Phylogeography of Walnut-Infesting Rhagoletis suavis (Diptera: Tephritidae) Flies

Phylogeography of Walnut-Infesting Rhagoletis suavis (Diptera: Tephritidae) Flies Abstract Discerning the biogeography and historical ranges of organisms is important to understanding the processes causing population divergence and speciation. Mountainous regions in North America have contributed to widespread divergence within animals and plants as species become geographically isolated and diverge. Here, we investigate patterns of divergence for six species of walnut-infesting flies in the Rhagoletis suavis species group (Diptera: Tephritidae) in the United States and Mexico based on sequence analysis of mitochondrial DNA (mtDNA) encoded Cytochrome Oxidase I and II genes (COI and COII). We resolved the relationship of the newly described species, Rhagoletis ramosae, found in the highlands of Mexico, within the R. suavis species group to test alternate hypotheses of migration and divergence. Bayesian phylogenetic analysis supported a clade in which R. ramosae was most closely related to Rhagoletis zoqui/Rhagoletis completa, found in eastern Mexico. This implies that the Sierra Madre Oriental and not the Sierra Madre Occidental have been a major conduit of migration, isolation, and speciation for walnut flies between Mexico and the United States. Comparisons of mtDNA divergence for R. suavis group flies with the Rhagoletis pomonella and Rhagoletis cingulata species groups suggested that despite current similarities in geographic distributions, these taxa do not share a common biogeographic history, diverging in different regions at different times in the past. Patterns displayed by Rhagoletis flies can be compared to patterns seen in other organisms through the Southwestern and Eastern United States, and Mexico to develop a fuller understanding of the biogeography of these regions of North America. Allopatry, the geographic separation of populations (Feder 2017), is often an important factor facilitating population divergence leading to speciation in many animals and plants (Mayr 1963, Coyne and Orr 2004, Nosil 2012). When populations reside in different areas with little or no gene flow between them, they can accumulate genetic differences independent from one another. Given enough time, the isolated populations can eventually accrue enough differences to become reproductively isolated and diverge into separate species. Understanding the geographic distribution of species through time and space (biogeography) is therefore a key element to discerning the processes giving rise to new biodiversity (Klicka and Zink 1997, Avise and Walker 1998, Avise et al. 1998). Knowing the historic distributions of taxa can also be important for discerning how climate and ecology interact with spatial isolation and local adaptation to affect dispersal and population divergence (Lieberman 2003). The genus Rhagoletis (Diptera: Tephritidae) contains a number of different species groups native to North America that have contributed to our understanding of the roles that geography and ecology can play in the speciation process (Bush 1966, Bush and Smith 1998, Feder et al. 2003). The Rhagoletis suavis species group includes six species (Rhagoletis suavis (Loew), Rhagoletis juglandis (Cresson), Rhagoletis completa (Cresson), Rhagoletis boycei (Cresson), Rhagoletis zoqui (Bush), and Rhagoletis ramosae (Hernández-Ortiz)) (Fig. 1A), which are currently largely geographically isolated from one another in the United States and Mexico (Fig. 1B). R. suavis flies exploit their host plants by infesting the husks of walnuts (Juglans spp. (Fagales: Juglandaceae)) (Fig. 1C) and are thought to have undergone allopatric speciation (Bush 1966), which is different from most Rhagoletis species groups that overlap geographically. Walnut flies also differ from other Rhagoletis species groups in that the species are morphologically distinct from one another in wing pattern and body coloration and display different courtship behaviors, which is thought to be associated with sexual selection (Bush 1966, Bush and Smith 1998). Some walnut fly species currently overlap with one another in a portion of their geographic ranges, likely the result of secondary contact (Bush 1966, Bush and Smith 1998). In the southwestern United States, R. juglandis co-occurs in part of its distribution with R. boycei, while in the east it co-exists with R. completa (Fig. 1B). In the midwestern United States, R. completa and R. suavis partially overlap, while in Mexico R. completa and R. zoqui co-occur in a narrow region in the area of Acahuales. A hybrid zone has been shown to exist between two species, R. completa and R. zoqui, in Acahuales, Mexico (see star in Fig. 1B; Rull et al. 2012). No other hybrid zone has yet been found between any other pair of overlapping R. suavis species in nature. Fig. 1. View large Download slide Phylogenetic relationships of R. suavis group flies, their geographic distributions, and the ranges of their walnut host trees in North America. (A) mtDNA cladogram of the R. suavis group showing relationships resolved from previous studies (Smith and Bush 1997, Bush and Smith 1998). The relationship of R. ramosae in the earlier studies is depicted as unknown by the dotted line and question mark (B) Species distributions of walnut-infesting Rhagoletis in the United States and Mexico. Site locations for flies sequenced for COI and COII are distributed across distributions of each species, and numerical designations for sites analyzed in the study are shown (Supplementary Table S2). The hybrid zone between R. completa and R. zoqui found in Acahuales, Tamaulipas, Mexico is indicated by a star; (C) Species distributions of walnuts (Juglans spp.). We used mtDNA sequencing to determine the primary connectivity of walnut flies with alternative hypotheses that divergence occurred to the east along the Sierra Madre Oriental Mountains (solid arrow = SMOr) or in the west through the Sierra Madre Occidental Mountains (dotted arrow = SMOc). Fig. 1. View large Download slide Phylogenetic relationships of R. suavis group flies, their geographic distributions, and the ranges of their walnut host trees in North America. (A) mtDNA cladogram of the R. suavis group showing relationships resolved from previous studies (Smith and Bush 1997, Bush and Smith 1998). The relationship of R. ramosae in the earlier studies is depicted as unknown by the dotted line and question mark (B) Species distributions of walnut-infesting Rhagoletis in the United States and Mexico. Site locations for flies sequenced for COI and COII are distributed across distributions of each species, and numerical designations for sites analyzed in the study are shown (Supplementary Table S2). The hybrid zone between R. completa and R. zoqui found in Acahuales, Tamaulipas, Mexico is indicated by a star; (C) Species distributions of walnuts (Juglans spp.). We used mtDNA sequencing to determine the primary connectivity of walnut flies with alternative hypotheses that divergence occurred to the east along the Sierra Madre Oriental Mountains (solid arrow = SMOr) or in the west through the Sierra Madre Occidental Mountains (dotted arrow = SMOc). Except for R. ramosae, the phylogenetic relationships of R. suavis group flies in the United States and Mexico have been determined based on mitochondrial DNA (mtDNA) (Fig. 1A) (Smith and Bush 1997, Rull et al. 2013). R. ramosae is native to the Eje Volcánico Trans Mexicano (the central Altiplano or trans-Mexican volcanic belt; abbreviated EVTM hereafter) (Rull et al. 2013). The EVTM in the highlands of Mexico is a hotspot of biodiversity because of the mountainous terrain and volcanic regions that separate populations (Knowles 2001, Bryson et al. 2011a , Domínguez-Domínguez et al. 2011). Pleistocene glacial–interglacial climatic cycles, and the resultant expansions and contractions of the Mexican pine–oak forest, have driven widespread divergence within plants (Luna Vega et al. 1999, Jaramillo-Correa et al. 2008), scarabs (Lobo and Halffter 2000), reptiles, amphibians (Bryson et al. 2011b), and mammals (Ceballos et al. 2010). From previous mtDNA phylogenetic analysis, two scenarios were proposed by Rull et al. (2013) regarding R. ramosae (Fig. 1C). The first hypothesis is that R. ramosae is most closely related to R. boycei or R. juglandis in the southwestern United States. If this hypothesis is true, then an important axis of past gene flow and current geographic isolation for walnut flies occurred in is the Sierra Madre Occidental Mountains (SMOc) in western Mexico, resulting in the divergence of R. ramosae in the south and R. boycei or R. juglandis to the north. The second hypothesis is that R. ramosae is most closely related to R. completa or R. zoqui. In this case, the divergence of R. ramosae could be attributed to an ~90 km gap in suitable Rhagoletis habitat that exists between the EVTM and Sierra Madre Oriental Mountains (SMOr) in east-central Mexico, potentially corresponding to the location and time of divergence in other species groups. Resolving the relationship of R. ramosae to other R. suavis taxa can allow us to compare the speciation history of the R. suavis species group to other Rhagoletis flies in the highlands of Mexico. For example, host races of the species Rhagoletis pomonella (Walsh) are also present throughout the United States and Mexico where they attack the fruit of several different species of hawthorns (Crataegus spp. (Rosales: Rosaceae)), as well as domesticated apple (Malus domestica (Miller) (Rosales: Rosaceae) (Bush 1969). Unlike walnut flies, R. pomonella has been proposed to speciate in the absence of complete geographic isolation in the process of shifting and ecologically adapting to new host plants (Bush 1969, Berlocher and Feder 2002). An isolated population of hawthorn-infesting R. pomonella exists in the EVTM and is estimated to have become separated from flies in the SMOr over one million years ago (Feder et al. 2003, 2005; Michel et al. 2007). It has been hypothesized that periodic gene flow from the EVTM into the SMOr and more northern United States populations of hawthorn flies played a role in the radiation of the R. pomonella group onto a number of novel hosts with differing fruiting times (Feder et al. 2003, 2005; Xie et al. 2007). By comparing the estimates of divergence time for R. ramosae with hawthorn-infesting R. pomonella in the EVTM, we can infer if the two flies became isolated by an historically shared or different geographic isolating event(s). The objective of the current study was to resolve the biogeography of the R. suavis species group through DNA sequence analysis of a portion of the mitochondrial genome of these flies. We sequenced the Cytochrome Oxidase subunit I and II (COI and COII) mtDNA genes from 33 individuals from the R. suavis species group and the outgroup taxon, Rhagoletis cingulata (Loew) (Diptera:Tephritidae), to determine the phylogenetic relationship of R. ramosae to the five other members of the R. suavis group and test the two hypotheses of migration proposed by Rull et al. (2013). We also compared the divergence pattern of the R. suavis flies to other species groups of Rhagoletis in North America to determine whether any similarities that exist may be due to a shared biogeographic history of isolation. Methods Species Collections Walnuts infested with Rhagoletis fly larvae were collected from 1983 to 2012 from a total of 16 sites across the United States and Mexico (Table 1, Fig. 1B) and taken back to the laboratory and held on wire mesh racks over plastic collecting trays in a greenhouse experiencing ambient environmental conditions. Trays were monitored on a daily basis, and after larvae fed within the husks of walnuts, left the fruit, and formed