Diptera: Tephritidae - PubAg - USDA

BEHAVIOR

Mating Competitiveness of Cast-191, a Promising Genetic Sexing Strain of the Mediterranean Fruit Fly (Diptera: Tephritidae) J. L. CLADERA,1, 2 M. T. VERA,2 G. CALCAGNO,3 J. C. VILARDI,3 D. O. MCINNIS,4 FIELD WORKING GROUP5

AND

Ann. Entomol. Soc. Am. 95(1): 133Ð138 (2002)

ABSTRACT Genetic sexing strains (GSS) are being developed for use in sterile insect technique programs against the Mediterranean fruit ßy, Ceratitis capitata (Wiedemann). Among other important factors, the utility of these new strains depends on the ability of laboratory-reared males to compete efÞciently with wild males in mating with wild females in the Þeld. The current study evaluated the mating ability of a new GSS, Cast 191, in competition with wild males. A well-known GSS, Seib 6 Ð96, was also included in the test. The effect of irradiation on the mating competitiveness of the Cast 191 strain was assessed. The results showed that nonirradiated Cast 191 males outcompeted wild males, whereas irradiated males performed as well as wild counterparts. Although Seib 6 Ð96 males performed less efÞciently than Cast 191, they did not differ from their own performance reported 2 yr ago. Mating duration of Cast males, either irradiated or not, was not signiÞcantly different from the mating duration of wild males. Seib 6 Ð96 had shorter mating duration times than wild and Cast males. These results demonstrated, besides an effect of irradiation on male mating success, a good mating performance of the strain Cast 191. KEY WORDS Ceratitis capitata, Mediterranean fruit ßy, sterile insect technique, genetic sexing strains, mating competitiveness

THE MEDITERRANEAN FRUIT ßy, Ceratitis capitata (Wiedemann), causes millions of dollars in direct damage annually to fruits and vegetables worldwide (Christenson and Foote 1960). Even more damaging economically is the loss of potential markets due to quarantine restrictions for both domestic and foreign exports (Malavasi et al. 1994). As a result, various technologies have been developed over the past several decades to suppress or even eradicate this pest from infested areas. In recent years, a strong emphasis has been placed on implementing environmentally safe, biological control rather than on the traditional use of chemical insecticides (Knipling 1979). One of these biological methods involves the release of sterile males to mate with wild females and thereby reduce the wild population (Cunningham et al. 1980, Klassen et al. 1994). Moreover, it has been shown experimentally in the Þeld that the release of all-male sterile Mediterranean fruit ßies is several times more effec1

E-mail: [email protected]. Instituto de Gene´ tica, INTA Castelar, CC 25, Castelar (1712), Argentina. 3 Departamento de Cs. Biolo´ gicas, Fac. Cs. Exactas y Nat., University of Buenos Aires, Argentina. 4 USDA/ARS, Honolulu, HI 96822. 5 Field Working Group: E. Stolar, D. Segura, N. Petit Marty, F. Krsticevic (Instituto Gene´ tica, INTA Castelar), P. Gomez Cendra, M. Rodriguero, K. Barborini, T. Heer (Departmento Cs. Biolo´ gicas, Fac. Cs. Exactas y Nat., University of Buenos Aires); A. Allinghi (Comisio´ n Nacional de Energõá Ato´ mica); G. Bonpland (ISCAMen Mendoza); and L. Hansen and G. Segade (INTA San Pedro). 2

tive than releasing both sexes (McInnis et al. 1994, Rendon et al. 2000), although Shelley and Whittier (1996) did not Þnd such an effect. Strains that can genetically separate males from females have been created and are commonly referred to as genetic sexing strains (GSS) (Whitten 1969). In the Mediterranean fruit ßy, separation of the sexes has been attained at the pupal stage, with males arising from normal brown color pupae and females from mutant white color pupae (Robinson and van Heemert 1982). The two components of this GSS are as follows: (1) a single recessive gene trait that provides a basis for developing a sexing strain and (2) a genetic translocation of the normal pupal color trait to the male Y chromosome. More advanced GSS use earlier stages of development, i.e., the egg or larval stages (Saul 1984, Franz et al. 1996). These sexing strains are more efÞcient in reducing operational costs of separating males from females. One such new GSS is the Cast 191 strain from Argentina, developed at INTA, Castelar, which possesses a mutation of the sw (slow) gene located on chromosome 2. The recessive mutation sw affects the rate of development as well as the eye color and iridescence (Manso and Lifschitz 1992, Cladera 1995). Female ßies of this strain are homozygous for the mutation and males are heterozygous carriers of a translocation linking chromosomes 2 and Y. It has been shown that the different developmental times of sw bearing individuals could be used to separate males from females in the pupal

