Synonymization of key pest species within the Bactrocera ... · Synonymization of key pest species within the Bactrocera dorsalis species complex (Diptera: Tephritidae): taxonomic

Systematic Entomology (2014), DOI: 10.1111/syen.12113

Synonymization of key pest species within theBactrocera dorsalis species complex (Diptera:Tephritidae): taxonomic changes based on a reviewof 20 years of integrative morphological, molecular,cytogenetic, behavioural and chemoecological data

M A R K K . S C H U T Z E 1,2, N I D C H A Y A A K E T A R A W O N G 3,W E E R A W A N A M O R N S A K 4, K A R E N F . A R M S T R O N G 5, A N T O N I SA . A U G U S T I N O S 6,7,8, N O R M A N B A R R 9, W A N G B O 6,7,10, K O S T A SB O U R T Z I S 6,7, L A U R A M . B O Y K I N 11, C A R L O S C Á C E R E S 6,7,S T E P H E N L . C A M E R O N 1, T O N I A . C H A P M A N 2,12, S U K S O MC H I N V I N I J K U L 13, A N A S T A S I J A C H O M I C 5, M A R C D E M E Y E R 14,E L L E N A D R O S O P O U L O U 15, A N N A E N G L E Z O U 2,12, S U N D A YE K E S I 16, A N G E L I K I G A R I O U - P A P A L E X I O U 17, S C O T T M . G E I B 18,D E B O R A H H A I L S T O N E S 2,12, M O H A M M E D H A S A N U Z Z A M A N 19,D AV I D H A Y M E R 20, A L V I N K . W . H E E 21, J O R G E H E N D R I C H S 6,7,A N D R E W J E S S U P 22, Q I N G E J I 10, F A T H I Y A M . K H A M I S 16,M A T T H E W N . K R O S C H 2,23, L U C L E B L A N C 24, K H A L I DM A H M O O D 25, A N N A R . M A L A C R I D A 26, P I N E L O P I M AV R A G A N I- T S I P I D O U 15, M A U L I D M W A T A W A L A 27, R I T S U O N I S H I D A 28,H A J I M E O N O 28, J E S U S R E Y E S 6,7, D A N I E L R U B I N O F F 24,M I C H A E L S A N J O S E 24, T O D D E . S H E L L Y 29, S U N Y A N E ES R I K A C H A R 30, K E N G H . T A N 31, S U J I N D A T H A N A P H U M 32,I H S A N H A Q 6,7,33, S H A N M U G A M V I J A Y S E G A R A N 1, S U K L . W E E 34,F A R Z A N A Y E S M I N 19, A N T I G O N E Z A C H A R O P O U L O U 17 andA N T H O N Y R . C L A R K E 1,2

1School of Earth, Environmental and Biological Sciences, Queensland University of Technology, Brisbane, Australia, 2PlantBiosecurity Cooperative Research Centre, Canberra, Australia, 3Department of Biotechnology, Faculty of Science, MahidolUniversity, Bangkok, Thailand, 4Department of Entomology, Faculty of Agriculture, Kasetsart University, Bangkok, Thailand,5Bio-Protection Research Centre, Lincoln University, Christchurch, New Zealand, 6FAO/IAEA Division of Nuclear Techniques inFood and Agriculture, Vienna, Austria, 7FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, Seibersdorf, Austria,8Department of Environmental and Natural Resources Management, University of Patras, Agrinio, Greece, 9Center for PlantHealth Science and Technology, Mission Laboratory, USDA-APHIS, Edinburg, TX, U.S.A., 10Fujian Agriculture and ForestryUniversity, Fuzhou, China, 11ARC Centre of Excellence in Plant Energy Biology and School of Chemistry and Biochemistry, TheUniversity of Western Australia, Perth, Australia, 12Elizabeth Macarthur Agricultural Institute, NSW Department of PrimaryIndustries, Menangle, Australia, 13Department of Agricultural Extension, Ministry of Agriculture and Cooperatives, Bangkok,Thailand, 14Invertebrates Section, Royal Museum for Central Africa, Tervuren, Belgium, 15Department of Genetics, Developmentand Molecular Biology, School of Biology, Faculty of Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece,

Correspondence: Mark K. Schutze, School of Earth, Environmental and Biological Sciences, Queensland University of Technology, GPO Box2434, Brisbane, Queensland 4001, Australia. E-mail: [email protected]

© 2014 The Royal Entomological Society 1

2 M. K. Schutze et al.

16International Centre of Insect Physiology and Ecology (ICIPE), Nairobi, Kenya, 17Department of Biology, University of Patras,Patras, Greece, 18Tropical Crop and Commodity Protection Research Unit, Pacific Basin Agricultural Research Center, USDAAgricultural Research Services, Hilo, HI, U.S.A., 19Institute of Food and Radiation Biology, Atomic Energy Research Establishment,Bangladesh Atomic Energy Commission, Dhaka, Bangladesh, 20Department of Cell and Molecular Biology, University of Hawaii,Honolulu, HI, U.S.A., 21Department of Biology, Faculty of Science, Universiti Putra Malaysia, Serdang, Malaysia, 22Central CoastPrimary Industries Centre, NSW Department of Primary Industries, Gosford, Australia, 23Centre for Water in the Minerals Industry,Sustainable Minerals Institute, University of Queensland, St Lucia, Australia, 24Department of Plant and Environmental ProtectionSciences, University of Hawaii at Manoa, Honolulu, HI, U.S.A., 25Pakistan Museum of Natural History, Islamabad, Pakistan,26Department of Biology and Biotechnology, University of Pavia, Pavia, Italy, 27Department of Crop Science and Production,Sokoine University of Agriculture, Morogoro, Tanzania, 28Graduate School of Agriculture, Kyoto University, Kyoto, Japan,29USDA-APHIS, Honolulu, HI, U.S.A., 30Department of Agriculture, Ministry of Agriculture and Cooperatives, Bangkok, Thailand,31Tan Hak Heng, Penang, Malaysia, 32Fruit Fly Molecular Genetic Laboratory, Department of Biotechnology, Faculty of Science,Mahidol University, Bangkok, Thailand, 33National Agricultural Research Centre, Islamabad, Pakistan and 34School ofEnvironmental and Natural Resource Sciences, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, Bangi, Malaysia

Abstract. Bactrocera papayae Drew & Hancock, Bactrocera philippinensis Drew &Hancock, Bactrocera carambolae Drew & Hancock, and Bactrocera invadens Drew,Tsuruta & White are four horticultural pest tephritid fruit fly species that are highlysimilar, morphologically and genetically, to the destructive pest, the Oriental fruit fly,Bactrocera dorsalis (Hendel) (Diptera: Tephritidae). This similarity has rendered thediscovery of reliable diagnostic characters problematic, which, in view of the economicimportance of these taxa and the international trade implications, has resulted in ongoingdifficulties for many areas of plant protection and food security. Consequently, a majorinternational collaborative and integrated multidisciplinary research effort was initiatedin 2009 to build upon existing literature with the specific aim of resolving biologicalspecies limits among B. papayae, B. philippinensis, B. carambolae, B. invadensand B. dorsalis to overcome constraints to pest management and international trade.Bactrocera philippinensis has recently been synonymized with B. papayae as a result ofthis initiative and this review corroborates that finding; however, the other names remainin use. While consistent characters have been found to reliably distinguish B. carambolaefrom B. dorsalis, B. invadens and B. papayae, no such characters have been found todifferentiate the latter three putative species. We conclude that B. carambolae is a validspecies and that the remaining taxa, B. dorsalis, B. invadens and B. papayae, representthe same species. Thus, we consider B. dorsalis (Hendel) as the senior synonym ofB. papayae Drew and Hancock syn.n. and B. invadens Drew, Tsuruta & White syn.n. Aredescription of B. dorsalis is provided. Given the agricultural importance of B. dorsalis,this taxonomic decision will have significant global plant biosecurity implications,affecting pest management, quarantine, international trade, postharvest treatment andbasic research. Throughout the paper, we emphasize the value of independent andmultidisciplinary tools in delimiting species, particularly in complicated cases involvingmorphologically cryptic taxa.

Introduction

The Bactrocera dorsalis species complex is a group of true fruitflies (Diptera: Tephritidae) that contains almost 100 morpholog-ically similar taxa (Drew & Hancock, 1994; Drew & Romig,2013). Most species in this complex are of no economic concern;however, the Oriental fruit fly, Bactrocera dorsalis (Hendel), andclosely related species in this complex, namely the Asian Papaya

fruit fly, Bactrocera papayae Drew & Hancock, the Philip-pine fruit fly, Bactrocera philippinensis Drew & Hancock, theCarambola fruit fly, Bactrocera carambolae Drew & Hancock,and the Invasive fruit fly, Bactrocera invadens Drew, Tsuruta &White (Fig. 1), are arguably amongst the world’s most impor-tant pests of horticulture (Clarke et al., 2005; Khamis et al.,2012). The native geographic distributions of these taxa spanthree continents: B. dorsalis ranges from the Indian subcontinent

© 2014 The Royal Entomological Society, Systematic Entomology, doi: 10.1111/syen.12113

B. papayae, B. invadens, and B. dorsalis synonymy 3

Fig. 1. Bactrocera (Bactrocera) dorsalis female. Photo credit: AnaRodriguez.

(and Andaman Island), across into southern China and mainlandSouth-east Asia, and as far as southern Thailand (Aketarawonget al., 2007; Drew & Romig, 2013); B. papayae occurs fromsouthern Thailand into much of the Indo/Malay Archipelago;B. carambolae is largely sympatric with B. papayae, in addi-tion to being widespread in the Andaman and Nicobar Islands(David & Ramani, 2011); B. philippinensis occurs only in thePhilippines; and B. invadens is presumed native to parts of theIndian subcontinent (Drew & Hancock, 1994; Drew et al., 2005;Drew & Romig, 2013). More recently, the range of B. dorsalishas expanded into Hawaii and a number of South Pacific Islands(Stephens et al., 2007), B. carambolae into northern SouthAmerica (van Sauers-Muller, 1991), B. papayae into the islandof New Guinea (Drew & Romig, 1997), and B. invadens acrossmost of sub-Saharan Africa (Khamis et al., 2009). These highlydisjunct geographic distributions, combined with prolonged dif-ficulties in differentiating putative species and developing reli-able diagnostic markers, have resulted in significant problemsassociated with international trade, quarantine, phytosanitation,food security and fundamental research with respect to thesehighly damaging and invasive pest taxa.

Bactrocera papayae, B. philippinensis, and B. carambolaewere described in a major revision of the B. dorsalis complex byDrew & Hancock (1994). With the exception of B. carambolae(see later), characters provided in the descriptions of thesetaxa are often variable and inadequate for confident speciesidentification, and the search for consistent diagnostic charactershas been extremely problematic (Clarke et al., 2005).