puparia in the trays, they were collected and stored in Petri dishes containing moist vermiculite. With the exception of samples collected in 2012, which were immediately frozen at −20°C as pupae, the Petri dishes containing pupae were stored at 4°C for 4–6 mo to simulate winter, after which time they were kept at 22°C in a constant temperature room and 14:10 h light:dark cycle until flies eclosed as adult and were frozen at −20°C for later genetic analysis. Table 1. Site locations, GPS coordinates, and collection information for walnut-infesting R. suavis species group flies and the outgroup, R. cingulata, sequenced for mtDNA ID  Location  Latitude  Longitude  Collector  boycei_1  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_2  Ft. Apache, AZ  34.0645  −109.9135  J.L. Feder  boycei_3  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_4  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_5  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_6  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_7  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_8  Sedona, AZ  34.9126  −111.7268  D. Howard  completa_1  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_2  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_3  Edmond, OK  35.6227  −96.4424  M.M. Glover  completa_4  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_5  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_6  Kearney, NE  40.6936  −99.0897  M.M. Glover  completa_7  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_8  Edmond, OK  35.6227  −96.4424  M.M. Glover  juglandis_1  Cibola, AZ  34.0242  −107.1303  J.L. Feder  juglandis_2  Carrizo, AZ  34.0007  −110.2639  J.L. Feder  juglandis_3  Pine, AZ  34.3846  −111.4547  J.L. Feder  juglandis_4  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  juglandis_5  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  ramosae_1  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_2  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_3  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  suavis_1  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_2  Blacksburg, VA  37.2255  −80.4259  D. Cochran  suavis_3  Buckner, MO  39.1342  −94.2011  R. Cook  suavis_4  Colfax, IA  41.6751  −93.2385  M.M. Glover  suavis_5  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_6  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_7  Blacksburg, VA  37.2255  −80.4259  D. Cochran  zoqui  Huamantla, Tlaxcala, Mexico  19.3261  −97.9254  J. Rull  cingulata  Coahuila, Nueva León, Mexico  25.8136  −100.3673  S.P. Egan  ID  Location  Latitude  Longitude  Collector  boycei_1  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_2  Ft. Apache, AZ  34.0645  −109.9135  J.L. Feder  boycei_3  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_4  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_5  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_6  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_7  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_8  Sedona, AZ  34.9126  −111.7268  D. Howard  completa_1  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_2  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_3  Edmond, OK  35.6227  −96.4424  M.M. Glover  completa_4  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_5  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_6  Kearney, NE  40.6936  −99.0897  M.M. Glover  completa_7  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_8  Edmond, OK  35.6227  −96.4424  M.M. Glover  juglandis_1  Cibola, AZ  34.0242  −107.1303  J.L. Feder  juglandis_2  Carrizo, AZ  34.0007  −110.2639  J.L. Feder  juglandis_3  Pine, AZ  34.3846  −111.4547  J.L. Feder  juglandis_4  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  juglandis_5  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  ramosae_1  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_2  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_3  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  suavis_1  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_2  Blacksburg, VA  37.2255  −80.4259  D. Cochran  suavis_3  Buckner, MO  39.1342  −94.2011  R. Cook  suavis_4  Colfax, IA  41.6751  −93.2385  M.M. Glover  suavis_5  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_6  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_7  Blacksburg, VA  37.2255  −80.4259  D. Cochran  zoqui  Huamantla, Tlaxcala, Mexico  19.3261  −97.9254  J. Rull  cingulata  Coahuila, Nueva León, Mexico  25.8136  −100.3673  S.P. Egan  ID = species identification. View Large Mitochondrial Sequencing Total genomic and mtDNA was extracted from individual flies using the Puregene DNA purification kit (Qiagen, Inc. Hilden, Germany). The polymerase chain reaction (PCR) was performed on purified DNA isolated from individual flies to amplify a portion of the mtDNA COI and COII genes. COI was amplified using the primer pairs Lep1F (5′-ATTCAACCAAT CATAAAGATATTGG-3′) and Lep1R (5′-TAAACTTCTGGATGTCCAAAAAATCA-3′) (Hebert et al. 2004). COII was amplified using either the primer pairs C1-J-2792 (George: 5′-ATACCTCGACGTTATTCAGA-3′) and TK-N-3772 (Eva: 5′-GAGACCAT TACTTGCTTTCAGTCATCT-3′) (Simons et al. 1994) or the newly designed R. suavis group specific forward (5′-CAACTTACATTCTTTCATGATC-3′) and reverse (5′-GAATTTACTCTATTTGTAATTC-3′) primers, which are internal to George and Eva primers. George and Eva primers did not amplify universally across the R. suavis species group, potentially due to a mutation in the primer site; thus, the analysis was only done on a fragment internal to these primers. PCR was performed with an initial denature step of 94°C for 5 min, followed by 35 cycles of 94°C for 20 s, 50°C/56°C for COI and COII, respectively, for 20 s, and 72°C for 30 s, and a final elongation step of 72°C for 10 min. Following SAP/exo cleanup, the PCR products were sequenced via Sanger sequencing at the University of Arizona Genomics Core (Tuscon, AZ). Phylogenetic Analysis Mitochondrial DNA sequences generated for a total of 33 flies from the 16 walnut collecting sites and R. cingulata, a closely related cherry-infesting fly used as an outgroup, were cleaned and trimmed using SEQUENCHER (GeneCorp, Ann Arbor, MI). The sequences are available on GenBank (accession numbers: MF977833–MF977898). The resulting sequences were then aligned against one another using MUSCLE (Edgar 2004) and the alignment refined, truncating the ends of the sequences to only include sites with information at each base pair (bp) using GBlocks (Castresana 2000) to generate reads with a total length of 541 bp for COI and 409 bp for COII for phylogenetic analysis. jModelTest2 (Guindon and Gascuel 2003, Darriba et al. 2012) identified the Hasegawa, Kishino and Yano (1985) model with invariable sites and rate variation among sites (HKY + I + G) as the nucleotide substitution model having the highest likelihood for explaining the walnut fly mtDNA data based on the Akaike information criterion (AIC). The HKY + I + G model was subsequently employed in a Bayesian phylogenetic analysis with COI, COII, and COI and COII concatenated sequences in MrBayes v.3.2.6 (Huelsenbeck and Ronquist 2001, Ronquist et al. 2012). Bayesian analyzes were conducted with two independent runs each with four Markov chains (three heated and one cold) for one million generations, with a 25% burn in. Default uninformative prior settings were used. Branch lengths for the Bayesian consensus mtDNA trees were estimated with maximum likelihood in PhyML 3.0 based on the HKY + I + G substitution model (Guindon and Gascuel 2003, Guindon et al. 2010). The resulting mtDNA trees were visualized with R. cingulata rooted as the outgroup with the APE package in R v3.3.3 (Paradis et al. 2004, R Core Team, 2017). Genetic Divergence We estimated the divergence times of the COII gene from R. suavis species and R. pomonella, as COI sequences for other species groups are not available, using a Bayesian Markov Chain Monte Carlo (MCMC) approach. COII sequences for the R. pomonella species in the eastern United States (n = 4), SMOr (n = 2), and EVTM (n = 2) were retrieved from Xie et al. (2008) through GenBank (Table 2). The sequences were aligned with the walnut fly COII sequences and truncated to the same 409 bp length fragment. Bayesian MCMC was run independently five times with 10 million generations each using Beast 2.4.7 (Bouckaert et al. 2014, Drummond et al. 2006). We used a gamma site model with four category counts and an HKY DNA substitution model. We used a relaxed lognormal clock with a rate of 3.54% divergence, a common estimate for the mtDNA molecular clock in insects (Papadopoulou et al. 2010). The tree was estimated with a Yule model and priors on the R. suavis and R. pomonella groups to constrain for monophyletic clades for the species group. Independent runs were combined using LogCombiner and summary tree produced using TreeAnnotator with at 10% burn-in (Drummond and Rambaut 2007). Table 2. Accession numbers and sampling information for mtDNA COII sequences for R. pomonella used for estimate divergence time ID  GenBank accession  Species  Population  Host plant  pom_EVTM1  EU109153  R. pomonella  EVTM  Hawthorn  pom_EVTM2  EU109154  R. pomonella  EVTM  Hawthorn  pom_East_haw1  EU109156  R. pomonella  US-East  Hawthorn  pom_East_haw2  EU109157  R. pomonella  US-East  Apple  pom_East_haw3  EU109158  R. pomonella  US-East  Hawthorn  pom_East_apple  EU109159  R. pomonella  US-East  Hawthorn  pom_SMOr1  EU109160  R. pomonella  SMOr  Hawthorn  pom_SMOr2  EU109161  R. pomonella  SMOr  Hawthorn  ID  GenBank accession  Species  Population  Host plant  pom_EVTM1  EU109153  R. pomonella  EVTM  Hawthorn  pom_EVTM2  EU109154  R. pomonella  EVTM  Hawthorn  pom_East_haw1  EU109156  R. pomonella  US-East  Hawthorn  pom_East_haw2  EU109157  R. pomonella  US-East  Apple  pom_East_haw3  EU109158  R. pomonella  US-East  Hawthorn  pom_East_apple  EU109159  R. pomonella  US-East  Hawthorn  pom_SMOr1  EU109160  R. pomonella  SMOr  Hawthorn  pom_SMOr2  EU109161  R. pomonella  SMOr  Hawthorn  View Large Results Most clades in the R. suavis group were well supported in the concatenated COI and COII tree (Fig. 2). For example, R. suavis, R. juglandis, R. boycei, and the clade containing, R. completa, R. zoqui, and R. ramosae each formed monophyletic groups with Bayesian posterior probabilities (PP) ≥0.99 (PP = 1.0, 1.0, 1.0, and 0.99, respectively). However, the phylogenetic relationships among R. completa, R. zoqui, and R. ramosae remained unresolved. R. ramosae possessed two unique bp differences not found in any other walnut fly species, one in COI (bp position 354) and one in COII (bp position 158). These two autapomorphies characterized R. ramosae as a unique clade. However, R. completa and R. zoqui could not be clearly distinguished, making the branching order within the clade containing R. completa, R. zoqui, and R. ramosae uncertain and the group polyphyletic based on mtDNA haplotypes (Fig. 2). Regardless, the COI and COII gene tree indicated that R. ramosae is more closely related to R. zoqui/R. completa and not to R. suavis, R. juglandis, or R. boycei. Fig. 2. View largeDownload slide Bayesian consensus mtDNA tree for combined COI and COII, in the R. suavis species group. Gene tree topologies were estimated for the concatenated COI (541 bp) and COII (409) gene fragments using a Bayesian phylogenetic analysis, with the posterior probabilities (PP) shown for each node (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7). The branch lengths were estimated through maximum likelihood, using PhyML, with R. cingulata used as the outgroup. R. ramosae is most closely related to R. completa/R. zoqui, with the three species forming a monophyletic clade. The position of R. suavis and R. juglandis is reversed