The competitiveness of sterile insects is critical to their effectiveness in reducing populations of the target pest by means of the sterile insect technique (SIT). This technique involves large scale production, irradiation, and release of large numbers of sterile insects in the target area, preferably only males, whose task is to mate with wild females so they lay infertile eggs. Sterile males must compete successfully with wild males for females in the field for the SIT to be successful. The loss of mating competitiveness has been demonstrated in mass-reared fruit flies (Lance et al. 2000; McInnis et al. 1996; Rendon et al. 2004) and remains a constant challenge.

Competitiveness can be assessed in different ways and for fruit flies there are well established protocols (FAO/IAEA/USDA 2003), which are constantly being improved (Cayol et al. 1999). In field cage mating competitiveness tests, irradiated males (for male only strains), or irradiated males and females (for bisex strains) together with wild males and females are released into field cages in a 1:1 ratio, mating pairs are captured, duration of matings is recorded and the proportion of matings of sterile males with wild females is used to calculate the Relative Sterility Index (RSI) (McInnis et al. 1996). Alternatively rather than capturing mating pairs, females may be offered the opportunity to oviposit into egging devices, e. g., artificial fruit, hanging in the field cages. Eggs are collected from the fruit and the percentage hatch is determined. This enables a Fried test to be used to calculate a male competitiveness C value (Fried 1971). Variation among replicates tends to be large, so an adequate number of replicates need to be carried out to obtain valid results.

In area-wide integrated pest management (AW-IPM) programs (Vreysen et al. 2007) with a SIT component there is always a compromise between a dose rate that gives high sterility but less competitive males and one that ensures better competitiveness but lower sterility (Lance & McInnis 2005; Suckling et al. 2011). Moths as a group require relatively large doses to induce sterility. Thus, Bakri et al. (2005) quoted 278 Gy as a mean sterilization dose for Tortricidae. However, one can take advantage of the existence in irradiated moths of the inherited sterility (IS) phenomenon (Carpenter et al. 2005), which is absent in irradiated flies. In most cases female moths are sterilized by a lower dose than male moths and the F₁ progeny resulting from a mating between partially sterile males and wild females are more sterile than their parents and these F₁ progeny are predominantly male. Carpenter et al. (2005) commented “the lower dose of radiation used to induce F₁ sterility increases the quality and competitiveness of released insects as measured by improved dispersal after release, increased mating ability and superior sperm competition”.

Saour (2014) noted that to use F₁ sterility, a dose is usually chosen that fully sterilizes females while only partially sterilizes males. He recommended a dose of 150 Gy for F₁ sterility against the European grapevine moth, Lobesia botrana (Denis & Schiffermüller) (Lepidoptera, Tortricidae), although males were still partially fertile at 400 Gy. In radiation biology trials with the light brown apple moth (LBAM), Epiphyas postvittana Walker (Lepidoptera: Tortricidae) a very small percentage of eggs from female moths irradiated with 200 Gy hatched and developed to the pupal stage, with 250 Gy eggs hatched but no progeny developed, and with 300 Gy no eggs hatched. For males there was 2% egg hatch and larval survival when male parents were irradiated with 300 Gy (Jang et al. 2012). Therefore they recommended a dose of 250–300 Gy for a SIT program against LBAM if parental sterility was the desired outcome but that a lower dose could be used if F₁ hybrid sterility (IS) was desired to obtain “increased competitiveness of moths”.

In moths different methods such as flight mills, wind tunnels, flight cylinders and mark-release-recapture trials (Stringer 2013) have been used to measure certain components of sterile insect competiveness. Suckling et al. (2011) used wind tunnels and mark-release-recapture in hedgerows and vineyards to test flight ability and dispersal of LBAM irradiated at various doses. Mark-release-recapture studies and mating tables with virgin females were used to measure field competiveness of diapausing and non-diapausing codling moths, Cydia pomonella L. (Lepidoptera: Tortricidae) in Canada (Bloem et al. 2004). A mating table consists of virgin female moths with clipped wings that are confined to small arenas hung in the tree canopy and that allow mating pairs to be collected and identified. Walk-in field cages have been used to demonstrate either the influence of radiation dose or the release ratio on population growth in the cactus moth, Cactoblastis cactorum (Berg) (Lepidoptera: Pyralidae) (Hight et al. 2005), the codling moth (Bloem et al. 1999) and the false codling moth, Thaumatotibia leucotreta (Meyrick) (Lepidoptera: Tortricidae) (Hofmeyer et al. 2005). Likewise, walkin field cages were used to assess mating compatibility of codling moth populations originating from various geographical areas (Taret et al. 2010). The use of field cages to test mating competitiveness of irradiated moths—as has been done with fruit flies—has not been reported.