0013-8746/02/0133Ð0138$02.00/0 䉷 2002 Entomological Society of America

134

ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA

(Cladera 1995) or egg stages (Pizarro et al. 1997). Currently, rearing of the GSS Cast 191 is being expanded to begin mass-production. Action programs that use GSS for the sterile insect technique (SIT) are effective when high proportions of wild females mate with sterile males and thus fail to reproduce. However, recent studies have indicated that wild females C. capitata tend to mate more often with wild than with sterile, laboratory-reared males (Shelly et al. 1994, McInnis et al. 1996, Liimatainen et al. 1997, Cayol et al. 1999, but see Wong et al. 1983). These results point out the critical need to monitor closely the mating quality of laboratory-reared insects in the Þeld. In an effort to deÞne and measure the mating performance of mass-reared versus wild ßies, standard quality control tests have been developed (IAEA 1998). The current method of choice to evaluate new strains is to conduct mating trials in outdoor Þeld cages between one or more laboratory strains and wild ßies under seminatural conditions. In the current study, males of the Cast 191 strain were compared with wild males to measure the mating success of this new genetic sexing strain in Þeld cages in Argentina. Another GSS was included in the test, Seib 6 Ð96, for which Þeld cage mating competitiveness had already been reported (Cayol et al. 1999). Materials and Methods Experimental Site. The Þeld cage experiments took place at the National Center of Agricultural Investigation, INTA, in Castelar (⬇58⬚ 40⬘ W, 34⬚ 40⬘ S), near Buenos Aires City, Argentina, between 13 and 24 March 2000. Male and female ßies of a wild strain, plus males of two distinct genetic sexing laboratory strains, were used in the study. Wild Flies. Wild ßies were obtained from several collections of infested peaches, Prunus persica L., at the INTA Experimental station in San Pedro, located in the fruit producing region around San Pedro, Buenos Aires province, Argentina. Fruit were collected from the trees and off the ground and then held under laboratory conditions (23Ð27⬚C, 50 Ð70% RH, and a photoperiod of 12:12 [L:D] h) until obtaining adults for experimental use in the Þeld. Adult ßies were provided water and food in the form of sugar (sucrose): protein (hydrolyzed yeast) in a 3:1 ratio (wt: wt). Laboratory Strains. 1. Castelar 191: Nonirradiated (Cast) and irradiated (Cast*) adult male ßies were obtained from the genetic sexing Castelar strain, Cast 191Ð39303. The original genetic background of this strain is the strain MI94, which was under mass-production at the Provincial Insectary in Mendoza, Argentina, in 1994, and which was itself obtained from wild material collected from Mendoza province 2Ð3 yr prior. Female ßies of this strain are homozygous for a mutation of the sw (slow) gene (Cladera 1995), and males are heterozygous carriers of the translocation: T(Y;2)191, with breakpoint located between the regions 17B and 17C of chromosome 2 (M. A. Delprat, unpublished data). This strain shows a reduced level