in the COI versus COII gene tree, with R. juglandis branching first, followed by R. suavis. Scale for legend = 0.01 substitutions/bp. Fig. 2. View largeDownload slide Bayesian consensus mtDNA tree for combined COI and COII, in the R. suavis species group. Gene tree topologies were estimated for the concatenated COI (541 bp) and COII (409) gene fragments using a Bayesian phylogenetic analysis, with the posterior probabilities (PP) shown for each node (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7). The branch lengths were estimated through maximum likelihood, using PhyML, with R. cingulata used as the outgroup. R. ramosae is most closely related to R. completa/R. zoqui, with the three species forming a monophyletic clade. The position of R. suavis and R. juglandis is reversed in the COI versus COII gene tree, with R. juglandis branching first, followed by R. suavis. Scale for legend = 0.01 substitutions/bp. The mtDNA Bayesian phylogeny for concatenated COI and COII sequences (Fig. 2, individual trees are shown in Supplementary Fig. S1) was congruent with previous analyses of the R. suavis group (Smith and Bush 1997, Bush and Smith 1998). The concatenated COI and COII gene tree had R. suavis as the first species to branch from the common ancestor of walnut flies, followed by R. juglandis speciating from a clade containing all the remaining taxa found in the southwestern United States and Mexico (Fig. 2). Rhagoletis boycei was the next species to become isolated and diverged in southwestern New Mexico and the southeastern and central portions of Arizona from the common ancestor of R. completa, R. zoqui, and R. ramosae (Fig. 2). Here, R. boycei is now found at higher elevations, while the co-occurring R. juglandis is present at lower elevations (Bush 1966). Finally, the clade containing R. completa, R. zoqui, and R. ramosae diversified from north to south, respectively, along an area including the SMOr and EVTM. The Bayesian tree estimates of divergence time followed the same pattern of speciation inferred for the walnut flies (Fig. 3). Estimated divergence times (and 95% highest posterior density, HPD) for the R. suavis, R. juglandis, and R. boycei clades were 0.749 (0.458–1.068), 0.635 (0.362–0.869), and 0.419 (0.240–0.619) MYA, respectively. We were unable to estimate divergence times among R. completa, R. zoqui, and R. ramosae due to lack of mtDNA resolution for this clade. However, given their low level of mtDNA differentiation, these taxa were inferred to form recently. Fig. 3. View largeDownload slide Bayesian estimates of divergence time for R. suavis and R. pomonella species. Levels of divergence were estimated for a 409 bp fragment of the COII gene for R. suavis species and R. pomonella in the eastern United States, SMOr, and the EVTM. Posterior probabilities for the tree topology are colored on the nodes (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7), with light blue bars indicating the 95% highest priority density for node ages. Fig. 3. View largeDownload slide Bayesian estimates of divergence time for R. suavis and R. pomonella species. Levels of divergence were estimated for a 409 bp fragment of the COII gene for R. suavis species and R. pomonella in the eastern United States, SMOr, and the EVTM. Posterior probabilities for the tree topology are colored on the nodes (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7), with light blue bars indicating the 95% highest priority density for node ages. The hawthorn-infesting population of R. pomonella in the EVTM formed a separate clade from the rest of the R. pomonella group with a divergence time of 0.395 (0.180–0.6322) MYA. The rest of the populations, including R. pomonella from the SMOr and apple and hawthorn-infesting populations in the eastern United States, did not form a monophyletic clade and exhibited recent divergence. Discussion The mtDNA phylogeny for COI and COII genes resolved the relationship of R. ramosae to the other members of the R. suavis group, indicating that R. ramosae is most closely related to R. completa and R. zoqui, differing only by two substitutions, one each in the COI and COII genes. These results suggest that R. ramosae in the EVTM diverged relatively recently from R. completa and R. zoqui in the SMOr. This is supported by the evidence of hybridization between R. ramosae and both R. completa and R. zoqui in a laboratory setting (Tadeo 2014). The fact that R. completa does not form a monophyletic clade could be due to gene flow and/or incomplete lineage sorting between the species in the SMOr. A hybrid zone between R. completa and R. zoqui in Acahuales, Mexico (Rull et al. 2012) supports the hypothesis that ongoing introgression may be contributing to the lack of mtDNA phylogenetic resolution for these two taxa. In addition, the finding that R. ramosae is most closely related to species found in the SMOr implies that the major geographic axis for differentiation of walnut flies in Mexico was along the SMOr and EVTM (solid arrow in Fig. 1C), and not the SMOc (dotted arrow in Fig. 1C). As discussed in Rull et al. (2013), these results suggest that divergence of these taxa likely occurred in conjunction with the separation of Juglans major (Heller) and J. mollis (Engelm.)/J. pyriformis (Liebmann) that exists in the SMOr. Thus, the SMOr Mountains may have been a corridor for past migration and isolation between Mexico and the southwestern and eastern United States for walnut fly populations. Mitochondrial DNA sequencing has its limitations for inferring phylogenetic relationships and in estimating divergence times between taxa (Rubinoff and Holland 2005). The current study only considered two mitochondrial genes, which may not give a complete picture of the genetics of speciation and its geographic context for the R. suavis group. Including nuclear genes in the study and increasing the scope of the genomic analysis through next-generation sequencing could provide greater resolution to the history of divergence for walnut flies. For example, mtDNA analysis showed low resolution in the clade containing R. completa, R. zoqui, and R. ramosae, but the phylogeny could be improved by increasing the number of loci. However, for the main objective of the current study, the mtDNA COI and COII genes were clearly sufficient in showing that R. ramosae is closely related to R. completa and R. zoqui and much more distantly related to the other walnut-infesting fly species in the group. Given the maternal inheritance of mtDNA, it is always possible that the close affinity of R. ramosae to R. completa and R. zoqui reflects a recent sweep of a mitochondrial haplotype through R. ramosae, potentially due to a cytoplasmic factor or associated endosymbiont, and not reflect the evolutionary relationship of the species to R. completa and R. zoqui. However, if this were the case, then the mtDNA haplotype and endosymbiont would have likely been transferred together to R. ramosae from one of the other taxa, as the acquisition of a cytoplasmic factor or endosymbiont alone would not result in R. ramosae displaying a close affinity to R. completa or R. zoqui. This would require hybridization between these species, which would support their close phylogenetic relationship. We found that R. suavis and R. pomonella, despite displaying broad similarities in their current geographic distributions through the eastern United States, EVTM, and SMOr, do not appear to share a common biogeographic history. First, both species groups are found in the EVTM, including the walnut fly, R. ramosae, and hawthorn-infesting R. pomonella (Rull et al. 2006, 2013). R. ramosae is relatively closely related to the other walnut-infesting flies R. completa and R. zoqui in the SMOr, differing by only one unique substitution in the COII gene (0.24% sequence divergence; Fig. 3). In comparison, hawthorn-infesting populations of R. pomonella in the EVTM are much more diverged from R. pomonella in the SMOr and eastern United States (Fig. 3), as also found in previous studies which show that R. pomonella in the EVTM are geographically isolated and genetically diverged from the R. pomonella group (Feder et al. 2003). R. pomonella in the EVTM are also post-zygotically isolated from flies in the EVTM, congruent with strong differentiation between the two populations (Rull et al. 2010). We estimated here that the population of R. pomonella in the EVTM diverged around 0.395 (95% HPD: 0.180–0.6322) MYA based on the COII gene (Fig. 3) Second, the R. suavis and R. pomonella groups both have populations in the SMOr and eastern United States but show different levels of divergence between the regions. For walnut flies, R. boycei, R. juglandis, and R. suavis show deep splits between the flies in the SMOr and EVTM, with estimated divergence times 0.419 (0.240–0.619), 0.635 (0.362–0.869), and 0.749 (0.458–1.068) MYA. In comparison, hawthorn-infesting populations of R. pomonella in the eastern United States and SMOr display no fixed difference in COII in the current or in a previous study, suggesting a close evolutionary relationship (Fig. 3; Xie et al. 2007). The cherry-infesting species R. cingulata, belonging to the R. cingulata group, also overlaps in its geographic distribution with R. suavis and R. pomonella species in North America. In particular, R. cingulata co-occurs with R. pomonella and R. suavis in the eastern United States and SMOr. However, cherry-infesting flies have not been found in the EVTM (Rull et al. 2011). Similar to R. pomonella, R. cingulata displays limited mtDNA divergence between the SMOr and the eastern United States, with approximately 0.34% divergence (H. Schuler, personal communication). Like walnut flies, populations of R. cingulata also exist in the southwestern United States, where R. pomonella is not native (Bush 1966). In contrast to the walnut fly species R. boycei and R. juglandis in the southwestern United States, which show deep divergence in COII relative to flies in Mexico, cherry-infesting R. cingulata displays no mtDNA divergence between the southwestern United States and SMOr (H. Schuler, personal Communication), possibly due to either gene flow or relatively recent isolation (Fig. 3). Thus, despite their broadly shared geographic distributions, walnut, hawthorn, and cherry-infesting flies have different patterns of genetic differentiation through the southwestern United States, eastern United States, and Mexico implying that they differ in the times of separation across North America. Except in the EVTM, the R. suavis group shows greater geographic mtDNA differentiation, and by inference past spatial isolation, than R. pomonella and R. cingulata flies across their common shared range. For example, the separation of walnut flies in the SMOr and in the eastern United States appears to have occurred historically earlier than that for hawthorn- and cherry-infesting flies from these two regions. It may be that the species R. suavis in the eastern United States represents a comparatively old range expansion from Mexico and the southwestern United States into the eastern United States, while the center of endemism for hawthorn- and cherry-infested flies may be the eastern United States itself. However, at the current time, we cannot rule out the alternate possibility that the walnut flies may have been originally native to the eastern United States and became isolated in the southwestern United States and Mexico after migrating from the east. Future nuclear DNA sequencing may resolve these issues, for example, by indicating where genomic diversity (allelic polymorphism) is greatest and where it is reduced, potentially allowing identification of the region of origin for walnut flies in North America. The general pattern of increased mtDNA differentiation among walnut flies