Bactrocera invadens was detected in Kenya in 2003, at whichpoint it was considered likely to represent a morphologicallyvariable form of B. dorsalis (Lux et al., 2003). Nevertheless,B. invadens was described as a new species following compar-ison of Kenyan material with specimens from the purportednative range of Sri Lanka (Drew et al., 2005). With the holotypedesignated from the invasive population in Kenya, B. invadenswas determined to be a new species, due in particular to a scu-tum colour ranging from pale red-brown to black, with the exis-tence of variable lanceolate-patterned intermediates; the scutum

of B. dorsalis is strictly defined as mostly black (Drew & Han-cock, 1994). Debate over the relationship between B. invadensand B. dorsalis has persisted, particularly as some specimens ofB. invadens possess a black scutum and are ‘almost inseparable’from B. dorsalis (Drew et al., 2005: 153). This situation is fur-ther compounded, as ‘occasionally the thorax [of B. dorsalis] isalmost entirely red-brown’ (Drew & Hancock, 1994: 18), with‘pale forms of B. dorsalis [occurring] in less than 20% of thepopulation’ (Drew & Romig, 2013: 100). Confusion over theunreliable discrimination between B. dorsalis and B. invadenshas resulted in significant problems with African horticultureand food security (Khamis et al., 2012).

Although the search for discriminatory characters has failedto yield universally accepted diagnoses, research into the bio-logical relationships among these taxa has generated con-siderable evidence that B. papayae, B. philippinensis, andB. invadens – but not B. carambolae – are the same biologi-cal species as B. dorsalis. Here, we review previous literature,in addition to that generated from recent research, much ofwhich was carried out in conjunction with a 6-year FAO/ Inter-national Atomic Energy Agency (IAEA) Coordinated ResearchProject (CRP) on the ‘Resolution of cryptic species complexesof tephritid pests to overcome constraints to SIT [Sterile InsectTechnique] application and international trade’ involving morethan 40 researchers from 20 countries. As will be shown,this multidisciplinary approach has generated consistent find-ings that form the basis for our taxonomic reassignments. Ourwork is done within an integrative taxonomic framework (sensuSchlick-Steiner et al., 2010) and we stress that while each indi-vidual dataset may be debated with respect to its relative valuefor species delimitation, the multiple lines of evidence across arange of disciplines undertaken by independent research groupsspanning a period of 20 years provide a compelling case for tax-onomic revision. In defining biological species limits throughintegrative taxonomy, we follow here the unified species conceptof de Queiroz (2007), which defines species as separately evolv-ing metapopulation lineages. In line with this theory, we pre-dict that at least one attribute (e.g. mating isolation, reciprocalgenetic monophyly or morphological discontinuity) would beconsistent with current taxonomy if these taxa are different bio-logical species. If they are the same biological species, we expectno such differences. Discussions of the theoretical and practicalapproaches underpinning the current study can be found in deQueiroz (2007) and Schlick-Steiner (2010), respectively, and asapplied specifically to tephritids in Clarke & Schutze (in press).

Our review is broken into three major sections: (i) evidencefor the synonymization of B. papayae and B. philippinensis withB. dorsalis; (ii) evidence for the synonymization of B. invadenswith B. dorsalis; and (iii) evidence for maintaining B. carambo-lae as a distinct species. Additionally, given that the recent revi-sion by Drew & Romig (2013) synonymized B. philippinensiswith B. papayae, but retained the other taxa, we include a sum-mary that explicitly addresses the arguments made by Drew &Romig (2013) against synonymizing B. papayae and B. invadenswith B. dorsalis. Following the review component of the paper,we provide the taxonomic component of the work which makesformal statements of synonymization.



At points we refer to two additional taxa in this review,namely Bactrocera occipitalis Drew & Hancock and Bactrocerakandiensis Drew & Hancock. Bactrocera occipitalis occurs inthe Philippines and is sympatric with B. philippinensis, whileB. kandiensis is endemic to Sri Lanka and sympatric withB. invadens. Both are members of the B. dorsalis complex, arepest species in their own right, and are morphologically similarto B. dorsalis (Drew & Hancock, 1994; Clarke et al., 2005;Khamis et al., 2012). This has resulted in some confusion asto their biological relationship with B. dorsalis, akin to that ofB. papayae, B. philippinensis, and B. invadens. However, as forB. carambolae, they possess subtle yet consistent differences inmorphology (Drew & Hancock, 1994; Iwahashi, 1999a; Drewet al., 2008; Drew & Romig, 2013) and molecular genetics(Boykin et al., 2014; Schutze et al., 2014), sufficient to regardthem as distinct species. They are, therefore, not a focus ofthis review, but occasionally they are closely associated with aparticular data set and are thus discussed.

Evidence for the synonymy of B. papayaewith B. dorsalis

Morphology

Bactrocera papayae and B. philippinensis were separatedfrom B. dorsalis based on the former two taxa having longeraculei (1.77–2.12 mm) than B. dorsalis (1.4–1.6 mm). Thepleural area immediately below the postpronotal lobe wasdescribed as brown (i.e. pale) in B. dorsalis and brown to fus-cous (i.e. dark) in B. papayae (Drew & Hancock, 1994). Bac-trocera papayae and B. philippinensis were, in turn, separatedfrom each other by the shape of scales on the distal end of theeversible membrane of the ovipositor, being long and narrow inB. papayae and short in B. philippinensis (Drew & Hancock,1994). Drew et al. (2008) provided an additional morphologicalcharacter state for the separation of B. papayae and B. philip-pinensis: the latter species had a costal band ‘usually expandinginto a fish-hook barb pattern around apex of R4+5’ (p. 220), butthe authors did note that ‘intraspecific variation in these charac-ters … causes difficulties in successful diagnosis’ (p. 219).

Subsequent studies on external morphology have failed todetect any consistent differences between B. papayae andB. philippinensis (Mahmood, 2004; Drew & Romig, 2013),leading Drew & Romig (2013) to synonymize B. philippinensiswith B. papayae. Geometric morphometric analysis of wingshape variation also failed to detect any species-level variationbetween these two taxa and B. dorsalis (Schutze et al., 2012b;Krosch et al., 2013).

The issue of genitalia length is more complex. Bactrocera dor-salis, B. papayae and B. philippinensis have been differentiatedbased on female aculeus length (Drew & Hancock, 1994) andmale aedeagus length (Drew et al., 2008) [NB: male aedeagusand female aculeus lengths are closely correlated in these taxa(Iwaizumi et al., 1997)]. However, although Mahmood (2004)could separate B. dorsalis from B. papayae and B. philippinensisbased on aculeus length, he was unable to separate the latter two

Latitude (degrees north)20151050

Aed

eagu

s le

ngth

(m

m)

3.2

3.0

2.8

2.6

2.4

2.2

R2 = 0.447

P < 0.0001

Fig. 2. Regression of aedeagus lengths (mm), taken from five Bactro-cera dorsalis s.l. males for each of 14 sample sites, against a latitudinalgradient from northern Thailand to eastern Malaysia. Note that somesites appear to contain fewer than five individuals; this is due to someindividuals having the same length of aedeagus (e.g. the northernmostlocation, which appears to have only three samples). Figure reproducedfrom Krosch et al. (2013) with permission.

species from each other. Further, while Iwahashi (2001) foundsignificant differences in aedeagus length between B. dorsalisand B. papayae populations, there was substantial intraspecificvariation and overlap in aedeagus size range for these taxa (Iwa-hashi, 2000). To resolve these apparent inconsistencies amongprevious studies, a more recent analysis used a structured latitu-dinal transect design running from northern Thailand to southernPeninsular Malaysia (Krosch et al., 2013), the reputed area ofoverlap or abutment of these taxa according to location datarecorded by Drew & Hancock (1994). In this study, Kroschet al. (2013) showed aedeagus length to be a continuous cli-nal variable significantly correlated with latitude (Fig. 2). Thisresult did not disagree with earlier literature, in that northernB. dorsalis had shorter aedeagi than southern B. papayae, butintermediate populations had no obvious disjunction in aedea-gus length that could be attributed to a species break. Rather,there was a continuous and significant linear decrease in aedea-gus length with increasing latitude northwards. We conclude thatgenitalia length variation among populations of B. dorsalis andB. papayae should be regarded as intraspecific variation.

Molecular genetics

A range of genetic tools have been applied to the pest speciesof the B. dorsalis complex since the 1994 revision; the relevantliterature is reviewed in Clarke et al. (2005) and Boykin et al.(2014). There has, however, remained ongoing uncertainty overspecies limits among these taxa because, while some studiesidentified significant genetic differences between and amongtaxa (Muraji & Nakahara, 2002; Naeole & Haymer, 2003),others found none (Yong, 1995; Armstrong & Ball, 2005;Tan et al., 2011, 2013). Importantly, many of these studies



were unable to differentiate population-level from species-levelvariation, because only relatively few individuals were analysedor only a limited number of geographic locations were sampled.This has been addressed by recent studies (see later) thathave incorporated greater numbers of specimens collected fromvirtually the entire geographic range of the target taxa, withresults failing to find evidence for genetic differentiation amongB. dorsalis, B. papayae and B. philippinensis, except at a levelcommensurate with intraspecific variation.

Two independent multi-locus phylogenetic analyses, usingdifferent sets of nuclear and mitochondrial loci, could not dif-ferentiate B. dorsalis, B. philippinensis, and B. papayae fromeach other, even though these tests could separate these taxafrom 12 other Bactrocera species from outside the complex (SanJose et al., 2013), and from other members of the B. dorsaliscomplex, including B. carambolae and B. occipitalis (Boykinet al., 2014). This result led San Jose et al. (2013: 684) to con-clude that B. dorsalis, B. papayae and B. philippinensis (alongwith B. invadens) ‘… probably represent a single geneticallyindistinguishable, phenotypically plastic, pest species that hasspread throughout the world’. In other forms of genetic analy-ses, a population-level study using cox1 sequence data identifiedhaplotypes common to B. dorsalis, B. papayae and B. philip-pinensis, failing to identify genetic structure indicative of twoor more species (Schutze et al., 2012b). Similar results werefound in two independent microsatellite studies on B. dorsalisand B. papayae populations from Thailand and west Malaysia,which also found no evidence of population structure commen-surate with the presence of two species (Krosch et al., 2013;Aketarawong et al., 2014). Rather, both studies revealed onepanmictic population across the Thai/Malay Peninsula with avery weak or absent isolation by distance signal. We concludethat published comprehensive genetic evidence is insufficient tomaintain the separation of B. dorsalis and B. papayae.

Cytogenetics

Mitotic and polytene chromosomes of PhilippineB. philippinensis and Malaysian B. papayae maintained atthe FAO/IAEA Insect Pest Control Laboratory (IPCL) atSeibersdorf, Austria, have been analysed and compared withmitotic chromosomes and polytene chromosome maps pub-lished for B. dorsalis (Zacharopoulou et al., 2011). Bactroceraphilippinensis and B. papayae presented the typical number ofmitotic chromosomes and polytene elements as described forB. dorsalis. The comparative analysis of polytene chromosomestructure and banding pattern failed to identify any differencesthat could be used as species-diagnostic markers. There was noevidence of chromosomal rearrangements, such as inversions,that could support the existence of well-separated species amongthese three entities of the complex (Augustinos et al., in press).

