Naturally infested fruits from northern Thailand (Chiang Mai province, 19°27'48.5"N; 98°57'50.3"E and 19°23'13.9"N; 98°57'58.9"E) and southern Thailand (Nakhon Si Thammarat province, 8°18'25.8"N; 99°37'50.3"E) were collected from host plants and brought to the fruit fly mass-rearing facility of the Department of Agricultural Extension (DOAE) in Pathumthani province. There they were placed on sawdust in containers to let the larvae mature and pupate. Larvae and pupae were kept at 25±2 °C, 80–90% RH and a 12: 12 (L: D) photoperiod. Emerged adult flies were provided a standard diet consisting of enzymatic yeast hydrolysate and sugar (1: 3) with water supplied ad libitum. Wild flies at least 14 days old were identified at the laboratory of the Department of Agriculture, Bangkok, using external morphological characters to confirm their identity in accordance with their taxonomic descriptions (Drew and Hancock 1994). The B. dorsalis colony from Chiang Mai was obtained from the following fruits: mango (Mangifera indica L.), rose apple (Eugenia javanica Lam.) and star fruit (Averrhoa carambola L.), while the B. dorsalis colony from Nakhon Si Thammarat was obtained from Kluai Leb Mu Nang banana (Musa sapientum L.) and guava (Psidium guajava L.). The identity of flies was further confirmed by using the additional diagnostic tools of pheromone and genetic analyses.

Five individual rectal (= pheromone) glands of individually methyl-eugenol fed and non-fed sexually mature wild males from southern Thailand, identified as B. dorsalis using morphological characters, were dissected out and stored in 95% alcohol. Samples were sent to the Laboratory of Entomology and Chemical Ecology, Department of Biology, Faculty of Science, Universiti Putra Malaysia for pheromone analysis. Samples were prepared using sample homogenization and solvent concentration under nitrogen before individual glands were transferred to conical glass vials for GC-MS analyses. The methyl-eugenol metabolites: 2-allyl-4, 5-dimethoxyphenol (DMP) and (E)-coniferyl alcohol (CF), were detected in all samples that were fed with methyl-eugenol. No samples were found to have the endogenous compounds present in B. carambolae males such as (3-methylbutyl) acetamide, ethyl benzoate, benzamide, 6-oxo-1-nonanol and 1, 6-nonanediol. These results confirmed that the southern colony of B. dorsalis (previously B. papayae) was not contaminated with B. carambolae, which is restricted to southern Thailand (Clarke et al. 2001).

The rest of the bodies of the B. dorsalis males from southern Thailand from which the rectal glands were removed for pheromone analysis, together with sexually mature complete females, were sent for genetic analysis in groups of five preserved in propylene glycol to the Diagnostics for Biosecurity laboratory at Lincoln University, Christchurch, New Zealand. DNA was extracted using PrepGem, and PCR amplified and sequenced for ITS1 using the PCR primers reported in Boykin et al. (2013). DNA sequences were aligned using Sequencher and a neighbor joining sequence-similarity representation developed with MEGA. Sequences of species-verified reference specimens from both Boykin et al. (2013) (specifically B. dorsaliss.s. from Taiwan, the Philippines and Malaysia, and B. carambolae from Suriname) and other in-house samples (specifically B. dorsalis [ex B. papayae] from Malaysia and B. carambolae from Suriname and Indonesia, (Mataram, Lombok)) were included in the alignment. The results further confirmed that the colony from the southern population was consistent with B. dorsaliss.s., as per Boykin et al. (2013).

Sexually mature B. dorsalis from each region were exposed to fresh mature Kluai Nam Wa banana (Musa sapientum L.) as an oviposition substrate and larval rearing medium. Bananas with eggs were removed from the rearing cages and placed on sawdust in a ventilated container. Pupae were collected daily and held in 20–25 °C room for maturation. This allowed synchronization of development for the sexual compatibility tests. Eight day-old pupae were manually sifted and transferred to standard quality control Plexiglas cages (30 × 40 × 30 cm) (FAO/IAEA/USDA 2003) with screen mesh windows on two sides and one window on the top for water supply. Standard sugar-yeast diet was provided under low-stress conditions after fly emergence.

Wildish B. dorsalis flies (2^nd and 3^rd generations) from northern (Chiang Mai) and southern (Nakhon Si Thammarat) Thailand were used for mating compatibility studies. Adult flies were sorted by sex within five days of emergence and virgin flies, once sexually mature at 21 and 23 days of age, were selected for the field cage tests. Based on preliminary studies to assure the sexual maturation of wild flies (Orankanok et al. 2013), flies were provided for mating compatibility test at 23 days of age for fertile wild males and at 21 days of age for fertile wild females. Flies were marked by immobilizing them and placing a small dot of acrylic color on each fly’s scutum for both males and females in the early morning of each test day. Colors used were alternated between the two populations. All marked flies for the different replicates were maintained in cylindrical plastic containers, 12.5 cm diameter × 15 cm height, with a triangular mesh on the lid where water-agar and sugar-protein diet was supplied.

Mating compatibility tests between populations from the two regions were performed near the DOAE fruit fly mass-rearing facility. Outdoor octagonal field cages (size 120 cm each side and 220 cm height made from 32 mesh nylon screen), with each cage containing a single potted mango tree (Mangifera indica L.) ca. 180 cm in height. These were used since mate choice experiments in large, walk-in field cages containing a host plant have proven useful tools in discriminating among closely related sibling species (Cayol et al. 1999, Petit-Marty et al. 2004, Vera et al. 2006, Orozco et al. 2007, Cáceres et al. 2009, Schutze et al. 2013, Bo et al. 2014) and as reviewed by Juárez et al. (2015); protocols for such trials are now well established and widely applied.