is consistent with the hypothesis that members of the R. suavis group have speciated allopatrically in the absence of gene flow (Bush 1966). Although R. suavis group flies tend to attack different species of walnut as their major native hosts (Supplementary Table S1, Rull et al. 2013), host specificity of R. suavis flies does not appear to have contributed to their divergence. Each species in the R. suavis group is capable of attacking any walnut present in its range (Rull et al. 2013). However, the geographic ranges of the different walnuts do not overlap greatly in North America, and the ranges of flies closely follow those of their primary walnut host, generating geographic isolation (Fig. 1). We, therefore, argue that it is the largely disjunct nature of the distribution of walnuts rather than any biological characteristic of the walnut hosts themselves that contributed to the radiation of the R. suavis group. The timing of walnut speciation, however, greatly predates that of R. suavis husk flies (Rull et al. 2013). The origin of the walnut genus, Juglans, in North America (c. 60 MYA), and of many present-day walnut species (40–45 MYA), is much older than that for R. suavis group flies (Fjellstrom and Parfitt 1995). As a result, host walnuts imposed a populations structure of geographic isolation on flies, as colonists moved into new regions and became largely associated with a new walnut host, possibly related to range expansions and contractions occurring during glacial and inter-glacial periods. In contrast to walnut flies, speciation in the R. pomonella group and for certain R. cingulata flies has been hypothesized to occur via host shifting without complete geographic isolation. Thus, R. pomonella and R. cingulata populations infesting hawthorns and black cherries, respectively, may have experienced higher levels of gene flow when they diverged compared to walnut flies. In addition, it appears that R. pomonella populations in the SMOr and R. cingulata populations in the SMOr and southwestern United States may have been more recently contiguous with populations in the eastern United States than is the case with walnut flies, the major exception is the hawthorn-infesting population of R. pomonella in the EVTM (Rull et al. 2006). Thus, R. pomonella and R. cingulata generally display reduced mtDNA differentiation than walnut-infesting flies across North America. Rhagoletis flies are not the only organisms to be affected by the mountainous geography of the highlands of Mexico experiencing geographic isolation and subsequent divergence (Knowles 2001, McCormack et al. 2008, Ruiz-Sanchez and Specht 2013). The different mountain ranges in central Mexico, including the SMOr, SMOc, and EVTM, have been shown to structure populations with between mountain ranges or even in some cases along a mountain range, as is seen in Mexican stoneroller fish in the SMOc (Domínguez-Domínguez et al. 2011). Similar to R. pomonella which shows strong differentiation between the EVTM and SMOr, many taxa exhibit distinct breaks in populations between the different mountain ranges in Mexico. McCormack et al. (2008) reported high mtDNA sequence divergence between Mexican Jay populations inhabiting the SMOc, SMOr, and EVTM with speciation occurring 0.7 MYA. Gene flow is also reduced due to geographic isolation in the EVTM in the Mexican shrub, Nolina parviflora, resulting in strong differentiation and population structure in the different mountain ranges (Ruiz-Sanchez and Specht 2013). Further work is needed to determine how congruent the patterns and timing of diversification of these organisms are with Rhagoletis. Our results suggest that there is no universal biogeographical process that affected all Rhagoletis species groups the same and in unison. It will be interesting to determine how the differences displayed by Rhagoletis flies match patterns seen in other organisms in the southwestern and eastern United States and Mexico, to develop a fuller and deeper understanding of the biogeography of these regions of North America. Supplementary Data Supplementary data are available at Insect Systematics and Diversity online. Acknowledgments We thank M.M. Doellman, C. Tait, H. Schuler, P. Meyers, E. Hernandéz, E. Tadeo, E. Acosta, L.A. Miller, C.C. Xu, G. McLain, M. DiGaetano, and Q.D. Read for help with fly collections and rearing in this study. This work was supported by the University of Notre Dame Advanced Diagnostics & Therapeutics Initiative (S.P.E., J.L.F.) and Environmental Change Initiative (S.P.E., J.L.F.), the Instituto de Ecología, A.C. (INECOL) and the Campaña Nacional Contra las Moscas de la Fruta (Convenios INECOL-SENASICA-IICA/CONACOFI) (M.A.), a Rice University Huxley Faculty Fellowship to S.P.E., and NSF and USDA grants to S.P.E. and J.L.F. References Cited Avise, J. C., and D. E. Walker. 1998. Pleistocene phylogeographic effects on avian populations and the speciation process. Proc. R. Soc. B Biol. Sci . 265: 457– 463. Google Scholar CrossRef Search ADS   Avise, J. C., D. Walker, and G. C. Johns. 1998. Speciation durations and Pleistocene effects on vertebrate phylogeography. Proc. R. Soc. B Biol. Sci . 265: 1707– 1712. Google Scholar CrossRef Search ADS   Berlocher, S. H., and J. L. Feder. 2002. Sympatric speciation in phytophagous insects: moving beyond controversy? Annu. Rev. Entomol . 47: 773– 815. Google Scholar CrossRef Search ADS PubMed  Bouckaert, R., J. Heled, D. Kühnert, T. Vaughan, C-H. Wu, D. Xie, M. A. Suchard, A. Rambaut, and Drummond, A. J. 2014. BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput. Biol . 10: e1003537. Google Scholar CrossRef Search ADS PubMed  Bryson, R. W., R. W. Murphy, M. R. Graham, A. Lathrop, and D. Lazcano. 2011a. Ephemeral Pleistocene woodlands connect the dots for highland rattlesnakes of the Crotalus intermedius group. J. Biogeogr . 38: 2299– 2310. Google Scholar CrossRef Search ADS   Bryson, R. W., R. W. Murphy, A. Lathrop, and D. Lazcano-Villareal. 2011b. Evolutionary drivers of phylogeographical diversity in the highlands of Mexico: a case study of the Crotalus triseriatus species group of montane rattlesnakes. J. Biogeogr . 38: 697– 710. Google Scholar CrossRef Search ADS   Bush, G. L. 1966. The taxonomy, cytology, and evolution of the genus Rhagoletis in North America (Diptera, Tephritidae). Bull. Museum Comp. Zool . 134: 431– 562. Bush, G. L. 1969. Sympatric host race formation and speciation in frugivorous flies of the genus Rhagoletis (Diptera, Tephritidae). Evolution (N. Y) . 23: 237– 251. Bush, G. L., and J. J. Smith. 1998. The genetics and ecology of sympatric speciation: a case study. Res. Popul. Ecol. (Kyoto) . 40: 175– 187. Google Scholar CrossRef Search ADS   Castresana, J. 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol. Biol. Evol . 17: 540– 552. Google Scholar CrossRef Search ADS PubMed  Ceballos G., J. Arroyo-Cabrales, and E. Ponce. 2010. Effects of Pleistocene environmental changes on the distribution and community structure of the mammalian fauna of Mexico. Quat. Res. 73: 464– 473. Coyne, J. A., and H. A. Orr. 2004. Speciation . Sinauer Associates, Sunderland, MA. Darriba, D., G. L. Taboada, R. Doallo, and D. Posada. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods . 9: 772. Google Scholar CrossRef Search ADS PubMed  Domínguez-Domínguez, O., M. Vila, R. Pérez-Rodríguez, N. Remón, and I. Doadrio. 2011. Complex evolutionary history of the Mexican stoneroller Campostoma ornatum Girard, 1856 (Actinopterygii: Cyprinidae). BMC Evol. Biol . 11: 153. Google Scholar CrossRef Search ADS PubMed  Drummond, A. J. and A. Rambaut. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol . 7: 214. Google Scholar CrossRef Search ADS PubMed  Drummond, A. J., S. Y. W. Ho, M. J. Phillips and A. Rambaut. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biol . 4: e88 Google Scholar CrossRef Search ADS PubMed  Edgar, R. C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res . 32: 1792– 1797. Google Scholar CrossRef Search ADS PubMed  Feder, J. L. 2017. Allopatric. In B.D. Roitberg (ed.) Ref. Modul. Life Sci . Elsevier Press, Amsterdam, Netherlands. Google Scholar CrossRef Search ADS   Feder, J. L., S. H. Berlocher, J. B. Roethele, H. Dambroski, J. J. Smith, W. L. Perry, V. Gavrilovic, K. E. Filchak, J. Rull, and M. Aluja. 2003. Allopatric genetic origins for sympatric host-plant shifts and race formation in Rhagoletis. Proc. Natl. Acad. Sci. U. S. A . 100: 10314– 10319. Google Scholar CrossRef Search ADS PubMed  Feder, J. L., X. Xie, J. Rull, S. Velez, A. Forbes, B. Leung, H. Dambroski, K. E. Filchak, and M. Aluja. 2005. Mayr, Dobzhansky, and Bush and the complexities of sympatric speciation in Rhagoletis. Proc. Natl. Acad. Sci. U. S. A . 102 ( Suppl): 6573– 6580. Google Scholar CrossRef Search ADS PubMed  Fjellstrom, R. G., and D. E. Parfitt. 1995. Phylogenetic analysis and evolution of the genus Juglans (Juglandaceae) as determined from nuclear genome RFLPs. Plant Syst. Evol . 197: 19– 32. Google Scholar CrossRef Search ADS   Guindon, S., and O. Gascuel. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst. Biol . 52: 696– 704. Google Scholar CrossRef Search ADS PubMed  Guindon, S., J. F. Dufayard, V. Lefort, M. Anisimova, W. Hordijk, and O. Gascuel. 2010. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst. Biol . 59: 307– 321. Google Scholar CrossRef Search ADS PubMed  Hasegawa, M., H. Kishino, and T. Yano. 1985. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J. Mol. Ecol. 22: 160– 174. Hebert, P. D. N., E. H. Penton, J. M. Burns, D. H. Janzen, and W. Hallwachs. 2004. Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator. Proc. Natl. Acad. Sci. U. S. A . 101: 14812– 14817. Google Scholar CrossRef Search ADS PubMed  Huelsenbeck, J. P., and F. Ronquist. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics . 17: 754– 755. Google Scholar CrossRef Search ADS PubMed  Jaramillo-Correa, J. P., E. Aguierre-Planter, D. P. Khasa, L. E. Eguiarte, D. Piñero, G. R. Furnier, and J. Bousquet. 2008. Ancestry and divergence of subtropical montane forest isolates: molecular biogeography of the genus Abies (Pinaceae) in southern México and Guatemala. Mol. Ecol . 17: 2476– 2490. Google Scholar CrossRef Search ADS PubMed  Klicka, J., and R. M. Zink. 1997. The importance of recent ice ages in speciation: a failed paradigm. Science . 277: 1666– 1669. Google Scholar CrossRef Search ADS   Knowles, L. L. 2001. Genealogical portraits of speciation in montane grasshoppers (genus Melanoplus) from the sky islands of the Rocky Mountains. Proc. R. Soc. B Biol. Sci . 268: 319– 324. Google Scholar CrossRef Search ADS   Lieberman, B. S. 2003. Paleobiogeography: the relevance of fossils to biogeography. Annu. Rev. Ecol. Evol. Syst . 34: 51– 69. Google Scholar CrossRef Search ADS   Lobo, J.M., and G. Hallffter. 2000. Biogeographical and Ecological Factors Affecting the Altitudinal Variation of Mountainous Communities of Coprophagous Beetles (Coleoptera: Scarabaeoidea): a Comparative Study. Ann. Entomol. Soc. Am. 93: 115– 126. Luna Vega, I., A. O. Alcántara, O. D. Espinosa, and J. J. Morrone. 1999. Historical relationships of the Mexican cloud forests: a preliminary vicariance model applying par- simony analysis of endemicity to vascular plant taxa. J Biogeogr . 26: 1299– 1305. Google Scholar CrossRef Search ADS   Mayr, E. 1963. Animal Species and Evolution . Harvard University Press, Harvard, MA. Google Scholar CrossRef Search ADS   McCormack, J. E., A. T. Peterson, E. Bonaccorso, and T. B. Smith. 2008. Speciation in the highlands of Mexico: genetic and phenotypic divergence in the Mexican jay (Aphelocoma ultramarina). Mol. Ecol . 17: 2505– 2521. Google Scholar CrossRef Search ADS PubMed  Michel, A. P., J. Rull, M. Aluja, and J. L. Feder. 