Western Australia is fortunate in that 2 key pests of perennial horticulture, the codling moth and the Oriental fruit moth, Grapholita molesta (Busck) (Lepidoptera: Tortricidae), have not established in the state, and since 1900 over 20 campaigns have been carried out to eradicate incursions (Woods et al. 2001). Eradication was achieved by tree and fruit removal and tightly scheduled use of insecticides over the entire growing season. These techniques are no longer economically or socially viable and other approaches such as the SIT are needed. The native tortricid species LBAM was chosen as the model species to develop alternative area-wide eradication strategies that include an SIT component.

Initial work identified the irradiation dose required to induce sterility in the species (Soopaya et al. 2011). Thereafter trials were carried out on competiveness of irradiated moths using doses between 200 and 300 Gy, the range most likely to be used in an eradication program. Techniques and walk-in field cages used for assessing Mediterranean fruit fly competitiveness were adapted for use with LBAM and bisex and male-only releases were investigated. LBAM females will lay eggs on any smooth surface with up to 150 eggs per batch (Brown et al. 2010) and up to 600 eggs per female in the laboratory (Bill Woods, unpublished data). The potential for easily separating and counting egg batches from individual females provides options not available for tephritid fruit flies.

The objective of the trials was to investigate the potential use of walk-in field cages to measure mating competitiveness of moths, and to determine whether the release of only sterile males resulted in any competitive advantage as had been shown with tephritid fruit flies. Three trials were carried out initially using the standard techniques as used with fruit flies, but in subsequent trials they were modified to make the techniques more applicable for use with moths.

Materials and Methods

All field cage trials were carried out at the Department of Agriculture and Food in South Perth, Western Australia. Trial 1 took place in Mar 2008, trial 2 in Apr 2009 and trial 3 in Sep and Oct 2009. Mean minimum and maximum temperatures (°C) for these months in Perth were 16.7 and 29.4 (Mar 2008), 13.5 and 27.6 (Apr 2009), 9.8 and 18.6 (Sep 2009), and 12.3 and 22.3 (Oct 2009). Over the period of the trials maximum and minimum temperatures at Perth varied from 26.4 to 32.7 (maximum) and 17.8 to 21.7 (minimum) (3–7 Mar 2008), from 24.9 to 29.7 (maximum) and 11.3 to 15.5 (minimum) (15–19 Apr 2009), from 16.8 to 19.8 (maximum) and 6.2 to 14.6 (minimum) (8–14 Sep 2009), and from 23.0 to 36.9 (maximum) and 9.8 to 13.1 (minimum) (14–17 Oct 2009) ( http://www.bom.gov.au/climate/data/). Sunset was near 18.50 h in Mar 2008 and 17.50 h in Apr 2009. In Oct 2009 the light level inside the field cages was reduced from around 400–600 lux at 19.30 h to near zero by 20.30 h. RH varied between 50–80% from 19.30 h to 22.30 h.

FIELD CAGES

Walk-in field cages consisted of a cylindrical tent 2 m tall × 3 m diam made with moth proof nylon. Each cage contained in the middle an artificial Ficus benjamina tree approximately 1.5 m tall. Moths were released close to dusk. We aimed for 25 ♂ and 25 ♀ moths per moth type in each field cage. However this was not always possible due to moth availability. In these situations, the numbers of irradiated and unirradiated males in a field cage was always equal or only varied by one moth due to mortality during transfer from containers to field cages. Mating pairs were captured in vented 40 mL plastic vials and mating type, time and place of capture were recorded. Temperature and humidity in the field cage were recorded every 30 min. LED headlamps either on red or white settings were used and one person manned each field cage for 4–7 h after moth release, depending on temperature and mating activity.

Table 1.

Field cage set up to assess mating competitiveness of the light brown apple moth (LBAM) in trials 1, 2 and 3. To the extent possible, untreated and irradiated moths were released into each field cage in a 1:1 ratio, i.e., 25 ♂ and 25 ♀ moths per moth type in each field cage. However this was not always possible due to moth availability. In these situations, the numbers of irradiated and unirradiated males in a field cage was always equal or only varied by one moth due to mortality during transfer from containers to field cages.

INDICES OF MATING COMPETITIVENESS

The formulas of various indices of mating competitiveness are described in the FAO/IAEA/USDA (2003) Quality Control Manual.