Six replicates of control tests (NxN and SxS) and inter-regional (NxS) mating compatibility test were carried out during December 2012, with the six replicates of the two control combinations completed in one day, and the six replicates of the inter-regional combination completed on another. General procedures followed those outlined in the FAO/IAEA/USDA (2003) manual. As B. dorsalis mates at dusk (Arakaki et al. 1984), for each replicate, 20 males from each of the two populations under study were released into the field cage 1 to 2 hours before sunset (16.00–16.30 hrs) and 30 minutes before the females, to give the males enough time to establish territories and form leks (Prokopy and Hendrichs 1979, Shelly and Kaneshiro 1991), followed by the release of 20 females from each of the same two populations (16.30–17.00 hrs). Only healthy marked flies were released; non-active or dead flies were replaced. Temperature, relative humidity and light intensity were recorded immediately after females were released and then every half hour. The formation of copulating pairs was observed continuously, and five minutes after initiation of a mating, the mating pairs were collected into small vials. For each mating couple, the following data were recorded: copulation start time, copulation location in cage, temperature, relative humidity, light intensity, male and female color. For position within cage, high elevation was defined as the top of the canopy of the tree or the ceiling of the field cage; low elevation was defined as the mid-lower canopy or the side-lower cage wall. The mated flies were not replaced or released back into the cage after separation (Cayol et al. 1999). Experiments concluded when flies became inactive, which occurred after sunset when light intensity dropped to 0–10 lux (Schutze et al. 2015b).

Sexual compatibility was measured using several indices. The Proportion of Flies Mating (PM) measures the suitability of the flies and the environment for mating and represents the overall mating propensity of the flies: Data are discarded if the proportion of flies mating is less than 20% (FAO/IAEA/USDA 2003). The Index of Sexual Isolation (ISI) takes into account the difference existing between homotypic and heterotypic matings; it ranges from -1 (complete negative assortative mating, that is, all matings are with members of the opposite population) to 0 (complete random mating or equal proportion of the possibilities of mating) to +1 (complete positive assortative mating or total sexual isolation, that is total mating isolation of the two populations and males and females only mated with their respective populations). The Male Relative Performance Index (MRPI) highlights any relative difference between males of both populations in terms of overall mating performance; it ranges from -1 (only males of the reciprocal population mated) to 0 (equal mating performance between males of both populations or males of both populations participated equally in mating) to +1 (only males of one population mated). The Female Relative Performance Index (FRPI) highlights any relative difference between females of both populations in terms of overall mating performance. The range of FRPI is similar to MRPI, but applied to females. The combined application of ISI, MRPI, and FRPI provides a comprehensive measure of mating compatibility, as it demonstrates the degree of isolation between populations or species and the relative participation of the sexes of each population or species (Cayol et al. 1999).

Six replications of ten pairs representing both combinations of crosses of flies between B. dorsalis from northern (N) and southern (S) Thailand were individually assessed for the number of eggs, pupae and F1 adults produced. Eggs of each cross were collected and observed from day 7 until 90 day-old parents. All daily collected eggs of each cross were counted, introduced into semi-ripe bananas and laid on fine sawdust in separately ventilated containers for larval maturation. Pupae were separated and transferred into the cages. Adults of each cross were examined for successful emergence.

Six replicates of field cage mating compatibility studies involving northern B. dorsalis and southern B. dorsalis (ex-B. papayae) were completed. The propensity for mating (PM) values was larger than 0.20 in all replicates, indicating that the conditions under which the tests were run were satisfactory. The mean proportion of flies mating in control tests of NN and SS were 0.57 and 0.33, respectively, and 0.38 for the inter-regional tests (NS).

The ISI value of 0.07 illustrates random mating between northern and southern B. dorsalis populations. Northern males showed slightly higher effectiveness at obtaining mates than males of southern population (MRPI = 0.21), while females of both origins participated equally in mating (FRPI = 0.05) (Figure 1).

Figure 1. 

Index of Sexual Isolation (ISI) and Relative Performance Indices for Males (MRPI) and Females (FRPI) for the field cage mating compatibility tests between the two B. dorsalis populations from northern and southern Thailand.

Total numbers of mated pairs for the control populations of B. dorsalis were lower for the southern population (SS = 80 pairs) and higher for the northern population (NN = 136 pairs). Total pairs across all combinations and comparisons for either homotypic or heterotypic couples of the inter-regional study were not different (SS = 21; NN = 28; SN = 17; NS = 21).

Couples were found mating between 2,000 lux and 35 lux and at temperatures between 28–31 °C. There was no difference in mating latency for couple formation in the two control studies (SS = 67.77 minutes; NN = 70.18 minutes). Averaged across all combinations and comparing homotypic or heterotypic couples of the inter-regional study, there were no differences in mating latency (SS = 69.58 minutes, NN = 70.61 minutes, SN = 71.60 minutes, NS = 72.47 minutes).

More couples in controls and inter-regional mating tests involving the two B. dorsalis populations were collected from high in the field cage (either from the ceiling of the cage or the upper canopy) relative to lower locations; there were no differences between the two population origins in terms of height of the couples inside the field cage (Table 1).

Six replicates of ten pairs for each of the two crosses among B. dorsalis from northern and southern Thailand were completed. No differences were found in the mean number of eggs produced, the mean number of pupae produced, percentage pupal recovery (i.e. egg to pupation percentage), and mean percent adult emergence (Table 2).