Sexual compatibility

Field-cage mate choice tests revealed that B. dorsalis,B. papayae and B. philippinensis mate randomly with each

other (Medina et al., 1998; Tan, 2000; Schutze et al., 2013).These field-cage tests included a host tree and were carriedout following internationally agreed standards used to deter-mine mating compatibility in support of fruit fly Sterile InsectTechnique (SIT) programmes (FAO/IAEA/USDA, 2003).Assessment of egg hatch rates, survival to adult emergence,sex ratios and hybrid fertility in this study found no evidenceof postzygotic incompatibility to the F1 generation followinghybridization of B. dorsalis, B. papayae and B. philippinensis.

Chemoecology

Males of B. papayae, B. philippinensis and B. dor-salis consume the potent male attractant methyl eugenol(ME) and biotransform it to two oxidized analogues(2-allyl-4,5-dimethoxyphenol and (E)-coniferyl alcohol) forstorage in the rectal gland and subsequent use in courtship inter-actions (Nishida et al., 1988; Tan & Nishida, 1996, 2012; Weeet al., 2002; Tan et al., 2013, in press). This results in a matingadvantage for males that consume ME (Shelly & Dewire, 1994;Shelly et al., 1996; Tan & Nishida, 1996). Perkins et al. (1990)recorded minor differences in the rectal volatiles of B. papayae(then known as ‘Mal B’ prior to its formal description),B. philippinensis (= ‘Phil B’) and B. dorsalis, but neverthelessconcluded that ‘…Mal B, Phil B, and B. dorsalis have similarglandular components’ (Perkins et al., 1990: 2486). Importantly,when referring to these minor differences, Perkins et al. (1990:2486) concluded that ‘the analysis of Phil B is most similarto that of B. dorsalis’, as there were relatively greater differ-ences between B. philippinensis and B. papayae than betweenB. philippinensis and B. dorsalis. This similarity, again withminor differences among these three taxa, was also reportedby Fletcher & Kitching (1995). Considering B. philippinensishas now been synonymized with B. papayae, for which rela-tively greater differences were reported, we cannot maintainB. papayae as distinct from B. dorsalis based on glandularchemistry. We conclude there are insufficient differences inthe pheromone chemistry of B. papayae, B. philippinensisand B. dorsalis to render them different biologicalspecies.

Summary

A large body of multidisciplinary research has failed to detectspecies-level differences among B. dorsalis, B. papayae andB. philippinensis. There is some population-level genetic struc-ture and morphological variation consistent with isolation bydistance for B. dorsalis s.l. in South-east Asia (Schutze et al.,2012b), but a lack of other genetic differentiation, absenceof chemoecological differences, identical banding pattern ofpolytene chromosomes, minimal population-level morpholog-ical differences across the geographic distribution, and noevidence for mating isolation lead us to conclude thatB. dorsalis, B. papayae and B. philippinensis constitute onebiological species.



Fig. 3. Colour variation patterns on scutum of Bactrocera dorsalis in Bangladesh. Letters (A–H) reflect increasing darkness of the scutum frompale-brown (A) to almost completely black (H). Modified and used with permission from Leblanc et al. (2013).

Evidence for the synonymy of B. invadenswith B. dorsalis

Morphology

According to Drew & Romig (2013), diagnostic charactersthat distinguish B. invadens from B. dorsalis are scutum colour,width of postsutural lateral vittae and aedeagus length. Bactro-cera invadens has a scutum colour ranging from pale brownto black, whereas B. dorsalis is predominantly black, withless than 20% of specimens exhibiting a pale-coloured scu-tum. Also, B. invadens has narrower postsutural lateral vittaethan B. dorsalis, and B. invadens males have longer aedeagithan B. dorsalis, averaging 2.84 versus 2.69 mm, respectively(Drew et al., 2008; Drew & Romig, 2013). Difference in aedea-gus length was reported by White (2006), with B. dorsalisaveraging 2.59 mm, compared with 2.84 mm for B. invadens;note, however, that the range of aedeagus lengths overlapped(2.46–2.70 and 2.61–2.96 mm, respectively). Bactrocera dor-salis and B. invadens are different in their abdominal colourpattern (White, 2006).

A recent and comprehensive morphological survey ofB. invadens and B. dorsalis from across their respective ranges(Africa, Indian subcontinent and Eastern Asia) by Schutze et al.(2014) revealed that aedeagus length of B. invadens and B.dorsalis ranged from 2.41 to 2.97 mm and 2.35 to 3.00 mm,respectively, while postsutural vittae width ranged from 0.13to 0.21 mm and 0.15 to 0.23 mm, respectively. Hence, there is

significant overlap in these characters and neither separates thetwo entities based on current taxonomy. Further, Schutze et al.(2014) showed scutum colour to vary from brown to black inall specimens from Africa and the Indian subcontinent, withthose from eastern Asia predominantly black. The full rangeof scutum colour variation (red-brown to black) was foundin populations in which B. dorsalis, but not B. invadens, arerecorded to occur, e.g. Pakistan (Schutze et al., 2014), India(David & Ramani, 2011; Schutze et al., 2014) and Bangladesh(Leblanc et al., 2013) (Fig. 3).

Morphometric analyses of wings and legs of B. invadens andB. dorsalis reveal very small differences between these two taxarelative to other Bactrocera species examined, particularly thosefrom outside the B. dorsalis complex (Khamis et al., 2012).However, Khamis et al. (2012) found B. invadens to be highlysimilar to another member of the complex from Sri Lanka,B. kandiensis. Wing shape analysis of multiple populations ofB. dorsalis and B. invadens from across their native and intro-duced ranges showed a high degree of similarity between B. dor-salis and B. invadens, particularly between African B. invadensand B. dorsalis from the northern Indian subcontinent, andbetween Sri Lankan B. invadens and B. dorsalis from China(Schutze et al., 2014).

While variation in abdominal colour pattern has been reported,there is significant overlap between B. dorsalis and B. invadens.White (2006), for example, noted that the transverse bandon tergum III in B. dorsalis is always narrow and neverbroadly extended along the lateral margin, whereas B. invadens



Table 1. Summary of published molecular phylogenetic studies undertaken on Bactrocera dorsalis and Bactrocera invadens.

Publication Loci Analyses Key result and conclusion

Tan et al. (2011) cox1, 16S, 12S Neighbour-joining B. dorsalis and B. invadens monophyletic.Khamis et al. (2012) cox1 Neighbour-joining B. invadens polyphyletic, emerging in two clades: one

including specimens of B. dorsalis and one includingspecimens of B. kandiensis. Inconclusive.

Frey et al. (2013) cox1 SNP panels Neighbour-joining, maximumlikelihood and Bayesianmethods

No support for a monophyletic B. invadens distinct fromother B. dorsalis s.l. taxa. Recommend taxa besynonymized.

San Jose et al. (2013) cox1, EF-1𝛼, PER Maximum likelihood andBayesian Inference

No support for a monophyletic B. invadens distinct fromother B. dorsalis s.l. taxa. Bactrocera invadens andB. dorsalis probably represent a single species.

Schutze et al. (2014) ITS 1, ITS 2, cox1, nad4 Likelihood and Bayesianinference

No support for a monophyletic B. invadens distinct fromother B. dorsalis s.l. taxa. Recommend B. invadensand B. dorsalis be synonymized.

possesses a transverse band that often extends ‘broadly alongthe lateral margin so that much of the hind marking is black’(White, 2006: 138). Abdominal colour pattern in B. invadensis, however, very variable (White, 2006), with Drew & Romig(2013) noting the transverse band of tergum III for B. invadensto range from narrow to broad. Bactrocera dorsalis similarlyexhibits wide variation, with some forms clearly possessingbroad lateral margins on tergum III. Narrow markings aresimilarly found on terga IV and V in B. dorsalis; yet, again theyare often broad in B. invadens (White, 2006). As for tergumIII, however, B. invadens possesses narrow lateral dark fuscousmarkings on terga IV and V (Drew & Romig, 2013) and someforms of B. dorsalis can possess broader markings on these terga(fig. 60B, Drew & Romig, 2013). Abdominal colour pattern ishighly variable, with significant overlap between B. dorsalis andB. invadens that cannot discriminate between them or be used asevidence to maintain them as distinct species.

Molecular phylogenetics

Several molecular phylogenetic studies have evaluated therelationship between B. invadens and B. dorsalis (Tan et al.,2011; Khamis et al., 2012; Frey et al., 2013; San Jose et al.,2013; Schutze et al., 2014), the operational details of whichare provided in Table 1. Frey et al. (2013) could not resolveB. invadens from B. dorsalis and recommended their syn-onymy. The multi-locus studies by San Jose et al. (2013) andSchutze et al. (2014) found B. invadens to be polyphyletic withother members of the B. dorsalis complex and both studiesconsidered them genetically indistinguishable. Khamis et al.(2012) showed that B. invadens split into two clades fol-lowing analysis of the cox1 barcoding gene region, with SriLankan and African specimens grouping with B. dorsalis andsome individuals of B. invadens from Sri Lanka grouping withB. kandiensis. Like Khamis et al. (2012), Schutze et al. (2014)incorporated B. kandiensis in their study; however, B. kandi-ensis resolved as a well-supported clade emerging from thebroader B. dorsalis/invadens group and without any B. invadensindividuals.

Cytogenetics

Mitotic and polytene chromosomes of Kenyan B. invadenshave been compared with mitotic chromosomes and polytenechromosome maps published for B. dorsalis (Zacharopoulouet al., 2011). Bactrocera invadens presented the typical num-ber of mitotic chromosomes and polytene elements as describedfor B. dorsalis. Comparative analysis of polytene chromosomestructure and banding pattern failed to identify any differencesthat could be used as species-diagnostic markers. The lack ofchromosomal differences was further supported by the analy-sis of polytene chromosomes of F1 bidirectional hybrids amongB. dorsalis (Thailand origin) and B. invadens (Kenyan origin).The lack of detectable asynapses among the homologous chro-mosomes of these two entities supports the hypothesis that thesetwo entities do not represent different species (Augustinos et al.,in press).


Field-cage mating tests and subsequent postzygotic compati-bility analysis of hybrid viability, survival, fertility and sex ratiosdemonstrated that B. invadens from Kenya mated randomly withB. dorsalis from Pakistan and China under semi-natural condi-tions, producing fully viable offspring to the hybrid F2 genera-tion (Bo et al., 2014).

Chemoecology

As is the case for B. papayae and B. philippinensis,male B. invadens flies respond to and consume ME andbiosynthesize the same rectal gland pheromone constituents(2-allyl-4,5-dimethoxyphenol and (E)-coniferyl alcohol) asB. dorsalis (Tan et al., 2011, in press; Tan & Nishida, 2012).

