Serge Quilici (corresponding author), CIRAD, UMR « Peuplements Végétaux et Bio-agresseurs en Milieu Tropical », CIRAD/Université de La Réunion, Saint Pierre, La Réunion F-97410, France. E-mail: email@example.com
The Natal fruit fly, Ceratitis rosa (Diptera: Tephritidae), is a major pest of fruit crops in Reunion island. The sterile insect technique (SIT), which has frequently been used against some fruit fly species, could also contribute to the control of this pest. However, mass-rearing procedures and irradiation can strongly reduce the mating competitiveness of males. In this study, we investigated the influence of two essential oils [ginger root oil (GRO) and orange oil (OO)] on the males of C. rosa, in terms of attractiveness and effect on mating success. The influence of adult food regime on mating success of the males was also studied.
During tests in choice situation between the two oils, mature males were more attracted to GRO compared with OO. This was not the case when males had not reached their sexual maturity, which indicates that this attractiveness is age-dependent. The addition of proteins to the adult diet increased the mating competitiveness of the males, compared with males fed with sugar only. Males fed with a ‘full’ diet (sugar and hydrolysed yeast) accounted for 85% of all matings compared with 15% for those fed with a sugar-only diet. Results were similar for wild males and for males from a laboratory-reared colony. Exposure of the males to both types of oils significantly increased the mating competitiveness of sugar-fed males, while only the exposure to GRO was able to increase that of males previously fed with a ‘full’ diet. This study provides promising results for the improvement of male competitiveness in the perspective of SIT programmes against C. rosa in the future.
The Natal fruit fly, Ceratitis rosa (Karsch) (Diptera: Tephritidae), is a key pest of fruit crops in Reunion Island because of its large geographical distribution throughout the island and its wide polyphagy (Etienne 1982; Quilici 1989). The sterile insect technique (SIT) is an environmentally friendly approach to suppress, or sometimes eradicate, insect pests and is widely used in control programmes against tephritid pests, particularly the Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Hendrichs et al. 1995, 2002). This technique could be integrated in the control of C. rosa and is currently being developed in South Africa (Barnes et al. 2002; Barnes and Venter 2006). Progress is being made in the development of a mass-rearing technique for C. rosa (Barnes 2010). The creation of a genetic-sexing strain will allow the release of sterile males only (Barnes et al. 2002). Overcoming these limiting factors will eventually open the way for the use of SIT in the various countries of Eastern and Southern Africa, or in some islands of the south-western Indian Ocean where it is present (De Meyer 2001a,b; Baliraine et al. 2004). However, for any SIT project, it will be important to consider the fruit flies complexes in individual countries, as the suppression of a single species such as C. rosa could allow other competitors to occupy its emptied ecological niche (Duyck et al. 2004).
However, the success of the SIT depends, to a large extent, on the ability of released sterile males to attract and successfully copulate with wild females. This behavioural capability is especially important for species such as C. capitata (and presumably C. rosa), in which females display a high degree of mate selection, apparently based on male courtship performance (Whittier et al. 1992, 1994). Unfortunately, the mass-rearing procedures and irradiation inherent to the SIT usually reduce the mating competitiveness of male flies, and released sterile males typically have a lower mating success than wild males (Röessler 1975; Shelly et al. 1994; Lance et al. 2000). Thus, an important challenge for SIT programmes is the development of simple and inexpensive means to enhance sterile males’ mating performance. In recent years, several studies focused on two possible modifications to the pre-release environment of sterile males to accomplish this goal: the exposure of males to certain semiochemicals and the manipulation of the pre-release adult diet.