Proportion of Mating (PM)

PM measures the suitability of the tested insects and the field cage environment for mating and represents overall mating activity. For fruit flies a value of 0.5 is considered adequate, whereas a proportion of less than 0.2 will not provide meaningful data. The proportion of moths mating (PM) was defined as:

Relative Sterility Index (RSI)

The RSI is the proportion of wild females that mate with sterile males when sterile and wild males are present in a 1:1 ratio. Thus RSI is an indicator of sterile male mating performance (McInnis et al. 1996), and it was defined as:

where SW is the number of matings between sterile males and wild females and WW is the number of matings between wild males and wild females. The RSI equals the number of sterile matings divided by the number of all matings, and its values can vary between 0 and 1, with 0.5 indicating equal performance of wild and sterile males.

Index of Sexual Isolation (ISI)

The ISI (Cayol et al. 1999) is a measure of sexual compatibility between 2 strains. It measures the preference of insects to mate with individuals of the same versus another strain, and the index takes into account both within strain and between strain matings. This index was defined as:

where WW is the number of matings between wild females and wild males, LL is the matings between laboratory females and laboratory males, WL is the matings between wild females and lab males and LW is the matings between laboratory females and wild males. Calculation of the index does not require that both sexes are represented, so it can be used in both bisex and male-only situations. Thus in the case of complete behavioral isolation, the ISI would be 1, and with no isolation (random mating), the ISI would be 0. Hence the smaller the ISI value, the less the behavioral isolation between the 2 strains. Also, positive or negative ISI values indicate positive or negative assortative mating.

Male Mating Competitiveness (C) Based on Egg Hatch

Fried (1971) proposed a method to estimate the competitiveness for sterilized insects that is independent of the ratio of treated to normal insects used, and that can he determined from data collected from a test involving only 1 sterile to fertile ratio, provided that the egghatch data are known for matings between untreated insects and matings between sterilized and untreated insects. Thus Fried's formula for calculating male mating competitiveness (C) based on egg hatch data is as follows:

where W is the number of wild males in the field cage, S is the number of sterile males in the field cage, Hc is the is the egg hatch percentage from wild females in the control cage, Ht is the egg hatch percentage from wild females in the field cage for a particular treatment, and Hs is the egg hatch percentage from females mated with sterile males in the laboratory sterility cage. A Fried test C value of 1 indicates equal competitiveness between sterile and wild males, and values < 1 indicate that the irradiated release strain is less competitive than the wild strain.

In trial 2, the C value was calculated in 2 ways but in all cases the same field cage egg hatch (Ht) was used in all computations with the above formula. Firstly, C was calculated by using hatch data from the laboratory cages as is normal practice to determine Hc and Hs. However we also used egg hatch from individual moths from the field cages, aggregated by mating type, to determine Hc and Hs and thus calculate the C value and we termed this C1. In trial 3, the C value was also calculated in this way. Mating pairs were not captured but mating type, e.g., sterile male × wild female, was calculated using the egg hatch value from the “wild” females. This was possible because females were collected and egged individually, and in the case of a bisex release treatment either the sterile or wild female was marked. This marking was not necessary for male-only releases. Moths each with 1 or more spermatophores, which indicated that they had mated, and with an egg hatch rate of less than 7% were considered to have mated with an irradiated male. This was based on published data (Jang et al. 2012) which indicated that eggs oviposited by untreated females mated with 300 Gy irradiated male moths had an average hatch of 3.8%. Other unpublished data (Don McInnis, personal communication) indicated that eggs oviposited by untreated females that had mated with males that had been irradiated with 300 Gy as pharate pupae had an average hatch of 5.1 ± 0.8%. Therefore egg hatch equal to or below 7% was considered to be from a wild female mated with an irradiated male and above 7% from a wild female mated with a wild male.

Results

Trial 3

In trial 3, the PM was calculated based upon the number of females that had spermatophores, and the average PM was 0.6. Values for RSI and C are presented in Table 4. Though the data for both parameters showed an advantage in favor of male-only releases, that advantage was not statistically significant for the RSI (t = -1.19; df = 6; P = 0.28) and the C values (t = -1.75; df = 6; P = 0.131). Individual egg hatch values of the different replicates used to calculate the C value are presented in Table 5.

Table 2.

Proportion of Mating (PM), Relative Sterility Index (RSI), and Index of Sexual Isolation (ISI) of irradiated male LBAM moths in field cages (Trial 1). Treatments were replicated 3 times.

Table 3.