Vol. 95, no. 1

of recombination and good rearing qualities (E. C. Stolar, unpublished data). As such, it is a potentially successful GSS that could be transferred to mass rearing. Cast 191Ð39303 is reared at the INTA Genetics Institute laboratories on a carrot-based larval diet (Terań 1977). Adult ßies were provided water and food, as stated above, in the form of sugar: protein (hydrolyzed yeast) in a 3:1 ratio and held in large cage, which contain 5,000 individuals/cage. For the current study, pupae were divided into two lotsÑ one nonirradiated (Cast), and one irradiated (Cast*) at a median dose of 118.5 Gy (maximum dose ⫽ 120, minimum dose ⫽ 117 Gy), under hypoxia in a Gammacell 220 60 Co irradiator according to protocol established by the National Commission of Atomic Energy (A. Allinghi, personal communication) ⬇2 d before adult emergence. 2. Seibersdorf: Males of this strain were from the Seib 6 Ð96 line whose females are homozygous for the wp mutation (white pupae, Ro¨ ssler 1979), whereas the males are heterozygous and carriers of the genetic translocation, T(Y;5)2Ð22 (Franz et al. 1994). This strain is currently under mass-production at the BioKm 8 Insectary (ISCAMEN, Mendoza, Argentina) and is used in ongoing SIT programs of Mediterranean fruit ßy eradication in the regions of Mendoza and Patagonia, Argentina. The genetic background of the Seib 6 Ð96 line is EgII, originally from Egypt and reared by the International Atomic Energy Agency (IAEA) in Seibersdorf, Austria. Pupae from this strain were obtained from the same ISCAMEN facility after irradiation under hypoxia in a IMCO 20 60Co irradiator at a median dose of 125 Gy (minimum dose ⫽ 100 Gy, maximum dose ⫽ 150 Gy) (Oscar Delongo and Gustavo Taret, personal communications) 2 d before emergence. Pupae were then shipped by air, on the same day as irradiation, to Buenos Aires (⬇1.5 h ßight) and carried to the INTA laboratory in Castelar. Sexing of Adults. All of the pupae obtained from the wild strain and the two genetic sexing strains were maintained in the laboratory under standard conditions for adult Mediterranean fruit ßies, i.e., 23Ð27⬚C, 50 Ð70% RH, and a photoperiod of 12:12 (L:D) h. To separate the wild adults efÞciently, ßies were aspirated mechanically from holding cages, and anesthetized by exposure to cold for an average of 10 min at 0⬚C (in no case did ßies stay longer than 20 min) within the Þrst 24 h of emergence. Although the pupae of the GSS strains already were received in a sexed condition, the emerging males were also anesthetized and checked for low level female contamination. In all cases, sexed ßies were placed in lots of 60 individuals per plastic container (volume 1 liter) with water and food (3:1 sugar: protein) and held until sexual maturity (11Ð14 d for wild ßies, 5Ð 6 d for laboratory ßies). For irradiated ßies, only individuals emerging within 48 Ð72 h after irradiation were used. Fly Marking. Females were marked with individual labels. On the day before each experiment, ßies were anesthetized by exposure to cold temperature (0⬚C) for ⬇10 min. Then, with careful handling using soft forceps, a letter (font size #3, Arial, 0.5 ⫻ 1.0 mm)

January 2002

CLADERA ET AL.: MATING COMPETITIVENESS OF A NEW C. capitata STRAIN

printed on colored paper was added with a dot of acrylic paint onto the dorsal mesonotum of each ßy, according to McInnis et al. (2001). Upon drying, the drop of paint sealed the printed label to the body of the ßy. After labeling, 25 females were placed in each 1-liter container with food and water and held under laboratory conditions until next day. Males were not individually marked, but strains were identiÞed with a small dot of paint on the mesonotum. For cages where males from two different strains were released, one of the strains (laboratory or wild, alternately) was painted. In the Þeld cages where more than two strains were released together, all of the males were painted with a different color to identify male origin. Colors were randomly assigned and interchanged every day to control for any effects of painting on mating performance. After marking, 50 males were placed in each container and held in the same manner as virgin females. Test Protocol. The Þeld experiments were conducted in an open Þeld where outdoor nylon screened cages (⬇2.5 m high by 2.5 m diameter) were erected over rooted mandarin (Citrus nobilis L.) trees ⬇1.5 m tall and ⬇2 yr old. Four experiments were conducted. In the Þrst three, 50 males of one of the laboratory strains (Seib*, Cast, or Cast*, respectively) were released with 50 wild males and 50 wild females, resulting in a 2:1 male:female sex ratio, and a 1:1 ratio of laboratory versus wild males. In experiment 4, 25 males of each of the four tested lines were released along with 50 wild females. In this case, the male: female sex ratio was maintained at 2:1, but the laboratory:wild male ratio changed to 3:1. Each test morning at ⬇0715 hours (⬇20 min after sunrise), males were released inside each Þeld cage. Fifteen minutes later the females were released into the cages. Dead, moribund, or nonßying ßies were replaced with healthy ßies in each cage. Mating pairs were collected over a 10-h period until the end of the experiment on that day. The following data were recorded for each mating pair: the mating start time, the location in the cage where the pair was collected (i.e., the screen or the tree, and the height [high, middle, or low]), the strain of the male, and the color and letter of the label on the female. The pairs were collected in plastic vials and placed in the shade inside the cage until the mating couple separated ad libitum. Once separated, the ßies were released again into the Table 1. Week