The males of many tephritid species are strongly attracted to certain plant-derived compounds, also called ‘male lures’ or ‘parapheromones’ (Jang and Light 1996). During the last 15 years, a series of studies have demonstrated that some of these attractants may have an important effect on the pheromone production and mating behaviour of males. For example, males of the Oriental fruit fly, Bactrocera dorsalis (Hendel), which feed on pure methyl eugenol (Shelly and Dewire 1994; Tan and Nishida 1996) or natural sources (flowers) of this compound (Shelly 2000, 2002), show a pronounced mating advantage over control (unfed) males. Males that have ingested methyl eugenol display an increased signalling activity and produce a more attractive pheromone (Shelly and Dewire 1994), presumably following the incorporation of certain metabolites of methyl eugenol into their pheromone (Nishida et al. 1997). There is growing evidence that plant substances influence male reproductive behaviour in other Bactrocera spp. as well (Nishida et al. 1993; Hee and Tan 1998; Tan and Nishida 2000).
Similarly, laboratory tests on C. capitata revealed that feeding on wounded fruits of the orange tree (Citrus sinensis L.) confers a mating advantage to males (Papadopoulos et al. 2001). Furthermore, this positive effect is quite long-lasting: males exposed to the fruits for 24 h had a mating advantage over control males for at least 10 days following exposure. Likewise, in field cage tests, Shelly and Villalobos (2004) found that male Mediterranean fruit flies feeding on the bark and fruits of guava trees (Psidium guajava L.) obtained significantly more matings than control males. However, tests with both orange and guava showed that males gained a mating advantage only when they were allowed to be in contact with the fruits (and for guava, the bark) and that exposure to fruit (or guava bark) aroma alone had no effect on male mating success. In addition, exposure to essential oils contained in the peel of citrus fruits, the root of ginger (Zingiber officinale Roscoe) and the seed of angelica (Angelica archangelica L.) has been shown to confer a pronounced mating advantage to male Mediterranean fruit flies (Shelly 2001; Shelly et al. 2004a; Papadopoulos et al. 2006). For example, Shelly (2001) found that C. capitata males exposed to 20 μl of angelica seed oil (ASO hereafter) or ginger root oil (GRO) for 6 h and tested 2 days later obtained 67–81% of the total matings in field cage tests. Further studies have shown that sterilized, mass-reared males exposed to GRO (Shelly and McInnis 2001; Shelly et al. 2002b, 2003, 2004b, 2007a) or OO (Shelly et al. 2004a, 2006, 2008) acquired a mating advantage over wild non-exposed males. The positive effect is generally quite long and lasted at least 3, 5 and 7 days following exposure to OO (Shelly et al. 2004a), GRO (Shelly 2001) and ASO (Shelly 1999), respectively. Furthermore, in contrast to the natural substrates, the oils conferred a mating advantage even when males were not allowed to be in direct contact with them.
Modification of the pre-release, adult diet, a standard sugar–agar gel, may represent another simple approach to increase the effectiveness of sterile males in SIT programmes. Previous studies using wild C. capitata showed that ingestion of proteins by wild adult males enhances various aspects of their sexual behaviour (Yuval et al. 1998, 2007; Kaspi et al. 2000). However, contrasting results have been found when laboratory-reared flies are used (Joachim-Bravo et al. 2009). For example, Papadopoulos et al. (1998) showed that wild adult males fed on a protein-rich diet (sugar : yeast hydrolysate; 1 : 4) mature earlier and signal more frequently than males fed on a protein-poor diet (sugar only), a result not confirmed in experiments using laboratory-reared males. Similarly, studies conducted in Hawaii failed to demonstrate an effect of diet on the mating performance of mass-reared males (Shelly and Kennelly 2002; Shelly and McInnis 2003). However, other studies on C. capitata did show that laboratory-reared males fed on a protein-rich diet are more likely to copulate than males deprived of protein (Blay and Yuval 1997; Kaspi and Yuval 2000; Shelly et al. 2002a). According to Yuval et al. (2007), while protein feeding universally increases the mating success of wild males, its effect on laboratory- or mass-reared males varies with strain, experimental settings and environmental conditions. Joachim-Bravo et al. (2009) also point out that the larval diet of the males should be taken into consideration. Thus, the potential role of dietary protein in improving the effectiveness of Mediterranean fruit fly SIT still requires additional studies.