Proportion of Mating (PM), Relative Sterility Index (RSI), Index of Sexual Isolation (ISI), and male mating competitiveness (C) of irradiated male LBAM moths in field cages (Trial 2). Treatments were replicated 4 times. Calculation of C involved the use of hatch data from the laboratory cages to determine Hc and Hs—as is normal practice. However the calculation of C1 involved the use of egg hatch data from individual moths from the field cages, aggregated by mating type, to determine Hc and Hs.

In the laboratory unirradiated females that had mated with irradiated males contained an average of 1.4 spermatophores/female with a maximum of 2 spermatophores/female, whereas unirradiated females that had mated with unirradiated males contained an average of 1.6 spermatophores/female with a maximum of 3 spermatophores/ female. In moths collected from field cages in the morning, the number of spermataphores from mated females varied from 1 to 2 with a mean of 1.03 for females mated with irradiated males and 1.13 for females mated with unirradiated males. Overall 7% of females from field cages had mated twice.

Discussion

Values of the ISI can vary from -1 to +1 with 0 indicating random mating, and positive or negative values indicating positive or negative assortative mating. In trial 1 where irradiated and non-irradiated moths were all from the same laboratory strain—thus, there should have been no mating isolation—values varied from 0.02 to 0.46 but were statistically not significantly different. In the case of the Mediterranean fruit fly, ISI values usually varied between 0.1 and 0.4 and values above 0.5 indicate a moderate to high level of positive assortative mating (FAO/IAEA/FAO 2003). In trial 2, where laboratory moths were of South Australian origin and wild moths were from Western Australia values varied from 0.14 to 0.31 demonstrating sexual compatibility and only a slight tendency for homotypic (i.e., like with like) mating (Cayol et al. 1999).

In our trials when both males and females were irradiated and released there was little difference in competitiveness at different doses when measured as proportion of matings by irradiated males (RSI). Therefore, an advantage of using a dose lower than 300 Gy would not come from an increase in sterile moth competitiveness; rather, an advantage might come from inherited sterility (Kean et al. 2008), and this would only be possible if the moths were irradiated below 250 Gy (Soopaya et al. 2011; Jang et al. 2012).

Table 4.

Relative sterility index (RSI), and male mating competitiveness (C) of irradiated male LBAM moths in experiments involving either treated males and females (bisex releases) or treated males-only that were released into walk-in field cages (Trial 3). Treatments were replicated 4 times. The RSI estimates involved the use of mating data based on the presence of spermatophores in the female.. The calculation of C1 involved the use of egg hatch data from individual moths from the field cages, aggregated by mating type, to determine Hc and Hs.

Table 5.

Egg hatch data for individual replicates in trial 3 from which the C values were calculated.

The greatest, but not significantly different, RSI values were obtained only when male-only releases were made in the field cages, suggesting there may be an advantage in this approach. It has been assumed that bisex releases of lepidopterans are effective because females attract mates and reduce matings of wild males with wild females, and this would outweigh the disadvantage of sterile females being a sperm sink for sterile males (Stringer et al. 2013). In the case of tephritid fruit fly SIT, the release of only males avoids the fruit damage caused by oviposition attempts by irradiated females (sterile stings), avoids matings between sterile males and sterile females and increases dispersal of sterile males searching for wild females (Rendon et al. 1998). In the present case, until an effective method of sexing pupae of LBAM is developed the advantage of releasing only males can be tested only at a small scale; hence its effectiveness at a large field scale will remain only hypothetical.

The Fried test gives a measure of true mating competitiveness as expressed by egg hatch. In trials 2 and 3, the obtained C values were above 0.3 which is considered acceptable for fruit flies (USDA/FAO/IAEA 2003). In trial 2, there was no difference in C values whether they were calculated using sterility and egg hatch data from untreated insects kept in small laboratory cages or by using egg hatch data of individual moths collected from field cages. The advantage of collecting and egging individual moths and using an aggregate of the values to calculate C means that all values can be obtained from the field cage matings, thus avoiding the need to set up additional laboratory cages. This is advantageous in terms of logistics and in terms of deriving a result that is closer to natural mating behavior in the wild.

In trial 3, a minimum intervention approach was used with both bisex and male-only releases. Not only were the RSI and C values greater than in the other trials, but C values of the male-only treatment were greater than those of the bisex treatment, although not significantly so. This approach probably best mirrors the real world situation as impacts from observers are minimized. However, this approach is dependent on the assignment of the different mating types on the basis of egg hatch. Both sterile ♂ × sterile ♀ and sterile ♂× wildish ♀ matings inevitably will produce some, but consistently low, egg hatch. Therefore, an upper limit on the average egg hatch value below which occur virtually all hatch values from these mating types, must be set—albeit somewhat arbitrarily; and in our studies with LBAM, we set it at 7%.