2007. The genetic structure of hawthorn-infesting Rhagoletis pomonella populations in Mexico: implications for sympatric host race formation. Mol. Ecol . 16: 2867– 2878. Google Scholar CrossRef Search ADS PubMed  Nosil, P. 2012. Ecological Speciation . Oxford University Press, Oxford, UK. Google Scholar CrossRef Search ADS   Papadopoulou, A., I. Anastasiou, and A. P. Vogler. 2010. Revisiting the insect mitochondrial molecular clock: the Mid-Aegean trench calibration. Mol. Biol. Evol . 27: 1659– 1672. Google Scholar CrossRef Search ADS PubMed  Paradis, E., J. Claude, and K. Strimmer. 2004. APE: analyses of phylogenetics and evolution in R language. Bioinformatics . 20: 289– 290. Google Scholar CrossRef Search ADS PubMed  R Core Development Team. 2017. R: a language and environment for statistical computing . R Foundation for Statistical Compution, Vienna, Austria. https://www.R-project.org Ronquist, F., M. Teslenko, P. van der Mark, D. L. Ayres, A. Darling, S. Hohna, B. Larget, L. Liu, M. A. Suchard, and J. P. Huelsenbeck. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst. Biol . 61: 539– 542. Google Scholar CrossRef Search ADS PubMed  Rubinoff, D. and B. S. Holland. 2005. Between two extremes: mitochondrial DNA is neither the panacea nor the nemesis of phylogenetic and taxonomic inference. Syst. Biol . 54: 952– 961. Google Scholar CrossRef Search ADS PubMed  Ruiz-Sanchez, E., and C. D. Specht. 2013. Influence of the geological history of the Trans-Mexican volcanic belt on the diversification of Nolina parviflora (Asparagaceae : Nolinoideae). J. Biogeogr . 40: 1336– 1347. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, J. L. Feder, and S. Berlocher. 2006. Distribution and host range of hawthorn-infesting Rhagoletis (Diptera:Tephritidae) in Mexico. Ann. Entomol. Soc. Am . 99: 662– 672. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, and J. L. Feder. 2010. Evolution of intrinsic reproductive isolation among four North American populations of Rhagoletis pomonella (Diptera: Tephritidae). Biol. J. Linn. Soc . 100: 213– 223. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, and J. L. Feder. 2011. Distribution and basic biology of black cherry-infesting Rhagoletis (Diptera: Tephritidae) in México. Ann. Entomol. Soc. Am . 104: 202– 211. Google Scholar CrossRef Search ADS   Rull, J., E. Tadeo, M. Aluja, L. Guillen, S. P. Egan, and J. L. Feder. 2012. Hybridization and sequential components of reproductive isolation between parapatric walnut-infesting sister species Rhagoletis completa and Rhagoletis zoqui. Biol. J. Linn. Soc . 107: 886– 898. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, E. Tadeo, L. Guillen, S. Egan, M. Glover, and J. L. Feder. 2013. Distribution, host plant affiliation, phenology, and phylogeny of walnut-infesting Rhagoletis flies (Diptera: Tephritidae) in Mexico. Biol. J. Linn. Soc . 110: 765– 779. Google Scholar CrossRef Search ADS   Simons, C., F. Frati, A. Beckenbach, B. Crespi, H. Liu, and P. Flook. 1994. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann. Entomol. Soc. Am . 87: 651– 701. Google Scholar CrossRef Search ADS   Smith, J. J., and G. L. Bush. 1997. Phylogeny of the genus Rhagoletis (Diptera: Tephritidae) inferred from DNA sequences of mitochondrial cytochrome oxidase II. Mol. Phylogenet. Evol . 7: 33– 43. Google Scholar CrossRef Search ADS PubMed  Tadeo, E. 2014. Diferenciación y aislamiento reproductivo entre especies de moscas del grupo suavis en México . PhD Thesis, Universidad Veracruzana. Xie, X., J. Rull, A. P. Michel, S. Velez, A. A. Forbes, N. F. Lobo, M. Aluja, and J. L. Feder. 2007. Hawthorn-infesting populations of Rhagoletis pomonella in Mexico and speciation mode plurality. Evolution (N. Y) . 61: 1091– 1105. Xie, X., A. P. Michel, D. Schwarz, J. Rull, S. Velez, A. A. Forbes, M. Aluja, and J. L. Feder. 2008. Radiation and divergence in the Rhagoletis pomonella species complex: inferences from DNA sequence data. J. Evol. Biol . 21: 900– 913. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Insect Systematics and Diversity Oxford University Press

Phylogeography of Walnut-Infesting Rhagoletis suavis (Diptera: Tephritidae) Flies

Loading next page...
 
/lp/ou_press/phylogeography-of-walnut-infesting-rhagoletis-suavis-diptera-iFoUrieOA0
Publisher
Entomological Society of America
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
eISSN
2399-3421
D.O.I.
10.1093/isd/ixy002
Publisher site
See Article on Publisher Site

Abstract

Abstract Discerning the biogeography and historical ranges of organisms is important to understanding the processes causing population divergence and speciation. Mountainous regions in North America have contributed to widespread divergence within animals and plants as species become geographically isolated and diverge. Here, we investigate patterns of divergence for six species of walnut-infesting flies in the Rhagoletis suavis species group (Diptera: Tephritidae) in the United States and Mexico based on sequence analysis of mitochondrial DNA (mtDNA) encoded Cytochrome Oxidase I and II genes (COI and COII). We resolved the relationship of the newly described species, Rhagoletis ramosae, found in the highlands of Mexico, within the R. suavis species group to test alternate hypotheses of migration and divergence. Bayesian phylogenetic analysis supported a clade in which R. ramosae was most closely related to Rhagoletis zoqui/Rhagoletis completa, found in eastern Mexico. This implies that the Sierra Madre Oriental and not the Sierra Madre Occidental have been a major conduit of migration, isolation, and speciation for walnut flies between Mexico and the United States. Comparisons of mtDNA divergence for R. suavis group flies with the Rhagoletis pomonella and Rhagoletis cingulata species groups suggested that despite current similarities in geographic distributions, these taxa do not share a common biogeographic history, diverging in different regions at different times in the past. Patterns displayed by Rhagoletis flies can be compared to patterns seen in other organisms through the Southwestern and Eastern United States, and Mexico to develop a fuller understanding of the biogeography of these regions of North America. Allopatry, the geographic separation of populations (Feder 2017), is often an important factor facilitating population divergence leading to speciation in many animals and plants (Mayr 1963, Coyne and Orr 2004, Nosil 2012). When populations reside in different areas with little or no gene flow between them, they can accumulate genetic differences independent from one another. Given enough time, the isolated populations can eventually accrue enough differences to become reproductively isolated and diverge into separate species. Understanding the geographic distribution of species through time and space (biogeography) is therefore a key element to discerning the processes giving rise to new biodiversity (Klicka and Zink 1997, Avise and Walker 1998, Avise et al. 1998). Knowing the historic distributions of taxa can also be important for discerning how climate and ecology interact with spatial isolation and local adaptation to affect dispersal and population divergence (Lieberman 2003). The genus Rhagoletis (Diptera: Tephritidae) contains a number of different species groups native to North America that have contributed to our understanding of the roles that geography and ecology can play in the speciation process (Bush 1966, Bush and Smith 1998, Feder et al. 2003). The Rhagoletis suavis species group includes six species (Rhagoletis suavis (Loew), Rhagoletis juglandis (Cresson), Rhagoletis completa (Cresson), Rhagoletis boycei (Cresson), Rhagoletis zoqui (Bush), and Rhagoletis ramosae (Hernández-Ortiz)) (Fig. 1A), which are currently largely geographically isolated from one another in the United States and Mexico (Fig. 1B). R. suavis flies exploit their host plants by infesting the husks of walnuts (Juglans spp. (Fagales: Juglandaceae)) (Fig. 1C) and are thought to have undergone allopatric speciation (Bush 1966), which is different from most Rhagoletis species groups that overlap geographically. Walnut flies also differ from other Rhagoletis species groups in that the species are morphologically distinct from one another in wing pattern and body coloration and display different courtship behaviors, which is thought to be associated with sexual selection (Bush 1966, Bush and Smith 1998). Some walnut fly species currently overlap with one another in a portion of their geographic ranges, likely the result of secondary contact (Bush 1966, Bush and Smith 1998). In the southwestern United States, R. juglandis co-occurs in part of its distribution with R. boycei, while in the east it co-exists with R. completa (Fig. 1B). In the midwestern United States, R. completa and R. suavis partially overlap, while in Mexico R. completa and R. zoqui co-occur in a narrow region in the area of Acahuales. A hybrid zone has been shown to exist between two species, R. completa and R. zoqui, in Acahuales, Mexico (see star in Fig. 1B; Rull et al. 2012). No other hybrid zone has yet been found between any other pair of overlapping R. suavis species in nature. Fig. 1. View large Download slide Phylogenetic relationships of R. suavis group flies, their geographic distributions, and the ranges of their walnut host trees in North America. (A) mtDNA cladogram of the R. suavis group showing relationships resolved from previous studies (Smith and Bush 1997, Bush and Smith 1998). The relationship of R. ramosae in the earlier studies is depicted as unknown by the dotted line and question mark (B) Species distributions of walnut-infesting Rhagoletis in the United States and Mexico. Site locations for flies sequenced for COI and COII are distributed across distributions of each species, and numerical designations for sites analyzed in the study are shown (Supplementary Table S2). The hybrid zone between R. completa and R. zoqui found in Acahuales, Tamaulipas, Mexico is indicated by a star; (C) Species distributions of walnuts (Juglans spp.). We used mtDNA sequencing to determine the primary connectivity of walnut flies with alternative hypotheses that divergence occurred to the east along the Sierra Madre Oriental Mountains (solid arrow = SMOr) or in the west through the Sierra Madre Occidental Mountains (dotted arrow = SMOc). Fig. 1. View large Download slide Phylogenetic relationships of R. suavis group flies, their geographic distributions, and the ranges of their walnut host trees in North America. (A) mtDNA cladogram of the R. suavis group showing relationships resolved from previous studies (Smith and Bush 1997, Bush and Smith 1998). The relationship of R. ramosae in the earlier studies is depicted as unknown by the dotted line and question mark (B) Species distributions of walnut-infesting Rhagoletis in the United States and Mexico. Site locations for flies sequenced for COI and COII are distributed across distributions of each species, and numerical designations for sites analyzed in the study are shown (Supplementary Table S2). The hybrid zone between R. completa and R. zoqui found in Acahuales, Tamaulipas, Mexico is indicated by a star; (C) Species distributions of walnuts (Juglans spp.). We used mtDNA sequencing to determine the primary connectivity of walnut flies with alternative hypotheses that divergence occurred to the east along the Sierra Madre Oriental Mountains (solid arrow = SMOr) or in the west through the Sierra Madre Occidental Mountains (dotted arrow = SMOc). Except for R. ramosae, the phylogenetic relationships of R. suavis group flies in the United States and Mexico have been determined based on mitochondrial DNA (mtDNA) (Fig. 1A) (Smith and Bush 1997, Rull et al. 2013). R. ramosae is native to the Eje Volcánico Trans Mexicano (the central Altiplano or trans-Mexican volcanic belt; abbreviated EVTM hereafter) (Rull et al. 