After confirming OO and GRO attractiveness for laboratory-reared males of C. rosa, our objective in this study was to evaluate the influence of the two essential oils on the mating success of mature males of the Natal fruit fly. We also examined whether the method of exposure to the oils (‘olfaction only’ or ‘olfaction plus contact’) influences the mating success. We then assessed the influence of adult food regime on mating success. Finally, we investigated the combined influence on male mating success of exposure to the semiochemicals and adult food regime.
Material and Methods
Most of the adult C. rosa flies used for the experiments were produced in the CIRAD Entomology Laboratory in Saint-Pierre. The rearing was maintained in a regulated rearing room with a temperature of 25 ± 2°C, a relative humidity of 70 ± 10% and a light intensity of 2000 ± 300 Lux, 12:12 L:D. The rearing method followed was the one developed for C. rosa by Etienne (1973).
Prior to fly emergence, the boxes containing the pupae were transferred to cubical emergence cages (37 × 37 × 37 cm) with three sides covered with a mesh and the others with plexiglass. Every day, the pupae boxes were transferred to new cages to obtain cohorts of adults emerging on the same day. After emergence, adults were separated by sex and transferred to a new maturation cage, depending on their sex and the food regime chosen. Except for the experiment on adult food regime, adults were fed with sugar, Enzymatic Yeast Hydrolysate (ICN, Biochemicals, Aurora, OH, USA) and water supplied ad libitum. Adults used in the experiments were 16–17 days old (sexually mature) except for those used in the experiments on the influence of age on oil attractiveness.
In one of the experiments on food regime, we used wild flies collected from infested fruits of loquat [Eriobotrya japonica (Thunb.) Lindley] in the south of the island at an altitude of 600 m above sea level. In the laboratory, the fruits were placed in plastic trays with a layer of fine sand to allow pupation of the larvae at the end of their development. Every week, pupae were collected by overflowing the sand with water and sifting the floating pupae. The pupae were then placed in a small cylindrical aerated plastic emergence box on a piece of cellulose impregnated with a mixture of Nipagin and sodium benzoate to prevent mould development. Every day, emerged adults were transferred to maturation boxes according to their sex and the food regime chosen. Wild adults used in the experiments were 16–20 days old.
The tested oils were GRO (Citrus and Allied Essences, New York, NY, USA) and orange oil (OO) (ref.: W282510, California origin, Kosher, Cold Pressed; from Sigma Aldrich Chimie S.A.R.L., Lyon, France). They were used in pure form during the experiments.
Field cage methodology
Experiments were conducted in large field cages (h = 2.5 m; diameter = 3 m) (Synthetic Industries Gainesville, GA, USA). Each cage contained three potted mango plants (Mangifera indica L.) about 1.7 m high. For experiments on the attractiveness of oils for males, cotton discs (diameter: 6 cm; thickness: 4 mm) impregnated with the tested semiochemicals were suspended at a height of 1.5 m. For experiments on mating success, cohorts of males or females of known age were released early in the morning in the cage at the beginning of the trial. The evolution of the climatic factors (temperature and relative humidity) during the experiments was monitored with a recorder Hobo, U12-012 (Onset Computer Corporation, Pocasset, MA, USA).
Type of experiments
Attractiveness of oils for males of various ages in field cages
The attractiveness of GRO and OO was tested on cohorts of 50 laboratory-reared males of 5, 10, 15 or 20 days old. These males were fed a full diet and had no previous exposure to oils prior to the experiment. A series of tests were carried out in choice conditions between the two oils, with six replicates for each age. Three cotton discs were used in each cage: Two of them were impregnated with 20 μl of oil (one for each of the tested oils), while the third one (control) had no oil. After the release, males present on the different cotton discs were counted every 5 min for 1 h.