Summary

Similar to B. papayae and B. philippinensis, a consider-able body of evidence spanning morphological, molecular,



Table 2. Arguments for maintaining the separation of Bactrocera dorsalis, Bactrocera papayae and Bactrocera invadens as provided in Drew &Romig (2013) and counter-evidence as detailed in the main text.

Arguments against synonymization(Drew & Romig, 2013) Counter-arguments

Male aedeagus and female aculeus length are significantlydifferent between B. papayae (mean length 3.00 mm) andB. dorsalis (mean length 2.69 mm) (original data fromDrew et al., 2008).

Genitalia lengths of these two species overlap (Iwahashi, 2001), aresignficantly variable among intraspecific populations (Mahmood, 1999;Drew et al., 2008), and vary clinally with geography (Krosch et al.,2013).

Wing shape variation allows the accurate separation ofB. dorsalis from B. papayae (original data from Schutzeet al. (2012a)).

Schutze et al. (2012a) concluded that ‘[these wing shape data] may supportthem [B. papayae and B. dorsalis] being considered conspecific’.Subsequent data and analyses found that wing shape between B. dorsalisand B. papayae (and B. philippinensis) varies along a geographic cline(Krosch et al., 2013) and is strongly correlated with regionalbiogeography (Schutze et al., 2012b).

B. invadens can be separated from B. dorsalis based onscutum colour range, aedeagus length and postsuturallateral vittae width.

Scutum colour varies from red-brown to black in B. invadens (Drew et al.,2005) and B. dorsalis (LeBlanc et al., 2013; Schutze et al., 2014).Aedeagus length and postsutural lateral vittae width are highly variableand overlap between populations of the two taxa (Schutze et al., 2014).

Two mitochondrial DNA genes (cox1 and nad5) renderB. invadens as markedly different from B. dorsalis(unpublished data referred to in Drew & Romig, 2013).

Published comprehensive molecular studies have failed to resolve B.invadens as distinct from B. dorsalis (Tan et al., 2011; Khamis et al.,2012; Frey et al., 2013; San Jose et al., 2013; Schutze et al., 2014). cox1was used in all the phylogenetic studies published to date, and in none ofthese are B. dorsalis and B. invadens monophyletic for this locus.

cytogenetic, sexual compatibility and chemoecological datanow exist to reject the current taxonomy of B. invadens asa species distinct from B. dorsalis. Existing morphologicaldiagnostic characters, particularly aedeagus length and post-sutural lateral vittae, vary continuously across B. dorsalis andB. invadens geographic distributions. Scutum colour is highlyvariable across Africa and the Indian subcontinent but becomespredominantly black in populations of East and South-east Asia.While the underlying mechanisms for this intraspecific vari-ation are unknown, preliminary experiments by the authors(M.K. Schutze, W. Bo and C. Caceres, unpublished data) onPakistan-origin B. dorsalis with variable scutum colours at theFAO-IAEA Insect Pest Control Laboratory (Seibersdorf) haverevealed scutum colour to be a simple heritable trait. More-over, in a published niche overlap study, Hill & Terblanche(2014) combined new data with those available for B. invadens(De Meyer et al., 2010) and B. dorsalis (Stephens et al., 2007),revealing highly overlapping niches for these two taxa (includ-ing B. papayae and B. philippinensis), leading the authors toconclude that these taxa probably constitute the same biologicalspecies. In their recent treatment of B. papayae, B. invadens andB. dorsalis, Drew & Romig (2013) maintain their status as dis-tinct species. Given the economic implications of the taxonomicstatus of these species, we address the specific points made byDrew & Romig (2013) in Table 2.

Evidence for the maintenance of B. carambolae

Bactrocera carambolae has been included in studies examiningpest members of the B. dorsalis complex due to its morpho-logical similarity to B. dorsalis and its sympatric distributionwith B. papayae (Drew & Hancock, 1994; Drew & Romig,2013). Whereas much of this research effort has been directed at

identifying diagnostic characters to distinguish B. carambolaefrom other dorsalis-complex species, recent work has increasedattention on the biological relationship of B. carambolae withother members of the complex. Reliable, yet subtle, differ-ences between B. carambolae and B. dorsalis have been foundbased on morphology, molecular genetics, chemoecology andbehaviour.

Morphology

Drew & Romig (2013) remark that B. carambolae is dis-tinguished from B. papayae by a broad medial longitudinalblack band on abdominal terga III–V, a broader costal bandapically and shorter male aedeagi and female ovipositor. Addi-tionally, the anterolateral corners of terga IV are fuscous toblack and distinctly rectangular in shape in B. carambolae,whereas the fuscous to black corners of terga IV of B. dor-salis and B. papayae are narrow (Drew & Romig, 2013). Thereis, however, considerable variation in abdominal colour pat-tern that does not directly correlate with aedeagus length orcostal band width (Iwahashi, 1999b). For instance, individu-als bearing abdominal colour patterns typical of B. dorsalis/B.papayae (i.e. narrow markings on terga IV) possess relativelylong aedeagi (2.66–3.24 mm), whereas those with a B. caram-bolae pattern (i.e. rectangular markings on terga IV) possessboth long and short aedeagi (2.37–2.68 mm; note the overlapwith long aedeagi) (Iwahashi, 1999b). However, apical costalband width closely correlates with aedeagus length, in thatindividuals with apically broad costal bands have consistentlyshort aedeagi, and those with distinctly narrow bands havelong aedeagi (Iwahashi, 1999b). Taken together, these data ledIwahashi (1999b) to conclude that B. carambolae can be



distinguished from B. papayae by shorter aedeagi (albeit withsmall overlap in length) when present with a broader apicalexpansion of the costal band, but that terga IV markings, whichare rectangular in B. carambolae, ranged from narrow to rect-angular in B. papayae. Although aedeagus length is an unreli-able taxonomic character, due to its variability over a wide geo-graphic distribution, the data of Iwahashi (1999b) were recordedfrom individuals collected from the single locality of Singapore.The range of aedeagus length at this location was very wide(2.37–3.24 mm) and similar to that observed for B. dorsaliss.l. from across a large latitudinal gradient from PeninsularMalaysia to northern Thailand (see Fig. 2). Such a wide rangein aedeagus length at a single location, with both short andlong aedeagi correlating with additional morphological char-acters (e.g. apical costal band width), strongly supports theexistence of two species, i.e. B. carambolae and B. dorsalis.

Molecular genetics

Despite B. carambolae being genetically similar to B. dor-salis, differences between these species exist for nuclear andmitochondrial DNA data. Armstrong & Cameron (2000), forinstance, examined rDNA ITS1 restriction sites and showedthat B. carambolae could be clearly separated from B. dor-salis [based on a ∼50 base pair (bp) insertion/deletion (indel)],whereas B. dorsalis remained indistinguishable from B. papayaeor B. philippinensis. In a later phylogenetic study of the cox1barcode region, Armstrong & Ball (2005) found that B. caram-bolae formed a discrete but weakly supported group. Basedon multi-locus phylogenetic analyses of mitochondrial DNAand nuclear DNA, Boykin et al. (2014) and Schutze et al.(2014) found that B. carambolae formed a reciprocally mono-phyletic sister group to a large clade consisting of B. dorsalis,B. papayae, B. invadens and B. philippinensis sampled fromalmost their entire native and invasive geographic distributions.Notably, a 44-bp indel in ITS1 is the key diagnostic feature thatconsistently differentiates B. carambolae from B. dorsalis, B.papayae, B. philippinensis and B. invadens; this was confirmedfollowing screening of B. carambolae individuals from boththeir native distribution in South-east Asia and their invasiverange of Suriname, South America (Boykin et al., 2014).Although the recent multi-locus phylogenetic study of San Joseet al. (2013) showed B. dorsalis, B. papayae, B. philippinensis,B. invadens and B. carambolae to be polyphyletic, six out ofeight B. carambolae specimens (all from Malaysia) neverthe-less formed a weakly supported monophyletic clade sister toremaining B. dorsalis s.l. specimens examined.

Cytogenetics

Differences in metaphase karyotype banding patterns betweenB. carambolae and B. dorsalis have been reported in B. caram-bolae larvae from Thailand (Baimai et al., 1999). Such differ-ences between groups in mitotic chromosome heterochromatinpatterns are considered indicative of cryptic species, as shownfor other dipteran taxa, including other tephritids, vinegar flies

and mosquitoes (Baimai, 1998). However, larvae examined byBaimai et al. (1999) were identified as B. carambolae based onthe subsequent adult external morphology of other larvae thatemerged from the same infested fruits; the identification of cyto-genetically examined larvae as B. carambolae was thus inferred.Subsequent analysis of cultured B. carambolae, originally fromSuriname, revealed that mitotic and polytene chromosomes ofB. carambolae present the typical number of mitotic chromo-somes and polytene elements as described for B. dorsalis, whilecomparative analyses of polytene chromosome structure andbanding patterns failed to identify any differences that canbe used as species-diagnostic markers (Augustinos et al., inpress). The available cytogenetic data are thus inconclusive todifferentiate B. carambolae from B. dorsalis.


One of the principal arguments used against sexual compat-ibility tests made under semi-natural conditions is that suchexperiments are too far removed from the wild to reflect biolog-ical reality, and that ‘… little assistance can be provided frominterbreeding experiments as they are unrelated to the behaviourof populations under field conditions’ (Drew & Romig, 2013:26). We argue that despite the inherent caveats of such tests,systematically executed and analysed mating and postzygoticcompatibility assessments (FAO/IAEA/USDA, 2003) provideextremely valuable insights for addressing species delimita-tion studies in tephritids, particularly when incorporated withadditional independent data (Clarke & Schutze, in press).Results obtained from mating tests made under semi-naturalconditions between B. carambolae and B. dorsalis s.l. areevidence of this, as demonstrated by Schutze et al. (2013) forwhich assortative mating occurred between these taxa. Besidesthe incompatibilities observed, a closer examination revealedbehavioural differences between B. carambolae and B. dorsalis,B. papayae and B. philippinensis, particularly with respect tomating location (Schutze et al., 2013). Behavioural variationwas not observed when comparing among B. dorsalis, B.papayae and B. philippinensis, in that there were no significantdifferences in time of sexual activity or position in the fieldcage of mating couples. These results mirrored the earlierfield cage study carried out for SIT purposes by McInnis et al.(1999), who showed reduced mating compatibility between B.dorsalis from Hawaii and B. carambolae from Suriname whencarried out under semi-natural conditions, in which only 7.8%of all couples resulted from interspecific crosses. Preliminarymating compatibility studies of B. dorsalis crossed with B.kandiensis have shown evidence of assortative mating similarto that observed for B. carambolae (M.K. Schutze, W. Bo andC. Caceres, unpublished data).