Day

1

1 2 3 4 5 6

2

135

cage, and the time of separation was noted. The temperature, relative humidity, light intensity, and the number of ßies found on the screen of the cage were recorded at 30-min intervals. Dead ßies found in the cage were recorded, though none of these was replaced once observations began. At the end of the 10 h of observation, all of the live ßies were aspirated from the cage, and the label identiÞcations of the live females were recorded. Data Analysis. The general conditions of the experiment were considered adequate on the basis of the percentage of the mated ßies in each cage - the accepted standard was 20% for a minimum (quality control manual of the IAEA [1998]). The measure of mating success of the laboratory males (L) compared with wild males (W) was expressed in terms of the relative sterility index (McInnis et al. 1996), RSI ⫽ LW/(LW⫹WW). (The alternative male relative performance index (Cayol et al. 1999), MRPI ⫽ (SWWW)/(SW⫹WW) is mathematically related to RSI by the expression MRPI ⫽ 2RSI Ð1 in tests were females from only one origin are released [wild in the present case]). For the three experiments where only one laboratory strain was released (Seib*, Cast, and Cast*), the RSI index was calculated for each replicate per test, and the possible differences among the various tests were evaluated with a nonparametric analysis of variance (ANOVA) or Kruskall-Wallis test (Daniel 1990). The experiments were also taken in pairs and contrasted using a Mann-Whitney (Daniel 1990) test of median values (Daniel 1990), with BonferroniÕs correction applied to each result. For the fourth experiment in which males of all strains were tested, a goodness-of-Þt chi-square test was calculated for each replicate, assuming that the expected proportion of copulations was the same for each type of male (1:1:1:1). Values for total, global, and heterogeneity chi-square were calculated among replicas for each test. With respect to the analysis of the mating start time and mating duration, a nonparametric ANOVA was performed, taking into account the origin of the male. To evaluate any possible relationship between the location of the mating pair in the cage and male origin, a Homogeneity chi-square was performed. All statistical analyses were performed with STATISTICA for Windows (StatSoft 2000).

Number of replications per test and weather conditions during the entire experiment (Buenos Aires, Argentina, March 2000) # replicates per test

Temp, ⬚C

RH, %

Seib*

Cast

Cast*

All

Mean

Max

Min

Mean

Max

Min

Wind Direction, m/s3

Rain, mm

Relative sunshinea

3 3 3 1 3 3

3 3 3 1 1 0

3 3 3 2 1 3

0 0 0 3 4 3

24.0 25.4 19.5 20.9 17.2 18.8

29.4 32.5 25.2 27.0 25.4 24.4

18.6 18.2 13.8 14.8 9.0 13.2

62 49 48 66 58 54

88 86 93 86 100 93

37 34 Ñ 42 31 28

NE 6 NW 6 SW 6 N6 N6 W 10

0 0 0 0.5 0 0

72.1 71.7 88.6 36.4 81.6 88.8

Data from the meteorological station of the Institute of Climate and Water, INTA, Castelar, Argentina, based on 24-h recordings. m/s, meters/second. a Percentage of the day with direct sunlight.

136


Vol. 95, no. 1

Table 2. Mean ⴞ SD for the number of copulations involving wild or laboratory males, number of females mating at least once, and percentages of females mating and remating for each test identified by the origin of the laboratory strain released

Wild male Lab male Mated females % mating % remating

Seib* (N ⫽ 16)

Cast (N ⫽ 11)

Cast* (N ⫽ 15)

Seib*/Cast/Cast* (N ⫽ 10)