Attractiveness according to exposure mode and food regime in the laboratory
The attractiveness of both oils for 15- to 16-day-old males was tested in the laboratory using (i) two exposure modes (‘olfaction’ or ‘olfaction plus contact’) and (ii) two food regimes (‘sugar+protein’ or ‘sugar only’). In the latter case, flies were exposed to the oils in the ‘olfaction plus contact’ mode. Cohorts of 30 males were maintained in small cages (15 × 10 × 20 cm) whose top had a metallic grid where the cotton disc impregnated with one type of oil was placed. In the ‘olfaction’ mode, the cotton wick was raised by two small plastic plugs to prevent any contact with the males. In the ‘olfaction plus contact’ mode, the cotton was placed directly on the grid. The number of males present on the cotton (or close to the cotton in the ‘olfaction’ mode) was counted every 15 min during the 2-h exposure period.
Influence of different factors on male mating success in field cages
All experiments on mating success were carried out in field cages according to the FAO/IAEA/USDA (2003) quality control manual. For each experiment, three cohorts were released into the cage in the morning: 30 mature virgin females that had previously received a ‘full’ (sugar and hydrolysed yeast) food regime, 30 control males and 30 treated males. Mating success was evaluated by the number of pairs formed for each type of male, collected after sunset. Depending on the trial, a total of eight or nine replicates were carried out.
Two days before the experiment, 14- to 15-day-old males were anesthetized with CO2 then marked with a small spot of acrylic paint on the thorax to distinguish them after formation of mating pairs. The two types of male cohorts (control or treated) were marked alternatively throughout the different replicates to avoid any bias linked with the marking technique.
For the test on essential oils, males were exposed in the laboratory to the oil for 2 h, 1 day before releasing them in the field cage. A volume of 20 μl of the chosen oil was deposited on a small cotton wick (2 × 2 cm) placed on the metallic grid at the top of the treatment cage.
For the test on food regime, we investigated whether the adult food regime has an influence on the mating success of the males. Two regimes were tested: sugar-only and ‘full’ diet regimes. In both cases, food was supplied in excess to males, and also to females that always received a ‘full’ food regime. A regime with hydrolysed yeast only could not be tested, as the adults did not survive more than 10 days. Those experiments carried out with laboratory-reared insects were repeated with wild flies to examine whether any modification in the behaviour was because of the rearing process.
To evaluate the combined effect of exposure to oils and food regime, males fed with a given food regime were either exposed or not exposed to each of the two oils. The mating success of each category of males (exposed or not) was then tested in field cages, with the two categories of males competing for females. This series of experiments were only conducted with laboratory-reared males.
The R software (R_Development_Core_Team 2004, version R 2.4.0.) was used for statistical analysis. anova followed by a Student test or a Dunnett test, at a 0.05 threshold, were used to compare the means to determine the effect of the different studied factors. The normality of residues and the homogeneity of variance were tested with a Shapiro test and a Bartlett test.
To determine the effect of male age on the attractiveness of oils, we used anova. When significant differences were observed with the anova, a multiple comparison test (Student test with a Bonferroni correction, at the 5% level) was performed to determine the significant differences between the ages.
To evaluate the effects of the exposure to oil and the effect of the exposure mode on male mating success, the cohorts of control and exposed males were compared independently with a Student test. Then, the effects of oil and exposure mode were evaluated with a Dunnett test, comparing control and exposed males. In this field cage mating test, we used the ratio of possible pairs: number of collected pairs, divided by the number of possible pairs. For the Student tests, we used the ratio of realized pairs: number of collected pairs for each modality, divided by the total number of pairs in the replicate. For analysing the interaction between food regime and exposure to oils, we used two-way anova.
Attractiveness of essential oils for the males of Ceratitis rosa
In all attractiveness experiments, it was noticed that males become very agitated in the presence of essential oils. Their flights were pronounced at the beginning, then oriented towards the source of odour. Males however never sat on the oil spot neither did they ingest the oil.