2013). The EVTM in the highlands of Mexico is a hotspot of biodiversity because of the mountainous terrain and volcanic regions that separate populations (Knowles 2001, Bryson et al. 2011a , Domínguez-Domínguez et al. 2011). Pleistocene glacial–interglacial climatic cycles, and the resultant expansions and contractions of the Mexican pine–oak forest, have driven widespread divergence within plants (Luna Vega et al. 1999, Jaramillo-Correa et al. 2008), scarabs (Lobo and Halffter 2000), reptiles, amphibians (Bryson et al. 2011b), and mammals (Ceballos et al. 2010). From previous mtDNA phylogenetic analysis, two scenarios were proposed by Rull et al. (2013) regarding R. ramosae (Fig. 1C). The first hypothesis is that R. ramosae is most closely related to R. boycei or R. juglandis in the southwestern United States. If this hypothesis is true, then an important axis of past gene flow and current geographic isolation for walnut flies occurred in is the Sierra Madre Occidental Mountains (SMOc) in western Mexico, resulting in the divergence of R. ramosae in the south and R. boycei or R. juglandis to the north. The second hypothesis is that R. ramosae is most closely related to R. completa or R. zoqui. In this case, the divergence of R. ramosae could be attributed to an ~90 km gap in suitable Rhagoletis habitat that exists between the EVTM and Sierra Madre Oriental Mountains (SMOr) in east-central Mexico, potentially corresponding to the location and time of divergence in other species groups. Resolving the relationship of R. ramosae to other R. suavis taxa can allow us to compare the speciation history of the R. suavis species group to other Rhagoletis flies in the highlands of Mexico. For example, host races of the species Rhagoletis pomonella (Walsh) are also present throughout the United States and Mexico where they attack the fruit of several different species of hawthorns (Crataegus spp. (Rosales: Rosaceae)), as well as domesticated apple (Malus domestica (Miller) (Rosales: Rosaceae) (Bush 1969). Unlike walnut flies, R. pomonella has been proposed to speciate in the absence of complete geographic isolation in the process of shifting and ecologically adapting to new host plants (Bush 1969, Berlocher and Feder 2002). An isolated population of hawthorn-infesting R. pomonella exists in the EVTM and is estimated to have become separated from flies in the SMOr over one million years ago (Feder et al. 2003, 2005; Michel et al. 2007). It has been hypothesized that periodic gene flow from the EVTM into the SMOr and more northern United States populations of hawthorn flies played a role in the radiation of the R. pomonella group onto a number of novel hosts with differing fruiting times (Feder et al. 2003, 2005; Xie et al. 2007). By comparing the estimates of divergence time for R. ramosae with hawthorn-infesting R. pomonella in the EVTM, we can infer if the two flies became isolated by an historically shared or different geographic isolating event(s). The objective of the current study was to resolve the biogeography of the R. suavis species group through DNA sequence analysis of a portion of the mitochondrial genome of these flies. We sequenced the Cytochrome Oxidase subunit I and II (COI and COII) mtDNA genes from 33 individuals from the R. suavis species group and the outgroup taxon, Rhagoletis cingulata (Loew) (Diptera:Tephritidae), to determine the phylogenetic relationship of R. ramosae to the five other members of the R. suavis group and test the two hypotheses of migration proposed by Rull et al. (2013). We also compared the divergence pattern of the R. suavis flies to other species groups of Rhagoletis in North America to determine whether any similarities that exist may be due to a shared biogeographic history of isolation. Methods Species Collections Walnuts infested with Rhagoletis fly larvae were collected from 1983 to 2012 from a total of 16 sites across the United States and Mexico (Table 1, Fig. 1B) and taken back to the laboratory and held on wire mesh racks over plastic collecting trays in a greenhouse experiencing ambient environmental conditions. Trays were monitored on a daily basis, and after larvae fed within the husks of walnuts, left the fruit, and formed puparia in the trays, they were collected and stored in Petri dishes containing moist vermiculite. With the exception of samples collected in 2012, which were immediately frozen at −20°C as pupae, the Petri dishes containing pupae were stored at 4°C for 4–6 mo to simulate winter, after which time they were kept at 22°C in a constant temperature room and 14:10 h light:dark cycle until flies eclosed as adult and were frozen at −20°C for later genetic analysis. Table 1. Site locations, GPS coordinates, and collection information for walnut-infesting R. suavis species group flies and the outgroup, R. cingulata, sequenced for mtDNA ID  Location  Latitude  Longitude  Collector  boycei_1  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_2  Ft. Apache, AZ  34.0645  −109.9135  J.L. Feder  boycei_3  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_4  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_5  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_6  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_7  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_8  Sedona, AZ  34.9126  −111.7268  D. Howard  completa_1  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_2  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_3  Edmond, OK  35.6227  −96.4424  M.M. Glover  completa_4  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_5  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_6  Kearney, NE  40.6936  −99.0897  M.M. Glover  completa_7  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_8  Edmond, OK  35.6227  −96.4424  M.M. Glover  juglandis_1  Cibola, AZ  34.0242  −107.1303  J.L. Feder  juglandis_2  Carrizo, AZ  34.0007  −110.2639  J.L. Feder  juglandis_3  Pine, AZ  34.3846  −111.4547  J.L. Feder  juglandis_4  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  juglandis_5  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  ramosae_1  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_2  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_3  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  suavis_1  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_2  Blacksburg, VA  37.2255  −80.4259  D. Cochran  suavis_3  Buckner, MO  39.1342  −94.2011  R. Cook  suavis_4  Colfax, IA  41.6751  −93.2385  M.M. Glover  suavis_5  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_6  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_7  Blacksburg, VA  37.2255  −80.4259  D. Cochran  zoqui  Huamantla, Tlaxcala, Mexico  19.3261  −97.9254  J. Rull  cingulata  Coahuila, Nueva León, Mexico  25.8136  −100.3673  S.P. Egan  ID  Location  Latitude  Longitude  Collector  boycei_1  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_2  Ft. Apache, AZ  34.0645  −109.9135  J.L. Feder  boycei_3  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_4  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_5  Sedona, AZ  34.9126  −111.7268  D. Howard  boycei_6  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_7  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  boycei_8  Sedona, AZ  34.9126  −111.7268  D. Howard  completa_1  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_2  Sacramento Mountains, NM  32.9149  −105.3547  J.L. Feder  completa_3  Edmond, OK  35.6227  −96.4424  M.M. Glover  completa_4  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_5  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_6  Kearney, NE  40.6936  −99.0897  M.M. Glover  completa_7  Lawrence, KS  38.9747  −95.2467  R. Cook  completa_8  Edmond, OK  35.6227  −96.4424  M.M. Glover  juglandis_1  Cibola, AZ  34.0242  −107.1303  J.L. Feder  juglandis_2  Carrizo, AZ  34.0007  −110.2639  J.L. Feder  juglandis_3  Pine, AZ  34.3846  −111.4547  J.L. Feder  juglandis_4  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  juglandis_5  Chiricahua Mountains, AZ  31.9344  −109.2844  J.L. Feder  ramosae_1  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_2  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  ramosae_3  Taxco, Guerrero, Mexico  18.5616  −99.6048  J. Rull  suavis_1  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_2  Blacksburg, VA  37.2255  −80.4259  D. Cochran  suavis_3  Buckner, MO  39.1342  −94.2011  R. Cook  suavis_4  Colfax, IA  41.6751  −93.2385  M.M. Glover  suavis_5  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_6  Mt. Clemens, MI  42.5861  −82.8684  R. Cook  suavis_7  Blacksburg, VA  37.2255  −80.4259  D. Cochran  zoqui  Huamantla, Tlaxcala, Mexico  19.3261  −97.9254  J. Rull  cingulata  Coahuila, Nueva León, Mexico  25.8136  −100.3673  S.P. Egan  ID = species identification. View Large Mitochondrial Sequencing Total genomic and mtDNA was extracted from individual flies using the Puregene DNA purification kit (Qiagen, Inc. Hilden, Germany). The polymerase chain reaction (PCR) was performed on purified DNA isolated from individual flies to amplify a portion of the mtDNA COI and COII genes. COI was amplified using the primer pairs Lep1F (5′-ATTCAACCAAT CATAAAGATATTGG-3′) and Lep1R (5′-TAAACTTCTGGATGTCCAAAAAATCA-3′) (Hebert et al. 2004). COII was amplified using either the primer pairs C1-J-2792 (George: 5′-ATACCTCGACGTTATTCAGA-3′) and TK-N-3772 (Eva: 5′-GAGACCAT TACTTGCTTTCAGTCATCT-3′) (Simons et al. 1994) or the newly designed R. suavis group specific forward (5′-CAACTTACATTCTTTCATGATC-3′) and reverse (5′-GAATTTACTCTATTTGTAATTC-3′) primers, which are internal to George and Eva primers. George and Eva primers did not amplify universally across the R. suavis species group, potentially due to a mutation in the primer site; thus, the analysis was only done on a fragment internal to these primers. PCR was performed with an initial denature step of 94°C for 5 min, followed by 35 cycles of 94°C for 20 s, 50°C/56°C for COI and COII, respectively, for 20 s, and 72°C for 30 s, and a final elongation step of 72°C for 10 min. Following SAP/exo cleanup, the PCR products were sequenced via Sanger sequencing at the University of Arizona Genomics Core (Tuscon, AZ). Phylogenetic Analysis Mitochondrial DNA sequences generated for a total of 33 flies from the 16 walnut collecting sites and R. cingulata, a closely related cherry-infesting fly used as an outgroup, were cleaned and trimmed using SEQUENCHER (GeneCorp, Ann Arbor, MI). The sequences are available on GenBank (accession numbers: MF977833–MF977898). The resulting sequences were then aligned against one another using MUSCLE (Edgar 2004) and the alignment refined, truncating the ends of the sequences to only include sites with information at each base pair (bp) using GBlocks (Castresana 2000) to generate reads with a total length of 541 bp for COI and 409 bp for COII for phylogenetic analysis. jModelTest2 (Guindon and Gascuel 2003, Darriba et al. 2012) identified the Hasegawa, Kishino and Yano (1985) model with invariable sites and rate variation among sites (HKY + I + G) as the nucleotide substitution model having the highest likelihood for explaining the walnut fly mtDNA data based on the Akaike information criterion (AIC). The HKY + I + G model was subsequently employed in a Bayesian phylogenetic analysis with COI, COII, and COI and COII concatenated sequences in MrBayes v.3.2.6 (Huelsenbeck and Ronquist 2001, Ronquist et al. 2012). Bayesian analyzes were conducted with two independent runs each with four Markov chains (three heated and one cold) for one million generations, with a 25% burn in. Default uninformative prior settings were used. Branch lengths for the Bayesian consensus mtDNA trees were estimated with maximum likelihood in PhyML 3.0 based on the HKY + I + G substitution model (Guindon and Gascuel 2003, Guindon et al. 2010). The resulting mtDNA trees were visualized with R. cingulata rooted as the outgroup with the APE package in R v3.3.3 (Paradis et al. 2004, R Core Team, 2017). Genetic Divergence We estimated the divergence times of the COII gene from R. suavis species and R. pomonella, as COI sequences for other species groups are not available, using a Bayesian Markov Chain Monte Carlo (MCMC) approach. COII sequences for the R. pomonella species in the eastern United States (n = 4), SMOr (n = 2), and EVTM (n = 2) were retrieved from Xie et al. (2008) through GenBank (Table 2). The sequences were aligned with the walnut fly COII sequences and truncated to the same 409 bp length fragment. Bayesian MCMC was run independently five times with 10 million generations each using Beast 2.4.7 (Bouckaert et al. 2014, Drummond et al. 2006). We used a gamma site model with four category counts and an HKY DNA substitution model. We used a relaxed lognormal clock with a rate of 3.54% divergence, a common estimate for the mtDNA molecular clock in insects (Papadopoulou et al. 2010). The tree was estimated with a Yule model and priors on the R. suavis and R. pomonella groups to constrain for monophyletic clades for the species group. Independent runs were combined using LogCombiner and summary tree produced using TreeAnnotator with at 10% burn-in (Drummond and Rambaut 2007). Table 2. Accession numbers and sampling information for mtDNA COII sequences for R. pomonella used for estimate divergence time ID  GenBank accession  Species  Population  Host plant  pom_EVTM1  EU109153  R. pomonella  EVTM  Hawthorn  pom_EVTM2  EU109154  R. pomonella  EVTM  Hawthorn  pom_East_haw1  EU109156  R. pomonella  US-East  Hawthorn  pom_East_haw2  EU109157  R. pomonella  US-East  Apple  pom_East_haw3  EU109158  R. pomonella  US-East  Hawthorn  pom_East_apple  EU109159  R. pomonella  US-East  Hawthorn  pom_SMOr1  EU109160  R. pomonella  SMOr  Hawthorn  pom_SMOr2  EU109161  R. pomonella  SMOr  Hawthorn  ID  GenBank accession  Species  Population  Host plant  pom_EVTM1  EU109153  R. pomonella  EVTM  Hawthorn  pom_EVTM2  EU109154  R. pomonella  EVTM  Hawthorn  pom_East_haw1  EU109156  R. pomonella  US-East  Hawthorn  pom_East_haw2  EU109157  R. pomonella  US-East  Apple  pom_East_haw3  EU109158  R. pomonella  US-East  Hawthorn  pom_East_apple  EU109159  R. pomonella  US-East  Hawthorn  pom_SMOr1  EU109160  R. pomonella  SMOr  Hawthorn  pom_SMOr2  EU109161  R. pomonella  SMOr  Hawthorn  View Large Results Most clades in the R. suavis group were well supported in the concatenated COI and COII tree (Fig. 2). For example, R. suavis, R. juglandis, R. boycei, and the clade containing, R. completa, R. zoqui, and R. ramosae each formed monophyletic groups with Bayesian posterior probabilities (PP) ≥0.99 (PP = 1.0, 1.0, 1.0, and 0.99, respectively). However, the phylogenetic relationships among R. completa, R. zoqui, and R. ramosae remained unresolved. R. ramosae possessed two unique bp differences not found in any other walnut fly species, one in COI (bp position 354) and one in COII (bp position 158). These two autapomorphies characterized R. ramosae as a unique clade. However, R. completa and R. zoqui could not be clearly distinguished, making the branching order within the clade containing R. completa, R. zoqui, and R. ramosae uncertain and the group polyphyletic based on mtDNA haplotypes (Fig. 2). Regardless, the COI and COII gene tree indicated that R. ramosae is more closely related to R. zoqui/R. completa and not to R. suavis, R. juglandis, or R. boycei. Fig. 2. View largeDownload slide Bayesian consensus mtDNA tree for combined COI and COII, in the R. suavis species group. Gene tree topologies were estimated for the concatenated COI (541 bp) and COII (409) gene fragments using a Bayesian phylogenetic analysis, with the posterior probabilities (PP) shown for each node (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7). The branch lengths were estimated through maximum likelihood, using PhyML, with R. cingulata used as the outgroup. R. ramosae is most closely related to R. completa/R. zoqui, with the three species forming a monophyletic clade. The position of R. suavis and R. juglandis is reversed in the COI versus COII gene tree, with R. juglandis branching first, followed by R. suavis. Scale for legend = 0.01 substitutions/bp. Fig. 2. View largeDownload slide Bayesian consensus mtDNA tree for combined COI and COII, in the R. suavis species group. Gene tree topologies were estimated for the concatenated COI (541 bp) and COII (409) gene fragments using a Bayesian phylogenetic analysis, with the posterior probabilities (PP) shown for each node (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7). The branch lengths were estimated through maximum likelihood, using PhyML, with R. cingulata used as the outgroup. R. ramosae is most closely related to R. completa/R. zoqui, with the three species forming a monophyletic clade. The position of R. suavis and R. juglandis is reversed in the COI versus COII gene tree, with R. juglandis branching first, followed by R. suavis. Scale for legend = 0.01 substitutions/bp. The mtDNA Bayesian phylogeny for concatenated COI and COII sequences (Fig. 2, individual trees are shown in Supplementary Fig. S1) was congruent with previous analyses of the R. suavis group (Smith and Bush 1997, Bush and Smith 1998). The concatenated COI and COII gene tree had R. suavis as the first species to branch from the common ancestor of walnut flies, followed by R. juglandis speciating from a clade containing all the remaining taxa found in the southwestern United States and Mexico (Fig. 2). Rhagoletis boycei was the next species to become isolated and diverged in southwestern New Mexico and the southeastern and central portions of Arizona from the common ancestor of R. completa, R. zoqui, and R. ramosae (Fig. 2). Here, R. boycei is now found at higher elevations, while the co-occurring R. juglandis is present at lower elevations (Bush 1966). Finally, the clade containing R. completa, R. zoqui, and R. ramosae diversified from north to south, respectively, along an area including the SMOr and EVTM. The Bayesian tree estimates of divergence time followed the same pattern of speciation inferred for the walnut flies (Fig. 3). Estimated divergence times (and 95% highest posterior density, HPD) for the R. suavis, R. juglandis, and R. boycei clades were 0.749 (0.458–1.068), 0.635 (0.362–0.869), and 0.419 (0.240–0.619) MYA, respectively. We were unable to estimate divergence times among R. completa, R. zoqui, and R. ramosae due to lack of mtDNA resolution for this clade. However, given their low level of mtDNA differentiation, these taxa were inferred to form recently. Fig. 3. View largeDownload slide Bayesian estimates of divergence time for R. suavis and R. pomonella species. Levels of divergence were estimated for a 409 bp fragment of the COII gene for R. suavis species and R. pomonella in the eastern United States, SMOr, and the EVTM. Posterior probabilities for the tree topology are colored on the nodes (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7), with light blue bars indicating the 95% highest priority density for node ages. Fig. 3. View largeDownload slide Bayesian estimates of divergence time for R. suavis and R. pomonella species. Levels of divergence were estimated for a 409 bp fragment of the COII gene for R. suavis species and R. pomonella in the eastern United States, SMOr, and the EVTM. Posterior probabilities for the tree topology are colored on the nodes (black: PP ≤ 0.9, grey: 0.7 ≤ PP < 0.9, white: PP < 0.7), with light blue bars indicating the 95% highest priority density for node ages. The hawthorn-infesting population of R. pomonella in the EVTM formed a separate clade from the rest of the R. pomonella group with a divergence time of 0.395 (0.180–0.6322) MYA. The rest of the populations, including R. pomonella from the SMOr and apple and hawthorn-infesting populations in the eastern United States, did not form a monophyletic clade and exhibited recent divergence. Discussion The mtDNA phylogeny for COI and COII genes resolved the relationship of R. ramosae to the other members of the R. suavis group, indicating that R. ramosae is most closely related to R. completa and R. zoqui, differing only by two substitutions, one each in the COI and COII genes. These results suggest that R. ramosae in the EVTM diverged relatively recently from R. completa and R. zoqui in the SMOr. This is supported by the evidence of hybridization between R. ramosae and both R. completa and R. zoqui in a laboratory setting (Tadeo 2014). The fact that R. completa does not form a monophyletic clade could be due to gene flow and/or incomplete lineage sorting between the species in the SMOr. A hybrid zone between R. completa and R. zoqui in Acahuales, Mexico (Rull et al. 2012) supports the hypothesis that ongoing introgression may be contributing to the lack of mtDNA phylogenetic resolution for these two taxa. In addition, the finding that R. ramosae is most closely related to species found in the SMOr implies that the major geographic axis for differentiation of walnut flies in Mexico was along the SMOr and EVTM (solid arrow in Fig. 1C), and not the SMOc (dotted arrow in Fig. 1C). As discussed in Rull et al. (2013), these results suggest that divergence of these taxa likely occurred in conjunction with the separation of Juglans major (Heller) and J. mollis (Engelm.)/J. pyriformis (Liebmann) that exists in the SMOr. Thus, the SMOr Mountains may have been a corridor for past migration and isolation between Mexico and the southwestern and eastern United States for walnut fly populations. Mitochondrial DNA sequencing has its limitations for inferring phylogenetic relationships and in estimating divergence times between taxa (Rubinoff and Holland 2005). The current study only considered two mitochondrial genes, which may not give a complete picture of the genetics of speciation and its geographic context for the R. suavis group. Including nuclear genes in the study and increasing the scope of the genomic analysis through next-generation sequencing could provide greater resolution to the history of divergence for walnut flies. For example, mtDNA analysis showed low resolution in the clade containing R. completa, R. zoqui, and R. ramosae, but the phylogeny could be improved by increasing the number of loci. However, for the main objective of the current study, the mtDNA COI and COII genes were clearly sufficient in showing that R. ramosae is closely related to R. completa and R. zoqui and much more distantly related to the other walnut-infesting fly species in the group. Given the maternal inheritance of mtDNA, it is always possible that the close affinity of R. ramosae to R. completa and R. zoqui reflects a recent sweep of a mitochondrial haplotype through R. ramosae, potentially due to a cytoplasmic factor or associated endosymbiont, and not reflect the evolutionary relationship of the species to R. completa and R. zoqui. However, if this were the case, then the mtDNA haplotype and endosymbiont would have likely been transferred together to R. ramosae from one of the other taxa, as the acquisition of a cytoplasmic factor or endosymbiont alone would not result in R. ramosae displaying a close affinity to R. completa or R. zoqui. This would require hybridization between these species, which would support their close phylogenetic relationship. We found that R. suavis and R. pomonella, despite displaying broad similarities in their current geographic distributions through the eastern United States, EVTM, and SMOr, do not appear to share a common biogeographic history. First, both species groups are found in the EVTM, including the walnut fly, R. ramosae, and