12.63 ⫾ 4.95 8.88 ⫾ 3.26 20.19 ⫾ 4.61 40.38 ⫾ 9.22 5.87 ⫾ 6.08

8.73 ⫾ 5.29 18.55 ⫾ 3.86 24.00 ⫾ 4.84 48.00 ⫾ 9.67 12.43 ⫾ 9.50

12.3 ⫾ 6.44 14.4 ⫾ 5.05 23.9 ⫾ 5.18 47.7 ⫾ 10.36 12.31 ⫾ 8.85

6.2 ⫾ 3.74 2.4 ⫾ 1.58/9.8 ⫾ 4.08/7.1 ⫾ 3.41 23.2 ⫾ 4.66 46.4 ⫾ 9.32 8.64 ⫾ 4.09

Results Experimental Conditions. Environmental conditions are presented in Table 1. The light intensity inside the cages ranged from a low daily median of 24,000 Lux to a high daily median of 74,000 Lux (SD 9,900), with an overall daily median of 55,000 Lux. The observed proportion of mated females to released females was ⬎20% in all the replicate Þeld cages, and only 2.3% of the observed copulations took place off the tree, indicating that the environmental conditions and plant substrate were suitable to simulate natural conditions. As a result, none of the replications was eliminated from the analysis. The high mating rate also indicates that insects were healthy and not signiÞcantly affected by manipulation (e.g., anesthesia, labels). Indeed, a consistent proportion (6 Ð12%) of rereleased females were able to remate on the same day (Table 2). Mating Competitiveness of the Sexing Strains. The number of copulations obtained for each male type per replica and test indicated that the proportions of mated wild males versus mated laboratory males varied among experiments 1Ð3, in which males from single laboratory strains were used (Kruskall-Wallis test: H(2) ⫽ 14.99, P ⫽ 0.0006). Cast males mated more successfully than wild males, while the opposite was observed for Seib* males. The average value of the relative sterility index (RSI) for Cast* males (0.55 ⫾ 0.18) was intermediate to those for Seib* males (0.42 ⫾ 0.13) and Cast males (0.70 ⫾ 0.14). These differences were signiÞcant for Mann-Whitney test of medians: Z ⫽ 2.09, P ⫽ 0.0362 for Cast* versus Seib* and Z ⫽ 2.02, P ⫽ 0.0429 for Cast* versus Cast. The difference between Seib* and Cast males is highly signiÞcant, even after applying the Bonferroni correction (Z ⫽ ⫺3.75, P ⫽ 0.0006), whereas the other two become nonsigniÞcant. In the experiment in which males from all the laboratory strains were used, the numbers of mating pairs were not in the expected proportion of 1:1:1:1 (goodness-of-Þt chi-square to the proportion 1:1:1:1 of expected mating pairs were as follows: ␹2(G,3) ⫽ 44.06, P ⫽ 0.000; ␹2(T, 30) ⫽ 92.52, P ⫽ 0.000; ␹2 (H, 27) ⫽ 48.46, P ⫽ 0.007). High numbers of matings involved Cast males in almost all of the replications (Table 2). This result is consistent with the results from the other tests, although some heterogeneity was observed among replicates as indicated by the signiÞcantly high heterogeneity chi-square above. In particular, there was an alternation for the highest mating type between Cast and Wild males across replications (not shown).

Mating Start Time and Mating Duration. Results (Table 3) analyzed with a nonparametric ANOVA for each variable indicated that the mating start time for each male type did not differ among the strains (Kruskal-Wallis test: H(3) ⫽ 1.644, P ⫽ 0.649; ␹2(3) ⫽ 1.542, P ⫽ 0.672), whereas mating duration did differ among male types (H(3) ⫽ 44.658, P ⬍ 0.001; ␹2(3) ⫽ 45.793, P ⬍ 0.001). Pairwise comparisons by MannWhitney tests of medians indicated that Cast males did not differ from Wild males, and Cast* males differed signiÞcantly both from Cast and wild males (Z ⫽ 2.27, P ⫽ 0.023 for Cast* versus Cast; Z ⫽ 2.33, P ⫽ 0.020 for Cast* versus wild). These latter differences, however, did not remain signiÞcant after applying the BonferroniÕs correction. Seib* males differed in their mating duration from the rest of the male types (P ⬍ 0.001) even after applying BonferroniÕs correction. Location of Mating Pairs in the Trees. Most (76%) of the matings took place at intermediate heights, whereas only a few were collected in the lower (7%) or upper (17%) third of the tree. The pairs recorded in the intermediate tree level occurred mostly on the underside of leaves (88%), but this was not homogeneous across the various male types, (␹2(6) ⫽ 20.59, P ⫽ 0.002). In particular, Cast and Cast* males mated on the upper side of leaves (13.8 and 10.3%, respectively) more frequently than Wild males (6.5%). Seib* males mated more often on the stems of the tree (7.9%) than the other male types (Wild, 2.0%; Cast, 1.83%; Cast*, 2.82%). Discussion GSS have signiÞcantly increased the efÞciency of Mediterranean fruit ßy SIT programs (McInnis et al. 1994, Rendon et al. 2000). For this reason, and economic reasons associated with rearing costs, several laboratories around the world are developing new

Table 3. Start times and durations of mating pairs collected in field cages for wild and laboratory strains of C. capitata Male origin