Attractiveness for males of different ages in field cages
In all experiments, no males landed on the control cotton discs. When both oils were presented simultaneously in the same cage, the number of attracted males was usually higher with GRO, although the attractiveness of both oils was not significantly different for 5- and 10-day-old males for most observation time periods (fig. 1). For 10-day-old males, the difference was significant only between the observations at 50 min post-release (T50) and 60 min post-release (T60), with a higher attractiveness of GRO. For 15- and 20-day-old males, GRO appears significantly more attractive than OO for most of the observation time periods. The difference between the two oils was larger for 20-day-old males. For this age, the difference between the two oils was significant after T20, whereas for 15-day-old males, significant differences only appear at T35, T45, T50, T55 and T60 (P = 0.030, 0.010, 0.005, 0.020 and 0.004, respectively). For males of all ages, OO seems to maintain a similar level of attractiveness, whereas the attractiveness of GRO increases with age and during the observation period (fig. 1).
Attractiveness for 15- to 16-day-old males with two exposure modes in laboratory tests
The evolution with time of the mean number, plus standard error, of males standing on (or near) the cotton impregnated with oil was compared for the two exposure methods: O (‘olfaction’) or OC (‘olfaction plus contact’). For OO, a significant difference was found at T5, T60 and T120, where the attractiveness was higher for the ‘olfaction plus contact’ exposure (P = 0.032, 0.010 and 0.013, respectively) (fig. 2b). For GRO, no significant difference in attractiveness was found between the two exposure modes, although the number of males attracted was slightly higher when only olfaction was allowed (fig. 2a).
Attractiveness for 15- to 16-day-old males previously fed with different diets in laboratory tests
Adult diet showed little effect on the response of males to the two oils in the ‘olfaction+contact’ mode. For OO (fig. 3b), a significant difference was only observed at T75 when the number of sugar-fed males attracted was higher than the number of males fed with the full diet (P = 0.031). For GRO, the number of sugar-fed males attracted was slightly higher than the number of males fed on the ‘full’ diet, but a significant difference was only observed at T90 and T120 (P = 0.037 and 0.008, respectively) (fig. 3a).
Influence of various factors on male mating success
Exposure to oils
Only replicates where the number of formed pairs was superior to three were taken into account for the analyses (nine replicates for the ‘olfaction’ mode of exposure and eight replicates for the ‘olfaction plus contact’ mode).
GRO had a significant effect on improving the male mating performance, whereas no significant difference was found with OO, regardless of the exposure method. Figure 4 shows, for each type of oil, the mean number of formed pairs, and standard errors, for each type of exposure (O or OC) and each type of males (control or exposed males). As for both oils the tests with the two modes of exposure were carried out on the same day, the homogeneity of controls was verified for both oils. As no significant difference was observed (P = 0.9627 for GRO and 0.7291 for OO), the two control cohorts were grouped for each oil. No significant effect of OO was observed for whatever type of exposure (P = 0.722 for the ‘olfaction’ mode of exposure and P = 0.299 for the ‘olfaction plus contact’ mode) (fig. 4b) although males exposed to OO accounted for 54% (exposure O) and 68% (exposure OC) of total matings. On the other hand, a significant effect of GRO on male mating success was observed (P = 0.037 for the ‘olfaction’ mode of exposure and P = 0.016 for the ‘olfaction plus contact’ mode) (fig. 4a), with exposed males accounting for 61% (exposure O) and 64% (exposure OC) of total matings.
Influence of food regime
For these experiments, all replicates (8) were used for the analysis. Laboratory-reared males fed with a ‘full’ regime (sugar+protein) had a significantly higher mating success than males fed only sugar (Student test; P < 0.001) (fig. 5a). A similar result was obtained with wild males collected as larvae in the field (Student test; P < 0.001) (fig. 5b). In general, males fed with a ‘full’ regime accounted for about 85% of the total number of successful mating pairs (139 and 90) in the laboratory-reared and wild males, respectively.