However, infrequently a mating between a wild male and a wild female is consummated with the transfer of a spermatophore, but without the production of viable eggs. This, combined with the wide variability of egg hatch from wild ♀ × wild ♂ matings (Bill Woods, unpublished data) poses a question as to the validity of this minimal intervention approach. Trials that would compare the egg hatch from females mated to marked male moths with the hatch of eggs of females mated to unmarked male moths would help determine the accuracy of mating assessments based on egg hatch alone. Additionally the conduct of a much larger number of replicates would reduce the standard error, improve the accuracy of mating assessment based on egg hatch, and increase the potential for detecting significant difference among treatments.

In general, male tortricid moths mate more than once (Howell 1991), and LBAM is no exception. Stringer (2013) reported an average of 4.8 matings per laboratory female of non-irradiated males compared to 1.3 matings per female by males irradiated with 300 Gy in laboratory trials. This may impact on the effectiveness of the SIT, and the most effective way to reduce this impact is by use of a larger sterile to wild male over-flooding ratio in the field. Of more concern is whether wild females mated to sterile males are more likely to remate than wild females mated first with wild males. These potential negative impacts on the SIT need to be assessed experimentally and would require that females can actively discriminate between fertile and sterile males and that the sperm from successive matings do not compete in a random manner.

Multiple mating of female moths was observed in this study, especially under laboratory conditions. When provided with multiple mating opportunities, LBAM females mated up to 3 times. In other laboratory trials, the pairing of 1 female with 3 male moths (Soopaya et al. 2011), resulted in the transfer of up to 5 spermataphores per dissected female (Bill Woods, unpublished data). Knight (2007) reported that codling moth females paired at the same 1 to 3 ratio, contained an average 2.2 spermatophores per female upon being dissected, ranging from 1 to 7 per female. Therefore some females of these 2 tortricid species will mate more than once under laboratory conditions when multiple partners are available. However in our trials, the situation in the field cages was very different and only limited multiple matings occurred, irrespective of whether males were irradiated or unirradiated. It could be argued that this was because females were only in the field cage for one night, so a longer mating period may have resulted in more matings. However a result similar to our result was observed in trials performed in Perth (Bill Woods, unpublished data) in which live virgin females were placed in traps in the field for 3 d, and later dissected for evidence of mating. An average of 1.1 spermatophores per female was observed with a maximum of 2 spermataphores per female.

Knight (2007) noted that the codling moth was polyandrous in crowded laboratory conditions and in orchards with high population densities, but Howell (1991) observed that mated codling moth females generally discontinue calling and do not remate. Stringer et al. (2013) used stable carbon isotopic analysis of LBAM spermataphores to determine whether females caged in small field plots had been mated with wild or a small number of released irradiated laboratory males. They showed that 97% of the mated females had mated only once (range 1–3) and 13% of matings were from irradiated males. Use of this technique to separate matings from sterile or wild males has potential to study the impact of multiple mating on the effectiveness of the SIT, but it appears that multiple mating is unlikely to have a major impact especially if the wild population is reduced to low levels before sterile moths are released.

Field cage tests are accepted as the best way to measure competitiveness of sterile fruit flies (Cayol et al. 1999; FAO/IAEA/USDA 2003; Cáceres et al. 2007). As Shelly (2004) commented “the success of the SIT depends, to a large degree, on the ability of released sterile males to attract and obtain matings with wild females. This consideration is especially important for a species, such as C. capitata, in which females display a high degree of mate discrimination”. This is probably not the case with most moth species whose females do not actively choose mates and leks do not occur. Male moths orientate along a pheromone plume to find the female and mate without any obvious courtship behavior (Suckling 2011). In these simple mating systems (Lance & McInnis 2005), males need to fly and respond to females that are calling, and this will usually result in a mating.

Our study assessed the mating competitiveness of LBAM in field cages for the first time. We do not yet know what are acceptable RSI and C values for irradiated moths but it appears that irradiation at the doses tested does not impact significantly on mating performance. The use of a bisex strain in which female as well as male competitiveness are important, requires the development of new mating indices to tease out the relative importance of the competitiveness of each sex. Modifying the male relative performance index (MRPI) and the female relative performance index (FRPI), which measure differences in mating between wild and sterile males and wild and sterile females, may have potential (Cayol et al. 1999). The practical difficulties of dealing with a species which mates after dark will need to be overcome, and light, temperature and wind thresholds for mating need to be investigated. The use of a modified Fried test with minimal human intrusion into the mating environment is an approach that showed promise in this study.