Mating start

Mating duration

Wild Seib* Cast Cast*

11:55 ⫾ 2:41 (545)a 11:37 ⫾ 2:10 (166)a 11:53 ⫾ 2:39 (302)a 12:00 ⫾ 2:29 (287)a

3:03 ⫾ 1:02 (536)a 2:28 ⫾ 0:56 (165)b 3:02 ⫾ 1:07 (290)a 2:50 ⫾ 1:05 (283)a

Means followed by a different letter differed signiÞcantly according to Mann-Whitney median test (P ⬍ 0.001) after BonferroniÕs correction. Mean values (hours:min) ⫾ SD and number of observations (N) for each male type.

January 2002

CLADERA ET AL.: MATING COMPETITIVENESS OF A NEW C. capitata STRAIN

Mediterranean fruit ßy GSS strains for use in SIT programs. These efforts are being focused on strains in which the sexes are separated early and are adaptable to mass rearing. The utility of these new strains for the SIT depends upon three factors: (1) the efÞciency in separating the sexes, (2) the ability for mass rearing, and (3) the ability of the laboratory strain males to compete efÞciently with wild males in mating with wild females. The Þrst two of these factors are in an advanced stage (E. C. Stolar et al., unpublished data; M. A. Delprat et al., unpublished data) for the Cast 191 GSS. The current study focused on male mating competitiveness. At the same time, it assessed the effect of irradiation on the mating competitiveness of this strain. The results showed that Cast males mated with higher frequency than wild males (RSI ⫽ 0.7). The difference in mating competitiveness between laboratory and wild insects can potentially have two components, a genetic one based on processes of inheritance and selection pressures of mass-rearing, and a second one based on environmental aspects of the mass-rearing process, including the negative effect of sterilizing irradiation. Finding laboratory ßies more competitive than wild ßies is unusual. High performance of laboratory ßies versus wild ßies was not revealed for several other laboratory strains in Þeld cage tests performed in Guatemala or Hawaii (McInnis et al. 1996, Lance et al. 2000), nor in the present work for Seib 6 Ð96. It is known that selection in massrearing fruit ßies usually leads to faster developing ßies and earlier sexual maturation (Leppla et al. 1983, Briceno and Eberhard 1998), but there is no clear relation between this fact and mating success. It cannot be ruled out that some aspect of laboratory rearing produced mating behavioral traits advantageous for Cast males. Regardless of the mechanism, the performance shown here is quite exceptional. A similar advantage has been achieved by a laboratory-reared hybrid strain in Hawaii, but only after several generations of directional selection of laboratory males for mating success with wild females in Þeld cages (D.O.M., unpublished data). The experimental design of our study, i.e., including both Cast and Cast* test populations, allowed evaluation of the effect of irradiation on the mating success of laboratory males. The performance of irradiated Cast191 males was lower than that of nonirradiated Cast191; yet it did not differ signiÞcantly from that of wild males. The negative impact of irradiation on mating success revealed here for Cast191 was also shown by Calcagno (2001) for Cast228, a related laboratory strain with the same genetic background, but carrying a different translocation. To get a beneÞt of the Cast strainÕs male mating advantage, this detrimental impact of irradiation on its performance should be minimized. The coefÞcient RSI (0.42) for the Seib 6 Ð96 line is consistent with prior values obtained for this strain in 1997 by Cayol and coworkers in Argentina (Cayol et al. 1999). The RSI value was actually smaller, ⬇0.3, in that experiment. This suggests that the Seib 6 Ð96 strain