Interaction between food regime and semiochemicals
In these trials, all replicates (8) were used for the analysis for all combinations, except for the combinations GRO – sugar-only regime, where only five replicates were taken into account because the number of mating pairs was very low for the other replicates. Males exposed to GRO had a significantly higher mating success than control males for any of the food regimes they were provided with before the experiment (fig. 6): sugar only (P = 0.008) or sugar plus protein (P = 0.012). For OO, exposed males had a significantly higher mating success than control males, although only when they were fed the sugar-only regime (P = 0.006). No significant difference was obtained between OO-exposed and control males (P = 0.290) when fed the ‘full’ food regime. The two-way anova showed a significant effect of ‘exposure to oil’ (P = 3.862 × 10−5) and also of ‘food regime’ (P = 0.010), while no significant interaction was found between these two factors (P = 0.265).
During the choice tests in field cages, the attractiveness of males to GRO increased with male age, while the attractiveness to OO did not change with age. The difference in attractiveness was clearly associated with male age, with 15-day-old and moreover 20-day-old males showing a pronounced preference for GRO. The significant differences between both oils that begin to appear with 10-day-old males become clearer when males were older, which suggests that the attractiveness for both oils may be of a different nature. Quilici et al. (2002) showed that the males of C. rosa reach their sexual maturity at the age of 15 days, for temperatures varying between 20 and 26°C. The maximum numbers of males attracted to both oils were obtained with 15- and 20-day-old males. Ten-day-old males are only weakly attracted and 5-day-old males show an even lower response, although still significantly different from the control.
Our study also showed that males exposed to GRO mate more than unexposed males, regardless of their previous food regime and the mode of exposure to the oil. It is probable that the exposure to oil influences mating behaviour and/or the quantity or quality of male pheromone, thus increasing the attractiveness of males for the females. Males exposed to OO did not mate significantly more than control males, whatever the exposure mode, when they previously received a ‘full’ food regime. However, when males were previously deprived of proteins, an exposure to OO did significantly increase their mating success. Our results demonstrate a strong influence of GRO on the sexual activity of C. rosa males, while OO apparently shows a lower influence. However, further studies would be useful with both oils to examine different exposure periods, oil quantities or concentrations and time periods between exposure and testing.
In C. capitata, Papadopoulos et al. (2006) also showed a significant effect of the exposure to both OO and GRO on male mating success in laboratory tests. However, when tests were conducted with cohorts of males exposed to each of the oils, the exposure to OO induced a better mating than exposure to GRO. In field conditions, both types of males showed a similar mating success. To confirm our results on C. rosa, it would be interesting to conduct competition tests in field cages, where both types of exposed males would compete for wild females.
The chemical basis for the positive effect of some essential oils on male mating success is still unknown, but some evidence suggests that α-copaene, a plant-borne sesquiterpene hydrocarbon known as a powerful attractant for male C. capitata (Flath et al. 1994a,b), is involved. Both GRO and OO are known to confer a mating advantage to Mediterranean fruit fly males, and both contain this chemical (Nishida et al. 2000 and references therein; Shelly et al. 2004a and references therein). Moreover, Mediterranean fruit fly males exposed to pure α-copaene had a mating advantage over control males (Shelly 2001). Unlike B. dorsalis males with methyl eugenol, C. capitata males do not appear to use the chemical to synthesize their sex pheromone, as neither α-copaene nor structurally related compounds are present in the male pheromone (Millar 1995). Instead, as behavioural tests revealed, exposure to α-copaene-containing oils appears to increase pheromone-calling activity by C. capitata males (Papadopoulos et al. 2006). However, the relative increase in mating frequency far exceeds that observed for signalling, indicating that other factors are probably also involved (Shelly 2001).