Chemoecology

Unlike B. papayae, B. invadens and B. philippinensis,B. carambolae males possess a pheromone blend that is consid-erably different from B. dorsalis s.l. (i.e. B. dorsalis, B. papayae,



B. philippinensis and B. invadens). Perkins et al. (1990) reportedthat B. carambolae has a glandular chemistry that is distinctfrom that of B. dorsalis, B. papayae and B. philippinensis.A similar result was found by Fletcher & Kitching (1995),which led them to conclude that B. carambolae and B. papayaerepresented distinct species. Male rectal gland compoundsfollowing ME feeding differ between B. carambolae andB. dorsalis, B. papayae and B. philippinensis. As mentionedearlier, these species accumulate 2-allyl-4,5-dimethoxyphenoland (E)-coniferyl alcohol, whereas B. carambolae males accu-mulate only (E)-coniferyl alcohol (Tan & Nishida, 1996; Wee& Tan, 2007). Each species has different sensitivities to, andconsumption rates of, ME (Wee et al., 2002; Tan et al., inpress). These differences are reflected in operational manage-ment programmes, as, while a distribution rate of four MEfibre blocks per hectare is typically used for B. dorsalis maleannihilation purposes, a several-fold higher rate per hectare wasneeded to eradicate B. carambolae from an area in Suriname(van Sauers-Muller, 2008).

Host use

While we acknowledge the limitations of using host dataas evidence for species delimitation, host range studies ofB. carambolae and B. dorsalis/B. papayae nevertheless revealdistinct differences. This is particularly the case for banana(Musa sp.), which is a well documented host for B. dorsalisand B. papayae (Clarke et al., 2005) but not for B. caram-bolae. Two independent and comprehensive host fruit surveysconducted in the native and introduced ranges of B. carambo-lae (Thai-Malay Peninsula and Suriname, respectively) foundthat, while South-east Asian Musa sp. yielded B. papayae, noB. carambolae emerged from banana in either South-east Asiaor Suriname despite large quantities of fruit being sampled(Clarke et al., 2001; van Sauers-Muller, 2005).

Summary

We consider this multidisciplinary evidence sufficient formaintaining B. carambolae as a discrete taxonomic entity.Although very closely related to B. dorsalis, B. carambolae hasbeen shown by multiple independent studies, across a range ofdisciplines, to differ consistently from B. dorsalis, B. papayae,B. invadens and B. philippinensis. Furthermore, these studiesprovide diagnostic features by which B. carambolae can bereliably distinguished from B. dorsalis.

Taxonomic changes

Based on the evidence discussed earlier (as summarized inTable 3), we propose the synonymy of B. papayae Drew& Hancock syn.n. and B. invadens Drew, Tsuruta & Whitesyn.n. under the senior name B. dorsalis (Hendel). Bactroceraphilippinensis has recently been synonymized with B. papayae

(Drew & Romig, 2013), and therefore B. philippinensis alsobecomes a new synonym of B. dorsalis. Bactrocera carambolae,while very closely related to B. dorsalis, exhibits a number ofconsistent differences from B. dorsalis and we consider it a validspecies.

Drew & Romig (2013) argue that present-day B. invadensis probably the same species as Fabricius’s Musca ferruginea,inferring that they represent a separate species from B. dorsalis.This is inconsistent, as these authors maintain M. ferruginea asa junior synonym of B. dorsalis. Our proposal to synonymizeB. invadens with B. dorsalis removes the circular ambiguity ofDrew & Romig’s argument and acknowledges the considerableevidence that these taxa represent the same biological species.

In the following we provide a revised description ofB. dorsalis that incorporates the morphological variationexhibited by B. papayae and B. invadens with reference tothe original species descriptions and subsequent taxonomictreatments (Drew & Hancock, 1994; Drew et al., 2005; Drew& Romig, 2013) in addition to data obtained from studies listedearlier.

Abbreviations for museum collections are as follows: BMNH,The Natural History Museum, London, U.K.; NMKE, NationalMuseum of Kenya, Nairobi, Kenya; MRAC, Royal Museum ofCentral Africa, Tervuren, Belgium.

Bactrocera (Bactrocera) dorsalis (Hendel)Musca ferruginea – Fabricius, 1794: 342. Preoccupied by

Musca ferruginea Scopoli, 1763 (see Hardy, 1969).Dacus ferrugineus – Fabricius, 1805: 274.Dacus dorsalis – Hendel, 1912: 18. Lectotype ♀ in BMNH.Bactrocera ferruginea – Bezzi, 1913: 95.Chaetodacus ferrugineus var. dorsalis – Hendel, 1915: 426.Chaetodacus ferrugineus – Bezzi, 1916: 104.Chaetodacus ferrugineus dorsalis – Bezzi, 1916: 104.Chaetodacus ferrugineus var. okinawanus – Shiraki, 1933: 62;

Hardy & Adachi, 1956: 8; Hardy, 1969: 402 [synonym].Dacus (Strumeta) dorsalis – Hardy & Adachi, 1956, 7; Hardy,

1969: 395; 1973: 41; 1974: 29.Strumeta dorsalis – Hering, 1956: 63.Strumeta ferruginea – Hering, 1956: 63.Strumeta dorsalis okinawana – Shiraki, 1968: 23.Dacus (Bactrocera) dorsalis – Hardy, 1977: 49; Drew, 1982:

60; Drew, 1989: 63.Dacus (Bactrocera) semifemoralis – Tseng, Chen & Chu, 1992:

46 (synonymized by Drew & Romig, 2013: 76).Dacus (Bactrocera) yilanensis – Tseng, Chen & Chi, 1992: 52

(synonymized by Drew & Romig, 2013: 76).Bactrocera (Bactrocera) dorsalis – Drew & Hancock, 1994: 17,

Lectotype designation; Norrbom et al., 1998: 90; Mahmood& Hasan, 2005: 8; White, 2006: 137; Drew et al., 2007: 3.

Bactrocera (Bactrocera) variabilis – Lin & Wang, in Lin et al.,2011, 896. Holotype in HEIQ (synonymized by Drew &Romig, 2013: 76).

Bactrocera (Bactrocera) philippinensis – Drew & Hancock,1994: 52 (synonymized with B. papayae by Drew & Romig,2013: 142).



Table 3. Summary of multidisciplinary evidence in support of the synonymization of Bactrocera papayae and Bactrocera invadens with Bactroceradorsalis, and for the maintenance of Bactrocera carambolae as a separate species.

B. dorsalis compared with:

Character B. papayae/philippinensis B. invadens B. carambolae

Morphology Aedeagus length Clinal variation Clinal variation Sympatric populations

Wing shape Clinal variation Clinal variation Sympatric populations

emaSnoitairavlanilCemaSeattivlaretallarutustsoPThoracic colour pattern Variable with overlap;

often indistinguishableVariable with overlap;

often indistinguishableVariable with overlap; often

indistinguishableAbdominal colour pattern Variable with overlap;

often indistinguishableVariable with overlap;

often indistinguishableB. carambolae with

consistent typical pattern

Molecular genetics Multi-locus phylogeny Same monophyletic clade Same monophyletic clade Reciprocally monophyleticcox1 haplotype network

(pop genetics)Shared haplotypes Shared haplotypes No shared haplotypes

Microsatellite(pop genetics)

Cytogenetics Polytene chromosomes Identical Identical Identical

Chemoecology Rectal gland extracts aftermethyl eugenol feeding

Sexual compatibility Prezygotic compatibility Fully random mating Fully random mating Nonrandom assortativemating

Postzygotic compatibility Hybrids viable and fertile Hybrids viable and fertile Reduced hybrid viabilityIdentical Identical Different

Panmictic Not compared Not compared

Aculeus spicules Variable with overlap Not compared Variable with overlap

Bactrocera philippinensis is included, yet note that this species has recently been synonymized with B. papayae (Drew & Romig, 2013). Grey – identical,highly variable with overlap, or clinal, and in accordance with intraspecific variation; black – consistent differences in accordance with interspecificvariation.

Bactrocera (Bactrocera) papayae – Drew & Hancock, 1994:48, syn.n.

Bactrocera (Bactrocera) invadens - Drew, Tsuruta & White,2005: 149, syn.n.

Diagnosis. First flagellomere shorter than ptilinal fissure,face with a black spot in each antennal furrow. Tomentumpattern without longitudinal gap in the middle of prescutum.Scutum colour (other than vittae) red-brown to black. Lateralvittae of scutum present, parallel-sided, yellow, ending at orjust behind intra-alar seta. Medial vitta of scutum absent.Scutellum largely yellow. Setae: anterior supra-alar present,prescutellar present, scutellar one pair. Anepisternal stripe notextended forward. Wing costal band width from vein subcostalto slightly below vein R4+5 at wing apex; confluent with veinR2+3 in depth. Cell basal costal without microtrichia. Cell costalwith microtrichia in the anterodistal corner only. Cell basalradial with microtrichae at the base. Wing length: 5.4–6.9 mm.Foretibia pale fuscous to dark fuscous, mid-tibia pale fuscousto fuscous, hind tibia fuscous to dark fuscous. Femora largelyfulvous. Abdominal terga free, except I and II. Terga markings:terga III–V with medial longitudinal black band, tergum III witha basal narrow transverse black band, terga IV and V with blacktriangular lateral markings, sometimes longitudinal black bandson lateral margins and sometimes without any mark. TergumIII (males) with pecten. Sternum V (males) with V-shaped

notch. Posterior surstylus lobe short. Males attracted to methyleugenol.

Note: Confident morphological diagnosis between B. dorsalisand other members of the B. dorsalis complex remains problem-atic. Many previously reported characters are subjective (e.g.‘broad’ vs ‘narrow’) or have since been demonstrated to varyover geographical clines and to bear no relationship to actualspecies limits (e.g. genitalic morphometrics). Specific charactersdistinguishing B. dorsalis from other members of the complexare lacking and a consensus of evidence should be consideredwhen differentiating this species.

Redescription. Male. Head. Vertical length 1.57–1.88 mm.Frons, of even width, length 1.36–1.57× breadth; red-brown tofulvous, with fuscous to dark-fuscous around frontal and orbitalsetae; anteromedial hump pale fuscous to fuscous, with shortpale hairs; orbital setae dark fuscous to black: one superiororbital, two inferior orbital; lunule red-brown to fuscous. Ocellartriangle black. Vertex pale fuscous to fuscous. Face fulvouswith medium- to large-sized black spot in each antennal furrow;length 0.4–0.53 mm. Gena fulvous, with brown subocular spotand red-brown to black seta present. Occiput red-brown to black,yellow to fulvous along eye margins; occipital row with two toeight dark to black postocular setae. Antenna, scape and pedicelfulvous to red-brown, first flagellomere fulvous to red-brownwith fuscous on apex and outer surface; a strong red-brown to



dark dorsal seta on scape; arista black (fulvous basally); lengthof segments: 0.13–0.22, 0.31–0.35, 0.55–0.90 mm.