137

maintained its mating quality in the intervening 3 yr of mass-production at the ISCAMEN rearing facility in Mendoza. The tendencies observed in the individual comparisons were maintained in the Þnal experiment, with all male types present. Again, nonirradiated Cast males consistently had the greatest success, with irradiated Cast and wild males in second place, followed by irradiated Seib 6 Ð96 males in last place. Though some of the observed differences among replications could be partly explained by environmental factors, the consistency of the ranks of male types, i.e., Cast and wild males above the median values, and Seib* males below the medians, strongly suggests inherent differences among the strains in mating performance. In addition, the data again indicate a measured debilitating effect caused by irradiation, as noticed previously (Calcagno 2001). While the mating start time for each male type did not differ among the strains, the mating duration did differ among male types. Seib* males differed in their mating duration from the rest of the male types (average, 35 min shorter than Wild). The mating duration of Cast males did not differ from that of Wild males, and Cast* males were only marginally different (average, 12Ð13 min shorter) from Cast and Wild males. Although the observed differences between the Cast-191 and Seib 6 Ð96 strains, both in terms of mating competitiveness and mating duration must be taken with caution, because the Cast-191 strain was not raised under mass-rearing conditions, the results obtained for the Cast-191 strain are promising, suggesting that it fulÞlls one of the prerequisites of a successful GSS. Acknowledgments The authors thank the following institutions: BioKm 8, ISCAMen mass-rearing facility, in Mendoza for the kind donation of irradiated pupae of the Seib 6 Ð96 strain, the national organization CNEA for irradiating the Cast-191 pupae, the institute INTA San Pedro for allowing the collections of infested peaches, and the Institute of Climate and Water, INTA Castelar, for providing data from the meteorological station. Three anonymous reviewers greatly improved an early version of the manuscript. The work was supported by USDA/ARS (Hawaii) funds for miscellaneous supplies for D.O.M. and the following grants: FONCyT PID 615 and IAEA RCP8308 to J.L.C. and IAEA RC 103/R1, CONICET PIP 0722/98, and UBA-PID TW09 to J.C.V.

References Cited Briceno, R. D., and W. G. Eberhard. 1998. Medßy courtship duration: a sexually selected norm changed by crowding. Ethol. Ecol. Evol. 10: 369 Ð382. Calcagno, G. E. 2001. Comportamiento Reproductivo en la Mosca del Mediterrańeo, Ceratitis capitata (Diptera: Tephritidae) y su aplicacio´ n en programas de control gene´ tico. Ph.D. dissertation, University of Buenos Aires, Argentina. Cayol, J. P., J. C. Vilardi, E. Rial, and M. T. Vera. 1999. New indices and method to measure the sexual compatibility and mating performance of medßy (Diptera: Tephriti-

138


dae) laboratory reared strains under Þeld cage conditions. J. Econ. Entomol. 92: 140 Ð145. Christenson, L. D., and R. H. Foote. 1960. Biology of fruit ßies. Annu. Rev. Entomol. 5: 171Ð192. Cladera, J. L. 1995. Self-sexing strain of Ceratitis capitata (Diptera: Tephritidae) based on a gene that affects the rate of development. Ann. Entomol. Soc Am. 88: 353Ð356. Cunningham, R. T., W. Routhier, E. J. Harris, G. Cunningham, N. Tanaka, L. Johnston, W. Edwards, and R. Rosander. 1980. A case study: eradication of medßy by sterile-male release. Citrograph 65(3): 63Ð 69. Daniel, W. W. 1990. Applied nonparametric statistic, 2nd ed. PWS-Kent, Boston. Franz, G., P. Kerremans, P. Rendon, and J. Hendrichs. 1996. Development and application of genetic sexing systems for the Mediterranean fruit ßy based on a temperature sensitive lethal, pp. 185Ð192. In B. A. McPheron and G. J. Steck (eds.), Fruit ßy pests: a world assessment of their biology and management. St. Lucie Press, Delray Beach, FL. Franz, G., E. Gencheva, and P. Kerremans. 1994. Improved stability of genetic sex-separation strains for the Mediterranean fruit ßy, Ceratitis capitata. Genome 37: 72Ð 82. Klassen, W., D. A. Lindquist, and E. J. Buyckx. 1994. Overview of the Joint FAO/IAEA DivisionÕs involvement in Fruit Fly Sterile Insect Technique Programs, pp. 3Ð26. In C. O. Calkins, W. Klassen, and P. Liedo (eds.), Fruit ßies and the sterile insect technique. CRC, Boca Raton, FL. Knipling, E. F. 1979. Basic principles of insect suppression and management. U.S. Dep. Agric. Handb. 512, Washington, DC. [IAEA] International Atomic Energy Agency. 1998. A manual of quality control for fruit ßies, version 4.0. IAEA, Vienna, Austria. Lance, D., D. O. McInnis, P. Rendon, and C. Jackson. 2000. Courtship among sterile and wild Ceratitis capitata (Diptera: Tephritidae) in Þeld cages in Hawaii and Guatemala. Ann. Entomol. Soc. Am. 93: 1179 Ð1185. Leppla, N. C., M. D. Huettel, D. L. Chambers, T. R. Ashley, D. H. Miyashita, T.T.Y. Wong, and E. J. Harris. 1983. Strategies for colonization and maintenance of the Mediterranean fruit ßy. Entomol. Exp. Appl. 33: 89 Ð96. Liimatainen, J., A. Hoikkala, and T. Shelly. 1997. Courtship behavior of Ceratitis capitata (Diptera: Tephritidae): comparison of wild and mass-reared males. Ann. Entomol. Soc. Am. 90: 836 Ð 843. Malavasi, A., G. G. Rohwer, and D. S. Campbell. 1994. Fruit ßy free areas: strategies to develop them, pp. 165Ð180. In C. O. Calkins, W. Klassen, and P. Liedo (eds.), Fruit ßies and the sterile insect technique. CRC, Boca Raton, FL. Manso, F. C., and E. Lifschitz. 1992. Nueva metodologõá gene´ tica para el mejoramiento de la eÞciencia de la te´ cnica del macho este´ ril en el control de la mosca del Mediterrańeo Ceratitis capitata. Cienc. Invest. 44(4): 225Ð228. McInnis, D. O., S. Tam, C. Grace, and D. Miyashita. 1994. Population suppression and sterility rates induced by