The application of small quantities (0.065 μl) of GRO or OO on the wings of C. capitata males did not increase their mating success (Papadopoulos et al. 2006). However, the application of OO on the abdomen increased the mating success of the males, while the application of GRO decreased it. These authors suggest that the oils might penetrate through the cuticle and that the negative effect of GRO could be linked with its toxicity at high concentration. The finding that abdominal application, and presumed subsequent absorption, of OO enhanced male mating success, further suggests that specific components of this oil, other than α-copaene, are used to synthesize a pheromone more likely to immobilize females and to facilitate mating (Papadopoulos et al. 2006). Howse and Knapp (1996) reported 18 compounds to be common in citrus peel volatiles and male Mediterranean fruit fly sex pheromone, which included limonene and linalool, but not α-copaene. Instead, it appears that GRO exposure confers a mating advantage through an external phenomenon (i.e. possibly alteration of cuticular scent) rather than an internal process (i.e. pheromone synthesis). Indeed, Shelly et al. (2007b) suggest the possibility that female preference for GRO-exposed males was mediated by a ‘perfume effect’, whereby the aroma of GRO affected the male cuticle in such a way that enabled females to distinguish between exposed and non-exposed individuals. Nevertheless, additional studies are clearly needed to confirm this phenomenon for both OO and GRO.
The mode of action of compounds influencing the mating success of C. rosa is not yet known, but it seems that these compounds are not ingested. Laboratory tests on attractiveness on 15-day-old males did not show any difference between the two exposure modes, which allow the males to feed or not to feed on the oils. During all attractiveness tests, males did not stay on the oil spots and did not try to ingest the oils.
In addition to α-copaene, various other compounds are present in such oils. GRO also contains, for instance, neptenone, cineol, geraniol, linalool, zingerone and gingerol (Grace Alloca, personal communication). In future studies, it would be interesting to examine the effect of pure α -copaene, or other selected volatiles present in the oil, on C. rosa male mating success.
The food regime influences the progression towards sexual maturity in males of some Ceratitis spp. Thus, in C. capitata, the absence of proteins in the male food regime delays the age when sexual maturity is reached, as was demonstrated by studies of male pheromone calling, using wild flies (Papadopoulos et al. 1998; Kaspi et al. 2000) or laboratory-reared flies (Papadopoulos et al. 1998). Quilici et al. (2002) also showed that an absence of protein in the food regime of laboratory-reared C. rosa males delays the age when they begin to emit their pheromone and decreases the frequency of calling. The present results concord with the previous study.
Indeed, we observed an effect of the food regime on the mating success of C. rosa males: A food regime without proteins induced a lower mating rate (15% of the total number of pairs) for laboratory-reared males as well as for wild males. As discussed earlier, for wild C. capitata too, the presence of proteins in the regime increases the mating success of males, while results are contrasting for laboratory-reared C. capitata (Yuval et al. 2007 and references therein; Joachim-Bravo et al. 2009).
During the tests on interaction between factors, the food regime influenced the total number of matings, which was lower when males were deprived of protein, for both oils. In the case of exposure to GRO, there was a significant difference in mating success of exposed and non-exposed males, regardless of their food regime. However, the exposure to OO did not significantly improve the mating success of males that previously received a ‘full’ regime. In a study with mass-reared irradiated males of C. capitata,Shelly et al. (2003) also observed a significant effect of GRO exposure on male mating success, but no effect of the food regime, and no interaction between those factors.
In various tephritid species, pre-release diet and the exposure of mass-reared irradiated males to certain semiochemicals make it possible to improve their mating success compared with sugar-only-fed and unexposed laboratory-reared irradiated males or wild males. The results of our study indicate that for eventual future programmes of SIT against C. rosa, an enhanced pre-release adult diet and the exposure of males to GRO would be a possible way to improve sterile male sexual performance in the field.
We thank Serge Glénac and Jim Payet for maintaining fruit fly rearings and for help in conducting the experiments. We also thank Pablo Liedo (ECOSUR, Mexico) for useful comments on an earlier version of this manuscript. This research was funded by CIRAD, the Conseil Régional de La Réunion and EU, with additional support from the International Atomic Energy Agency (Research Agreement 12856), within the framework of the FAO/IAEA Co-ordinated Research Project on ‘Improving sterile male performance in fruit fly SIT programmes’.