0.84 (7.35, <0.001) 0.52 (2.79, 0.02-0.01) 0.26 (1.23, 0.4-0.2) G. patellonigra (Lab) 10 - 0.73 (3.02,0.02-0.01) -- G. turnicola 13 0.68 (3.32, 0.01-0.001) 0.79 (4.27,0.01-0.001) 0.91 (7.28, <0.001) 0.20 (0.68,0.9-0.5)

~ r:: I:l ~(1).... 1:0 00 H:>- Scientific Notes 317

REFERENCES CITED HAGSTRUM, D. W. 1971. Carapace width as a tool for evaluating the rate of development in spiders in the laboratory and field. Ann. Ent. Soc. America 64: 757-60. HUMPHREYS, W. F. 1975. The influence of burrowing and thermoregula tory behavior on the water relations of Geolycosa godeffroyi (Araneae: Lycosidae) an Australian wolf spider. Oecologia 21: 291 311. ---. 1976. Population dynamics of an Australian wolf spider. Oecologia 21: 291-311. KUENLER, E. J. 1958. Niche relations of three species of lycosid spiders. Ecology 39 (3): 494-500. MCQUEEN, D. J. 1978. Field studies of growth, reproduction, and mortality in the burrowing wolf spider Geolycosa domifex (Hancock). Canadian J. Zool. 56: 2037-49. SHOOK, R. S. 1979. Ecology of the wolf spider, Lycosa carolinensis Walckenaer (Araneae, Lycosidae) in a desert community. J. Arachnol. 6 (1): 53-64. WALLACE, H. K. 1942. A revision of the burrowing spiders of the genus Geolycosa (Araneae, Lycosidae). American MidI. Nat. 27 (1): 1-61. 318 Florida Entomologist 67 (2) June, 1984 Errata ERRATA-THE GENUS fizophora IN FLORIDA. J. A. Slater and R. M. Baranowski. Florida Ent. 66 (4): 416-40. On page 436 of this manuscript there is an error that is likely to cause nomenclatorial confusion on the status of Ozophora 1"eperta and O. inornata. The CORRECT listing of these 2 species is as follows: Ozophora reperta Blatchley 1954. Ozophora inornata Barber, American Mus. Novit. N. 1682, p. 5-6. NEW SYNONYMY. The authors recognize inornata as the new junior synonym of Blatchleyts species, reperta. The Editor regrets not catching the mistake in time. Book Announcement 319

ADVANCES AND CHALLENGES IN INSECT REARING. 1984. E. G. King and N. C. Leppla, eds. VSDA-ARS. 306 + xvi pages. 80 illustrations, tables, keys. Order from: Superintendent of Documents, V.S. Gov't. Printing Office, Washington, D.C. 20402, Stock No. 001-000-04393-5. $8.00, domestic price. "Advances and Challenges in Insect Rearing" fills the need for an up-to