Vol. 95, no. 1

variable sex-ratio sterile insect releases of Ceratitis capitata (Diptera: Tephritidae) in Hawaii. Ann. Entomol. Soc. Am. 87: 231Ð240. McInnis, D. O., D. R. Lance, and C. G. Jackson. 1996. Behavioral resistance to the sterile insect technique by the Mediterranean fruit ßy (Diptera: Tephritidae) in Hawaii. Ann. Entomol. Soc. Am. 89: 739 Ð744. McInnis, D. O., P. Rendon, and J. Komatsu. 2002. Mating and remating of medßies (Diptera: Tephritidae) in Þeld cages: ßy histories revealed by an individual marking technique. Fla. Entomol. (in press). Pizarro, J. M., F. C. Manso, and J. L. Cladera. 1997. New allele at a locus affecting developmental time in Mediterranean fruit ßy (Diptera: Tephritidae) and its potential use in genetic sexing at the egg stage. Ann. Entomol. Soc. Am. 90: 220 Ð222. Rendon, P., D. O. McInnis, D. L. Lance, and J. Stewart. 2000. Comparison of medßy male-only and bisex releases in large-scale Þeld trials, pp. 517Ð526. In Keng-Hong, Tan [ed.] Area-wide control of fruit ßies and other insect pests. Penerbit Universiti Sains Malaysia, Penang, Malaysia. Robinson, A. S., and C. van Heemert. 1982. Ceratitis capitata is a suitable case for genetic sexing. Genetica 58: 229 Ð237. Ro¨ ssler, Y. 1979. The genetics of the Mediterranean fruit ßy: a white pupae mutant. Ann. Entomol. Soc. Am. 72: 583Ð 585. Saul, S. S. 1984. Genetic sexing in the Mediterranean fruit ßy, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae): conditional lethal translocations that preferentially eliminate females. Ann. Entomol. Soc. Am. 77: 280 Ð283. Shelly, T. E., and T. S. Whittier. 1996. Mating competitiveness of sterile male Mediterranean fruit ßies (Diptera: Tephritidae) in male-only releases. Ann. Entomol. Soc. Am. 89: 754 Ð758. Shelly, T. E., T. S. Whittier, and K. Y. Kaneshiro. 1994. Sterile insect release and the natural mating system of the Mediterranean fruit ßy, Ceratitis capitata (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 87: 470 Ð 481. StatSoft. 2000. STATISTICA for Windows 5.5 (computer program manual). StatSoft, Tulsa, OK. Tera´ n, H. 1977. Comportamiento alimentario y su correlacio´ n a la reproduccio´ n en hembras de Ceratitis capitata Wied. Rev. Agron. NOA 14: 17Ð34. Whitten, M. J. 1969. Automated sexing of pupae and its usefulness in control by sterile insects. J. Econ Entomol. 62: 272Ð273. Wong, T.T.Y., J. I. Nishimoto, and H. M. Couey. 1983. Mediterranean fruit ßy (Diptera: Tephritidae): further studies on selective mating response of wild and of unirradiated and irradiated, laboratory-reared ßies in Þeld cages. Ann. Entomol. Soc. Am. 76: 51Ð55. Received for publication 29 March 2001; accepted 30 August 2001.