Thorax. Scutum colour variable red-brown to black, withlanceolate dark patterning in intermediate forms. When black,brown to orange-brown below and behind lateral postsuturalvittae, around transverse suture, between postpronotal lobesand notopleura, inside postpronotal lobes. Pleural area dark fus-cous to black with red-brown below postpronotal lobe. Yellowmarkings as follows: postpronotal lobe; notopleural cal-lus; anepisternal stripe reaching midway between anteriormargin of notopleural callus and anterior notopleural seta dor-sally, continuing to katepisternum as a transverse spot, anteriormargin straight to convex; anatergite (posterior apex black);anterior 55–75% katatergite (remainder black); two narrow tobroad parallel-sided lateral postsutural vittae ending at or justbehind intra-alar seta; scutellum yellow with a black basal band.Postnotum black. Setae: apical scutellar; prescutellar acros-tichal; intra-alar; anterior and posterior supra-alar; enepisternal;anterior and posterior notopleural; and two pairs scapular.

Legs. Femora entirely fulvous, except for small elongate darkfuscous spot on outer apical surface of fore femur of somespecimens; foretibia pale fuscous to dark fuscous, mid-tibia palefuscous to fuscous, hind tibia fuscous to dark fuscous; mid-tibiaeeach with an apical black spur; tarsi fulvous.

Wing. Length 5.4–6.9 mm; cells basal costal and costal colour-less; microtrichia in anterodistal corner of cell c only; remainderof wing colourless except fuscous cell subcostal, narrow fus-cous costal band confluent with R2+3 and remaining very nar-row or widening slightly if it overlaps this vein, to end justbeyond apex of R4+5 (in some specimens there is an expansionaround extremity of R4+5, which may be slight or expanding intoa hook-like pattern), a narrow pale fuscous anal streak endingbefore the wing margin; with or without dense aggregation ofmicrotrichia around A1 +CuA2; supernumerary lobe of weak tomedium development.

Abdomen. Oval; terga free except I and II; tergum III withpecten. Tergum I, sterna I and II wider than long. Abdominalcolour pattern variable: tergum I from orange-brown withpale fuscous laterally, to narrow black lateral margins, toentirely dark fuscous; tergum II orange-brown with narrowbasal transverse black band, which may reach dark fuscousto black lateral margins; terga III–V orange-brown with black‘T’-shaped pattern, comprising a transverse black band acrossanterior margin of tergite III, which may be shallow or deep andin some individuals reaches the lateral margins where it mayexpand to leave only a small postero-submedial orange-brownarea, and a medial longitudinal dark fuscous to black bandover terga III–V. Terga IV and V with narrow fuscous lateralmargins or triangular (narrow or broad) black markings onanterolateral corners that may extend posteriorly along thelateral margin. A pair of ceromata on tergum V. Abdominalsterna dark.

Female. As for male, except no dense aggregation ofmicrotrichia around A1 +CuA2; supernumerary lobe weak; nopecten present on abdominal tergite III. Oviscape orange-brown,may tend fuscous apically, dorsoventrally flattened and taper-ing posteriorly in dorsal view; ratio of length of oviscape to

tergum V 0.7 –1.2:1. Aculeus needle-shaped with four pairs ofsubapical setae.

Illustrations. Illustrations of what we now consider B. dor-salis are provided in Drew & Hancock (1994) (their figures 19,63 and 70), Drew et al. (2005 (their figures 1 and 4) and Drew& Romig (2013 (their figures 60, 87 and 136). The variationbetween these figures encompasses the variation we recognizewithin B. dorsalis as defined here.

Material examined. Bactrocera (Bactrocera) dorsalisLECTOTYPE, ♀, TAIWAN (FORMOSA): Koshun, 9.v.08(BMNH); PARALECTOTYPES, 3♀, 10♂, same data as lec-totype (BMNH); PARALECTOTYPE, 1♂, Tainan, v.1912(BMNH) (designated by Drew & Hancock, 1994).

Bactrocera (Bactrocera) invadens HOLOTYPE, ♂, KENYA:Coast, Matuga, 12.iii.2003, ICIPE sample T1, methyl eugenoltrap (NMKE); PARATYPES, 4♂, KENYA: Matuga, 12.iii.2003(MRAC); PARATYPES, 5♀, 2♂, CAMEROON: Essé,23.viii.2004 (MRAC); PARATYPES, 3♀, 2♂, TANZANIA:Morogoro, xii.2003 (MRAC); PARATYPES, 1♂, BENIN:Cotonou (IITA Station), 16.ix.2004 (MRAC): PARATYPES,1♂, SENEGAL: Keur Moussa, 8.vi.2004 (MRAC).

Bactrocera (Bactrocera) papayae HOLOTYPE, ♀, PENIN-SULAR MALAYSIA: Kuala Kangsar, 20.ix.1988 (BMNH).

Lists of specimens examined as part of this revision areprovided in earlier publications (Schutze et al., 2012a,b, 2014;Krosch et al., 2013; Boykin et al., 2014). These were collectedfrom the entire geographic range of B. dorsalis, including alltype localities of B. dorsalis (Taiwan), B. papayae (PeninsularMalaysia) and B. invadens (Kenya). Many thousands of otherspecimens across multiple depositories and identified as B.dorsalis, B. papayae and B. invadens have also been examinedby several authors.

Distribution. The synonymization of B. invadens andB. papayae with B. dorsalis considerably expands the knowndistribution of B. dorsalis. Hence, the distribution now extendsthroughout much of sub-Saharan Africa, across the Indian sub-continent to China, throughout the South-east Asian Indo/MalayArchipelago, and as far east as New Guinea, the islands of theSouth Pacific and Hawaii, into the Philippines and Palau. Bac-trocera dorsalis records from the Andaman and Nicobar Islandsin the Indian Ocean are unconfirmed: Drew & Romig (2013) listB. dorsalis from Andaman Island, but it is not in the faunal listsprovided by Ranganath & Veenakumari (1995a,b) and David &Ramani (2011).

Conclusions

This review of research data represents an example of com-prehensive integrative taxonomy (sensu Schlick-Steiner et al.,2010; Yeates et al., 2011). As previously noted (e.g. Kroschet al., 2013), laboratories from countries spanning Africa,Asia, Australasia, Europe and North America have directed



independent multidisciplinary research efforts toward the res-olution of B. dorsalis and closely related pest taxa over thepast 20 years. Crucially, while such research has been ongo-ing for this time, a coordinated research project establishedby the UN-FAO/IAEA has intensified research focus over thelast 5 years and fostered a more systematic and integrativeapproach within a significantly improved collaborative environ-ment. Each research group has applied specific expertise to inde-pendently address questions pertaining to the biological relation-ships among B. dorsalis and related species, and virtually alllines of evidence have drawn the same conclusion: that there isinsufficient evidence to maintain B. invadens, B. papayae andB. philippinensis as biological species distinct from B. dor-salis. This integrative research has further revealed complex-ities regarding intraspecific variation in B. dorsalis over itsgeographic distribution. In particular, this approach has shownthat the greatest variability is in the Indian subcontinent,where the genus Bactrocera as a whole is considered to haveoriginated and greatly diversified following its collision withEurasia some 35–50 Ma (Krosch et al., 2012). Further, as bio-logical data relevant to pest management have been publishedwhile these taxa existed under different names – particularly forB. invadens, for which considerable biological data have beengenerated recently – there is now a need to compile and recon-cile this literature under a revised understanding that these taxarepresent a single biological species.

The situation in India and Sri Lanka warrants particularattention to clarify relationships among the following species:(i) B. kandiensis; (ii) B. dorsalis; (iii) populations previouslyreferred to as B. invadens; and (iv) other B. dorsalis complexspecies. The study of Schutze et al. (2014) found that allB. dorsalis material examined from the subcontinent is genet-ically similar to B. dorsalis from Asia and B. invadens fromAfrica, while B. kandiensis emerged as a distinct clade but withevidence of introgression with B. dorsalis. Khamis et al. (2012)found that B. invadens splits into two clades, one mixed withB. dorsalis and the other mixed with B. kandiensis. This regionis clearly fertile ground for future research into the B. dorsaliscomplex.

The applied impacts of these name changes are significant,covering aspects of trade, quarantine and field control; examplesfor each of these follow.

Trade. Currently, the distributions of B. dorsalis complexspecies dealt with in this paper are almost entirely disjunct,with most countries having only one of the species (Thai-land is unique in having three – B. dorsalis, B. papayae andB. carambolae). Under the rules of the International Plant Pro-tection Convention, a country with one of the species but notanother has a sovereign right to impose risk reduction treat-ments on commodity imports from a country with another pestmember of the B. dorsalis complex species; this reality sever-ally restricts trade for many nations. Recognition that B. dor-salis, B. invadens, B. papayae and B. philippinensis are onespecies will ease restrictions to fresh commodity trade betweencountries where these are native or nonregulated invasive taxa.

Quarantine. A key element of quarantine is risk assessment.As demonstrated by Hill & Terblanche (2014), basic quarantine

issues such as enhanced predictive power to determine thelikely spread, and ultimate distribution, of invasive membersof the B. dorsalis complex are greatly improved by accuratelyrecognizing species boundaries.

Pest management. Many pest management tools will beimproved by resolving the species limits, of which applicationof the SIT is one. The SIT requires mass release of sterilizedmale flies to mate with, and so make infertile, wild conspecificfemales. A great body of SIT knowledge exists for B. dorsalis(see chapters in Dyck et al., 2005), but much less so for theother members of the complex. The recognition of conspecificityamong key pest taxa within the complex will allow B. dorsalisSIT to be applied in new countires and regions (Aketarawonget al., 2014).

Clarke et al. (2005) identified the B. dorsalis complex as notonly of great economic importance, but also a group with acomplex evolutionary history worthy of deeper study. We hopethat, following this synonymization, attention can now focus onoutstanding issues of economic importance and on elucidatingthe nature of intraspecific variation to further understand theevolutionary biology of B. dorsalis and related taxa.

Acknowledgements

M.K.S., A.R.C., K.F.A., T.A.C., A.C., A.E., D.H. and M.N.K.were supported entirely or in part through CRC National PlantBiosecurity and Plant Biosecurity CRC projects CRC20115,CRC20183, PBCRC3024 and PBCRC3066. These authorsacknowledge the support of the Australian Government’sCooperative Research Centres Program. S.L.C. was sup-ported by an Australian Research Council Future Fellowship(FT120100746). L.L. and M.S.J. were supported through aCooperative Agreement from the United States Department ofAgriculture through Farm Bill funding (project 3.0251) admin-istered by the College of Tropical Agriculture and HumanResources of the University of Hawaii. K.F.A. and L.M.B.were supported by the Tertiary Education Commission of NewZealand. The authors sincerely acknowledge support providedby the UN-FAO/IAEA via the Coordinated Research Project‘Resolution of cryptic species complexes of tephritid peststo overcome constraints to SIT application and internationaltrade’. We thank Bruce Halliday of the CSIRO for nomen-clatural advice. The authors thank the numerous anonymousreviewers for improving the manuscript.

References

Aketarawong, N., Bonizzoni, M., Thanaphum, S. et al. (2007) Inferenceon the population structure and colonization process of the invasiveoriental fruit fly Bactrocera dorsalis (Hendel). Molecular Ecology,16, 3522–3532.

Aketarawong, N., Isasawin, S. & Thanaphum, S. (2014) Evidence ofweak genetic structure and recent gene flow between Bactroceradorsalis s.s. and B. papayae, across Southern Thailand and WestMalaysia, supporting a single target pest for SIT applications. BMCGenetics, 15, 40 [WWW document]. URL http://www.biomedcentral.com/1471-2156/15/70 [accessed on 16 July 2014].



Armstrong, K.F. & Ball, S.L. (2005) DNA barcodes for biosecurity:invasive species identification. Philosophical Transactions of theRoyal Society of London, Series B, 360, 1813–1823.

Armstrong, K. & Cameron, C. (2000) Species identification of tephritidsacross a broad taxonomic range using ribosomal DNA. Area-wideControl of Fruit Flies and Other Insect Pests (ed. by K.H. Tan), pp.703–710. Penerbit Universiti Sains Malaysia, Pulau Pinang.

Augustinos, A.A., Drosopoulou, E., Gariou-Papalexiou, A. et al. (inpress) The Bactrocera dorsalis species complex: comparative cyto-genetic analysis in support of Sterile Insect Technique applications.BMC Genetics.

Baimai, V. (1998) Heterochromatin accumulation and karyotypic evolu-tion in some dipteran insects. Zoological Studies, 32, 75–88.

Baimai, V., Phinchongsakuldit, J. & Trinachartvanit, W. (1999)Metaphase karyotypes of fruit flies in Thailand (III): six members ofthe Bactrocera dorsalis complex. Zoological Studies, 38, 110–118.

Bezzi, M. (1913) Indian trypaneids (fruit flies) in the collection of theIndian Museum, Calcutta. Memoirs of the Indian Museum, 3, 53–175.

Bezzi, M. (1916) On the fruit flies of the genus Dacus (s.l.) occurringin India, Burma, and Ceylon. Bulletin of Entomological Research, 7,99–121.

Bo, W., Ahmad, S., Dammalage, T. et al. (2014) Mating compatibilitybetween Bactrocera invadens and Bactrocera dorsalis (Diptera:Tephritidae). Journal of Economic Entomology, 107, 623–629.

Boykin, L.M., Schutze, M.K., Krosch, M.N. et al. (2014) Multi-genephylogenetic analysis of south-east Asian pest members of theBactrocera dorsalis species complex (Diptera: Tephritidae) does notsupport current taxonomy. Journal of Applied Entomology, 138,235–253.

Clarke, A.R. & Schutze, M.K. (in press) The complexities of knowingwhat it is you are trapping. Trapping Tephritid Fruit Flies forDetection, Control, and Regulation (ed. by T. Shelly, N. Epsky, E.Jang, J. Reyes-Flores and R. Vargas). Springer, Dordrecht.

Clarke, A.R., Allwood, A., Chinajariyawong, A. et al. (2001) Seasonalabundance and host use patterns of seven Bactrocera Macquartspecies (Diptera: Tephritidae) in Thailand and Peninsular Malaysia.The Raffles Bulletin of Zoology, 49, 207–220.

Clarke, A.R., Armstrong, K.F., Carmichael, A.E. et al. (2005) Invasivephytophagous pests arising through a recent tropical evolutionaryradiation: the Bactrocera dorsalis complex of fruit flies. AnnualReview of Entomology, 50, 293–319.

David, K.J. & Ramani, S. (2011) An illustrated key to fruit flies(Diptera: Tephritidae) from Peninsular India and Andaman andNicobar Islands. Zootaxa, 3021, 1–31.

De Meyer, M., Robertson, M.P., Mansell, M.W. et al. (2010) Ecologicalniche and potential geographic distribution of the invasive fruit flyBactrocera invadens (Diptera, Tephritidae). Bulletin of EntomologicalResearch, 100, 35–48.

Drew, R.A.I. (1982) Taxonomy. Economic Fruit Flies of the SouthPacific Region (ed. by R.A.I. Drew, G.H.S. Hooper and M.A.Bateman), pp. 1–92. Queensland Department of Primary Industries,Brisbane.

Drew, R.A.I. (1989) The tropical fruit flies (Diptera: Tephritidae:Dacinae) of the Australasian and Oceanian regions. Memoirs of theQueensland Museum, 26, 1–521.

Drew, R.A.I. & Hancock, D. (1994) The Bactrocera dorsalis complexof fruit flies (Diptera: Tephritidae: Dacinae) in Asia. Bulletin ofEntomological Research(Suppl. 2), i–iii + 1–68.

Drew, R.A.I. & Romig, M.C. (1997) Overview – Tephritidae in thePacific and Southeast Asia. Management of Fruit Flies in the Pacific:A Regional Symposium, Naji, Fiji 28–31 October 1996. ACIARProceedings No. 76 (ed. by A.J. Allwood and R.A.I. Drew), pp.46–53. ACIAR, Canberra.

Drew, R.A.I. & Romig, M. (2013) Tropical Fruit Flies of South-EastAsia (Tephritidae: Dacinae). CABI, Wallingford.

Drew, R.A.I., Tsuruta, K. & White, I.M. (2005) A new species of pestfruit fly (Diptera: Tephritidae: Dacinae) from Sri Lanka and Africa.African Entomology, 13, 149–154.

Drew, R.A.I., Romig, M.C. & Dorji, C. (2007) Records of Dacine fruitflies and new species of Dacus (Diptera: Tephritidae) in Bhutan. TheRaffles Bulletin of Zoology, 55, 1–21.

Drew, R.A.I., Raghu, S. & Halcoop, P. (2008) Bridging the morphologi-cal and biological species concepts: studies on the Bactrocera dorsalis(Hendel) complex (Diptera: Tephritidae: Dacinae) in South-east Asia.Biological Journal of the Linnaean Society, 93, 217–226.

Dyck, V.A., Hendrichs, J. & Robinson, A.S. (eds) (2005) Sterile InsectTechnique Principles and Practice in Area-Wide Integrated PestManagement. Springer, Dordrecht.

Fabricius, J.C. (1794) Entomologia systematica emendata et aucta,Vol. 4, 472 pp. Impensis Christ. Gottl. Proft., Copenhagen.

Fabricius, J.C. (1805) Systema antliatorum secundum ordines, genera,species, 372 + 30 pp. Apud Carolum Reichard, Brunswick.

FAO/IAEA/USDA (2003) Manual for Product Quality Control andShipping Procedures for Sterile-Mass-Reared Tephritid Fruit Flies,Version V . IAEA, Vienna.

Fletcher, M.T. & Kitching, W. (1995) Chemistry of fruit flies. ChemicalReviews, 95, 789–828.

Frey, J.E., Guillén, L., Frey, B. et al. (2013) Developing diagnostic SNPpanels for the identification of true fruit flies (Diptera: Tephritidae)within the limits of COI-based species delimitation. BMC Evolution-ary Biology, 13, 106. DOI: 10.1186/1471-2148-13-106.

Hardy, D.E. (1969) Taxonomy and distribution of the Oriental fruitfly and related species (Tephritidae-Diptera). Proceedings of theHawaiian Entomological Society, 20, 395–428.

Hardy, D.E. (1973) The fruit flies (Tephritidae-Diptera) of Thailand andbordering countries. Pacific Insects Monograph, 31, 1–353.

Hardy, D.E. (1974) The fruit flies of the Philippines (Diptera: Tephriti-dae). Pacific Insects Monograph, 32, 1–266.

Hardy, D.E. (1977) Family Tephritidae. A Catalogue of the Diptera ofthe Oriental Region, Vol. 3 (ed. by M.D. Delfinado and D.E. Hardy),pp. 44–134. University Press of Hawaii, Honolulu, Hawaii.

Hardy, D.E. & Adachi, M.S. (1956) Insects of Micronesia, Diptera:Tephritidae. Bishop Museum Bulletin, 14, 1–28.

Hendel, F. (1912) H. Sauter’s Formosa-Ausbeute. Genus Dacus (Dipt.).Supplementa Entomologica, 1, 13–24.

Hendel, F. (1915) H. Sauter’s Formosa-Ausbeute. Tephritinae. AnnalsMusei Nationalis Hungarici, 13, 424–467.

Hering, E.M. (1956) Trypetidae (Dipt.) von Ceylon. Verhandlungen derNaturforschenden Gesellshaft in Basel, 67, 62–74.

Hill, M.P. & Terblanche, J.S. (2014) Niche overlap of congenericinvaders supports a single-species hypothesis and provides insightinto future invasion risk: implications for global management of theBactrocera dorsalis complex. PLoS One, 9, e90121.

Iwahashi, O. (1999a) Distinguishing between two sympatric species,Bactrocera occipitalis and B. philippinensis (Diptera: Tephritidae),based on aedeagal length. Annals of the Entomological Society ofAmerica, 92, 182–187.

Iwahashi, O. (1999b) Distinguishing between the two sympatric speciesBactrocera carambolae and B. papayae (Diptera: Tephritidae) basedon aedeagal length. Annals of the Entomological Society of America,92, 639–643.

Iwahashi, O. (2000) Speciation of Bactrocera dorsalis complex based onaedeagal length. Area-Wide Control of Fruit Flies and Other InsectPests (ed. by K.H. Tan), pp. 591–594. Penerbit Universiti SainsMalaysia, Pulau Pinang.

Iwahashi, O. (2001) Aedeagal length of the Oriental fruit fly, Bactroceradorsalis (Hendel) (Diptera: Tephritidae), and its sympatric species in



Thailand and the evolution of a longer and shorter aedeagus in theparapatric species of B. dorsalis. Applied Entomology and Zoology,36, 289–297.

Iwaizumi, R., Kaneda, M. & Iwahashi, O. (1997) Correlation of lengthof terminalia of males and females among nine species of Bactrocera(Diptera: Tephritidae) and differences among sympatric species ofB. dorsalis complex. Annals of the Entomological Society of America,90, 664–666.

Khamis, F.M., Karam, N., Ekesi, S. et al. (2009) Uncovering the tracksof a recent and rapid invasion: the case of the fruit fly pest Bactrocerainvadens (Diptera: Tephritidae) in Africa. Molecular Ecology, 18,4798–4810.

Khamis, F.M., Masiga, D.K., Mohamed, S.A. et al. (2012) Taxo-nomic identity of the invasive fruit fly pest, Bactrocera invadens:concordance in morphometry and DNA barcoding. PLoS One, 7,e44862.

Krosch, M.N., Schutze, M.K., Armstrong, K.F. et al. (2012) A molecularphylogeny for the Tribe Dacini (Diptera: Tephritidae): Systematic andbiogeographic implications. Molecular Phylogenetics and Evolution,64, 513–523.

Krosch, M.N., Schutze, M.K., Armstrong, K.F. et al. (2013) Piecingtogether an integrative taxonomic puzzle: microsatellite, wing shapeand aedeagus length analyses of Bactrocera dorsalis s.l. (Diptera:Tephritidae) find no evidence of multiple lineages in a proposed con-tact zone along the Thai/Malay Peninsula. Systematic Entomology,38, 2–13.

Leblanc, L., Hossain, M.A., Khan, S.A. et al. (2013) A preliminarysurvey of the fruit flies (Diptera: Tephritidae: Dacinae) of Bangladesh.Proceedings of the Hawaiian Entomological Society, 45, 51–58.

Lin, X.F., Wang, X.J. & Zeng, L. (2011) Three new species of the genusBactrocera Macquart (Diptera, Tephritidae, Dacinae) from Hainan,China. Acta Zootaxonomica Sinica, 36, 896–900.

Lux, S.A., Copeland, R.S., White, I.M. et al. (2003) A new invasive fruitfly species from the Bactrocera dorsalis (Hendel) group detected inEast Africa. Insect Science and its Application, 23, 355–361.

Mahmood, K. (1999) Intraspecific variations in two pest species of theOriental fruit fly, Bactrocera dorsalis (Hendel) (Diptera: Tephritidae)complex. Pakistan Journal of Zoology, 31, 315–321.

Mahmood, K. (2004) Identification of pest species in Oriental fruit fly,Bactrocera dorsalis (Hendel) (Diptera: Tephritidae) species complex.Pakistan Journal of Zoology, 36, 219–230.

Mahmood, K. & Hasan, S.A. (2005) Revision of subfamily Dacinae(Fruit flies) (Diptera: Tephritidae) of South-East Asia. PakistanJournal of Zoology (Suppl. Ser. 7), 1–65.

McInnis, D.O., Rendon, P., Jang, E. et al. (1999) Interspecific matingof introduced, sterile Bactrocera dorsalis with wild B. carambolae(Diptera: Tephritidae) in Suriname: a potential case for cross-speciessterile insect technique. Annals of the Entomological Society ofAmerica, 92, 758–765.

Medina, F.I.S., Carillo, P.A.V., Gregorio, J.S. et al. (1998) The matingcompatibility between Bactrocera philippinensis and Bactroceradorsalis. 5th International Symposium on Fruit Flies of EconomicImportance (ed. by K.H. Tan), p. 155, Penang.

Muraji, M. & Nakahara, S. (2002) Discrimination among pest speciesof Bactrocera (Diptera: Tephritidae) based on PCR-RFLP of themitochondrial DNA. Applied Entomology and Zoology, 37, 437–446.

Naeole, C.K.M. & Haymer, D.S. (2003) Use of oligonucleotide arraysfor molecular taxonomic studies of closely related species in theoriental fruit fly (Bactrocera dorsalis) complex. Molecular EcologyNotes, 3, 662–665.

Nishida, R., Tan, K.-H., Serit, M. et al. (1988) Accumulation ofphenylpropanoids in the rectal glands of males of the Oriental fruitfly, Dacus dorsalis. Experientia, 44, 534–536.

Norrbom, A.L., Carroll, L.E., Thompson, F.C. et al. (1998) Systematicdatabase of names. Fruit Fly Expert Identification System and Sys-tematic Information Database, Myia, Vol. 9 (ed. by F.C. Thompson),pp. 65–251. Backhuys Publishers, The Netherlands.

Perkins, M.V., Fletcher, M.T., Kitching, W. et al. (1990) Chemicalstudies of rectal gland secretions of some species of Bactroceradorsalis complex of fruit flies (Diptera: Tephritidae). Journal ofChemical Ecology, 16, 2475–2487.

de Queiroz, K. (2007) Species concepts and species delimitation.Systematic Biology, 56, 879–886.

Ranganath, H.R. & Veenakumari, K. (1995a) Notes on the dacine fruitflies (Diptera: Tephritidae) of Andaman and Nicobar Islands. TheRaffles Bulletin of Zoology, 43, 235–238.

Ranganath, H.R. & Veenakumari, K. (1995b) Notes on the dacine fruitflies (Diptera: Tephritidae) of Andaman and Nicobar Islands II. TheRaffles Bulletin of Zoology, 47, 221–224.

San Jose, M., Leblanc, L., Geib, S.M. et al. (2013) An evaluation ofthe species status of Bactrocera invadens and the systematics of theBactrocera dorsalis (Diptera: Tephritidae) complex. Annals of theEntomological Society of America, 106, 684–694.

van Sauers-Muller, A. (1991) An overview of the Carambola fruitfly Bactrocera species (Diptera: Tephritidae), found recently inSuriname. Florida Entomologist, 74, 432–440.

van Sauers-Muller, A. (2005) Host plants of the Carambola Fruit Fly,Bactrocera carambolae Drew & Hancock (Diptera: Tephritidae), inSuriname, South America. Neotropical Entomology, 34, 203–214.

van Sauers-Muller, A. (2008) Carambola fruit fly situation in LatinAmerica and the Caribbean. Proceedings of the Caribbean FoodCrops Society, 44, 135–144.

Schlick-Steiner, B.C., Steiner, F.M., Seifert, B. et al. (2010) Integrativetaxonomy: a multisource approach to exploring biodiversity. AnnualReview of Entomology, 55, 421–438.

Schutze, M.K., Jessup, A. & Clarke, A.R. (2012a) Wing shape as apotential discriminator of morphologically similar pest taxa within theBactrocera dorsalis species complex (Diptera: Tephritidae). Bulletinof Entomological Research, 102, 103–111.

Schutze, M.K., Krosch, M.N., Armstrong, K.F. et al. (2012b) Populationstructure of Bactrocera dorsalis s.s., B. papayae and B. philippinensis(Diptera: Tephritidae) in southeast Asia: evidence for a single specieshypothesis using mitochondrial DNA and wing-shape data. BMCEvolutionary Biology, 130. DOI: 10.1186/1471-2148-12-30.

Schutze, M.K., Jessup, A., Ul-Haq, I. et al. (2013) Mating compatibilityamong four pest members of the Bactrocera dorsalis fruit fly speciescomplex (Diptera: Tephritidae). Journal of Economic Entomology,106, 695–707.

Schutze, M.K., Mahmood, K., Pavasovic, A. et al. (2014) One and thesame: integrative taxonomic evidence that the African Invasive FruitFly Bactrocera invadens (Diptera: Tephritidae), is the same speciesas the Oriental Fruit Fly Bactrocera dorsalis. Systematic Entomology,DOI: 10.1111/syen.12114.

Shelly, T.E. & Dewire, A.-L.M. (1994) Chemically mediated matingsuccess in male Oriental fruit flies (Diptera: Tephritidae). Annals ofthe Entomological Society of America, 87, 375–382.

Shelly, T., Resilva, S., Reyes, M. et al. (1996) Methyl eugenol andmating competitiveness of irradiated male Bactrocera philippinensis(Diptera: Tephritidae). Florida Entomologist, 79, 481–488.

Shiraki, T. (1933) A systematic study of Trypetidae in the JapaneseEmpire. Memoirs of the Faculty of Science and Agriculture, TaihokuUniversity, 8, 1–509.

Shiraki, T. (1968) Fruit flies of the Ryukyu Islands. United StatesNational Museum Bulletin, 263, 1–104.

Stephens, A.E.A., Kriticos, D.J. & Leriche, A. (2007) The current andfuture potential geographical distribution of the oriental fruit fly,



Bactrocera dorsalis (Diptera: Tephritidae). Bulletin of EntomologicalResearch, 97, 369–378.

Tan, K.H. (2000) Behaviour and chemical ecology of fruit flies.Area-wide Control of Fruit Flies and Other Insect Pests (ed. by K.H.Tan), pp. 647–656. Penerbit Universiti Sains Malaysia, Pulau Pinang.

Tan, K.H. & Nishida, R. (1996) Sex pheromone and mating competitionafter methyl eugenol consumption in Bactrocera dorsalis complex.Fruit Fly Pests – A World Assessment of their Biology and Manage-ment (ed. by B.A. McPheron and G.S. Steck), pp. 147–153. St. LuciePress, Delray Beach, Florida.

Tan, K.H. & Nishida, R. (2012) Methyl eugenol – its occurrence,distribution, and role in nature, especially in relation to insect behaviorand pollination. Journal of Insect Science, 20, 74 pp.

Tan, K.H., Tokushima, I., Ono, H. et al. (2011) Comparison of phenyl-propanoid volatiles in male rectal pheromone gland after methyleugenol consumption, and molecular phylogenetic relationship offour global pest fruit fly species: Bactrocera invadens, B. dorsalis,B. correcta and B. zonata. Chemoecology, 21, 25–33.

Tan, K.H., Wee, S.L., Ono, H. et al. (2013) Comparison of methyleugenol metabolites, mitochondrial COI, and rDNA sequences ofBactrocera philippinensis (Diptera: Tephritidae) with those of threeother major pest species within the dorsalis complex. Applied Ento-mology and Zoology, 48, 275–282.

Tan, K.H., Nishida, R., Jang, E.B. et al. (in press) Pheromones,male lures, and trapping of tephritid fruit flies. Trapping Tephri-tid Fruit Flies for Detection, Control, and Regulation (ed. by

T. Shelly, N. Epsky, E. Jang, J. Reyes-Flores and R. Vargas). Springer,Dordrecht.

Tseng, Y.H., Chen, C.C. & Chu, Y.I. (1992) The fruit flies, genus DacusFabricius of Taiwan (Diptera: Tephritidae). Journal of the TaiwanMuseum, 45, 15–91.

Wee, S.L. & Tan, K.H. (2007) Temporal accumulation of phenyl-propanoids in male fruit flies, Bactrocera dorsalis and B. carambo-lae (Diptera: Tephritidae) following methyl eugenol consumption.Chemoecology, 17, 81–85.

Wee, S.L., Hee, A.K.W. & Tan, K.H. (2002) Comparative sensitivityto and consumption of methyl eugenol in three Bactrocera dorsalis(Diptera: Tephritidae) complex sibling species. Chemoecology, 12,193–197.

White, I.M. (2006) Taxonomy of the Dacina (Diptera: Tephritidae) ofAfrica and the Middle East. African Entomology Memoir, 2, 1–156.

Yeates, D.K., Seago, A., Nelson, L. et al. (2011) Integrative taxonomy,or iterative taxonomy? Systematic Entomology, 36, 209–217.

Yong, H.S. (1995) Genetic differentiation and relationships in five taxaof the Bactrocera dorsalis complex (Insecta: Diptera: Tephritidae).Bulletin of Entomological Research, 85, 431–435.

Zacharopoulou, A., Augustinos, A.A., Sayed, W.A.A. et al. (2011)Mitotic and polytene chromosomes analysis of the oriental fruit fly,Bactrocera dorsalis (Hendel) (Diptera: Tephritidae). Genetica, 139,79–90.

Accepted 15 September 2014


Documents

Synonymization of key pest species within the Bactrocera ... · Synonymization of key pest species within the Bactrocera dorsalis species complex (Diptera: Tephritidae): taxonomic