hawthorn-infesting R. pomonella (Rull et al. 2006, 2013). R. ramosae is relatively closely related to the other walnut-infesting flies R. completa and R. zoqui in the SMOr, differing by only one unique substitution in the COII gene (0.24% sequence divergence; Fig. 3). In comparison, hawthorn-infesting populations of R. pomonella in the EVTM are much more diverged from R. pomonella in the SMOr and eastern United States (Fig. 3), as also found in previous studies which show that R. pomonella in the EVTM are geographically isolated and genetically diverged from the R. pomonella group (Feder et al. 2003). R. pomonella in the EVTM are also post-zygotically isolated from flies in the EVTM, congruent with strong differentiation between the two populations (Rull et al. 2010). We estimated here that the population of R. pomonella in the EVTM diverged around 0.395 (95% HPD: 0.180–0.6322) MYA based on the COII gene (Fig. 3) Second, the R. suavis and R. pomonella groups both have populations in the SMOr and eastern United States but show different levels of divergence between the regions. For walnut flies, R. boycei, R. juglandis, and R. suavis show deep splits between the flies in the SMOr and EVTM, with estimated divergence times 0.419 (0.240–0.619), 0.635 (0.362–0.869), and 0.749 (0.458–1.068) MYA. In comparison, hawthorn-infesting populations of R. pomonella in the eastern United States and SMOr display no fixed difference in COII in the current or in a previous study, suggesting a close evolutionary relationship (Fig. 3; Xie et al. 2007). The cherry-infesting species R. cingulata, belonging to the R. cingulata group, also overlaps in its geographic distribution with R. suavis and R. pomonella species in North America. In particular, R. cingulata co-occurs with R. pomonella and R. suavis in the eastern United States and SMOr. However, cherry-infesting flies have not been found in the EVTM (Rull et al. 2011). Similar to R. pomonella, R. cingulata displays limited mtDNA divergence between the SMOr and the eastern United States, with approximately 0.34% divergence (H. Schuler, personal communication). Like walnut flies, populations of R. cingulata also exist in the southwestern United States, where R. pomonella is not native (Bush 1966). In contrast to the walnut fly species R. boycei and R. juglandis in the southwestern United States, which show deep divergence in COII relative to flies in Mexico, cherry-infesting R. cingulata displays no mtDNA divergence between the southwestern United States and SMOr (H. Schuler, personal Communication), possibly due to either gene flow or relatively recent isolation (Fig. 3). Thus, despite their broadly shared geographic distributions, walnut, hawthorn, and cherry-infesting flies have different patterns of genetic differentiation through the southwestern United States, eastern United States, and Mexico implying that they differ in the times of separation across North America. Except in the EVTM, the R. suavis group shows greater geographic mtDNA differentiation, and by inference past spatial isolation, than R. pomonella and R. cingulata flies across their common shared range. For example, the separation of walnut flies in the SMOr and in the eastern United States appears to have occurred historically earlier than that for hawthorn- and cherry-infesting flies from these two regions. It may be that the species R. suavis in the eastern United States represents a comparatively old range expansion from Mexico and the southwestern United States into the eastern United States, while the center of endemism for hawthorn- and cherry-infested flies may be the eastern United States itself. However, at the current time, we cannot rule out the alternate possibility that the walnut flies may have been originally native to the eastern United States and became isolated in the southwestern United States and Mexico after migrating from the east. Future nuclear DNA sequencing may resolve these issues, for example, by indicating where genomic diversity (allelic polymorphism) is greatest and where it is reduced, potentially allowing identification of the region of origin for walnut flies in North America. The general pattern of increased mtDNA differentiation among walnut flies is consistent with the hypothesis that members of the R. suavis group have speciated allopatrically in the absence of gene flow (Bush 1966). Although R. suavis group flies tend to attack different species of walnut as their major native hosts (Supplementary Table S1, Rull et al. 2013), host specificity of R. suavis flies does not appear to have contributed to their divergence. Each species in the R. suavis group is capable of attacking any walnut present in its range (Rull et al. 2013). However, the geographic ranges of the different walnuts do not overlap greatly in North America, and the ranges of flies closely follow those of their primary walnut host, generating geographic isolation (Fig. 1). We, therefore, argue that it is the largely disjunct nature of the distribution of walnuts rather than any biological characteristic of the walnut hosts themselves that contributed to the radiation of the R. suavis group. The timing of walnut speciation, however, greatly predates that of R. suavis husk flies (Rull et al. 2013). The origin of the walnut genus, Juglans, in North America (c. 60 MYA), and of many present-day walnut species (40–45 MYA), is much older than that for R. suavis group flies (Fjellstrom and Parfitt 1995). As a result, host walnuts imposed a populations structure of geographic isolation on flies, as colonists moved into new regions and became largely associated with a new walnut host, possibly related to range expansions and contractions occurring during glacial and inter-glacial periods. In contrast to walnut flies, speciation in the R. pomonella group and for certain R. cingulata flies has been hypothesized to occur via host shifting without complete geographic isolation. Thus, R. pomonella and R. cingulata populations infesting hawthorns and black cherries, respectively, may have experienced higher levels of gene flow when they diverged compared to walnut flies. In addition, it appears that R. pomonella populations in the SMOr and R. cingulata populations in the SMOr and southwestern United States may have been more recently contiguous with populations in the eastern United States than is the case with walnut flies, the major exception is the hawthorn-infesting population of R. pomonella in the EVTM (Rull et al. 2006). Thus, R. pomonella and R. cingulata generally display reduced mtDNA differentiation than walnut-infesting flies across North America. Rhagoletis flies are not the only organisms to be affected by the mountainous geography of the highlands of Mexico experiencing geographic isolation and subsequent divergence (Knowles 2001, McCormack et al. 2008, Ruiz-Sanchez and Specht 2013). The different mountain ranges in central Mexico, including the SMOr, SMOc, and EVTM, have been shown to structure populations with between mountain ranges or even in some cases along a mountain range, as is seen in Mexican stoneroller fish in the SMOc (Domínguez-Domínguez et al. 2011). Similar to R. pomonella which shows strong differentiation between the EVTM and SMOr, many taxa exhibit distinct breaks in populations between the different mountain ranges in Mexico. McCormack et al. (2008) reported high mtDNA sequence divergence between Mexican Jay populations inhabiting the SMOc, SMOr, and EVTM with speciation occurring 0.7 MYA. Gene flow is also reduced due to geographic isolation in the EVTM in the Mexican shrub, Nolina parviflora, resulting in strong differentiation and population structure in the different mountain ranges (Ruiz-Sanchez and Specht 2013). Further work is needed to determine how congruent the patterns and timing of diversification of these organisms are with Rhagoletis. Our results suggest that there is no universal biogeographical process that affected all Rhagoletis species groups the same and in unison. It will be interesting to determine how the differences displayed by Rhagoletis flies match patterns seen in other organisms in the southwestern and eastern United States and Mexico, to develop a fuller and deeper understanding of the biogeography of these regions of North America. Supplementary Data Supplementary data are available at Insect Systematics and Diversity online. Acknowledgments We thank M.M. Doellman, C. Tait, H. Schuler, P. Meyers, E. Hernandéz, E. Tadeo, E. Acosta, L.A. Miller, C.C. Xu, G. McLain, M. DiGaetano, and Q.D. Read for help with fly collections and rearing in this study. This work was supported by the University of Notre Dame Advanced Diagnostics & Therapeutics Initiative (S.P.E., J.L.F.) and Environmental Change Initiative (S.P.E., J.L.F.), the Instituto de Ecología, A.C. (INECOL) and the Campaña Nacional Contra las Moscas de la Fruta (Convenios INECOL-SENASICA-IICA/CONACOFI) (M.A.), a Rice University Huxley Faculty Fellowship to S.P.E., and NSF and USDA grants to S.P.E. and J.L.F. References Cited Avise, J. C., and D. E. Walker. 1998. Pleistocene phylogeographic effects on avian populations and the speciation process. Proc. R. Soc. B Biol. Sci . 265: 457– 463. Google Scholar CrossRef Search ADS   Avise, J. C., D. Walker, and G. C. Johns. 1998. Speciation durations and Pleistocene effects on vertebrate phylogeography. Proc. R. Soc. B Biol. Sci . 265: 1707– 1712. Google Scholar CrossRef Search ADS   Berlocher, S. H., and J. L. Feder. 2002. Sympatric speciation in phytophagous insects: moving beyond controversy? Annu. Rev. Entomol . 47: 773– 815. Google Scholar CrossRef Search ADS PubMed  Bouckaert, R., J. Heled, D. Kühnert, T. Vaughan, C-H. Wu, D. Xie, M. A. Suchard, A. Rambaut, and Drummond, A. J. 2014. BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput. Biol . 10: e1003537. Google Scholar CrossRef Search ADS PubMed  Bryson, R. W., R. W. Murphy, M. R. Graham, A. Lathrop, and D. Lazcano. 2011a. Ephemeral Pleistocene woodlands connect the dots for highland rattlesnakes of the Crotalus intermedius group. J. Biogeogr . 38: 2299– 2310. Google Scholar CrossRef Search ADS   Bryson, R. W., R. W. Murphy, A. Lathrop, and D. Lazcano-Villareal. 2011b. Evolutionary drivers of phylogeographical diversity in the highlands of Mexico: a case study of the Crotalus triseriatus species group of montane rattlesnakes. J. Biogeogr . 38: 697– 710. Google Scholar CrossRef Search ADS   Bush, G. L. 1966. The taxonomy, cytology, and evolution of the genus Rhagoletis in North America (Diptera, Tephritidae). Bull. Museum Comp. Zool . 134: 431– 562. Bush, G. L. 1969. Sympatric host race formation and speciation in frugivorous flies of the genus Rhagoletis (Diptera, Tephritidae). Evolution (N. Y) . 23: 237– 251. Bush, G. L., and J. J. Smith. 1998. The genetics and ecology of sympatric speciation: a case study. Res. Popul. Ecol. (Kyoto) . 40: 175– 187. Google Scholar CrossRef Search ADS   Castresana, J. 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol. Biol. Evol . 17: 540– 552. Google Scholar CrossRef Search ADS PubMed  Ceballos G., J. Arroyo-Cabrales, and E. Ponce. 2010. Effects of Pleistocene environmental changes on the distribution and community structure of the mammalian fauna of Mexico. Quat. Res. 73: 464– 473. Coyne, J. A., and H. A. Orr. 2004. Speciation . Sinauer Associates, Sunderland, MA. Darriba, D., G. L. Taboada, R. Doallo, and D. Posada. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods . 9: 772. Google Scholar CrossRef Search ADS PubMed  Domínguez-Domínguez, O., M. Vila, R. Pérez-Rodríguez, N. Remón, and I. Doadrio. 2011. Complex evolutionary history of the Mexican stoneroller Campostoma ornatum Girard, 1856 (Actinopterygii: Cyprinidae). BMC Evol. Biol . 11: 153. Google Scholar CrossRef Search ADS PubMed  Drummond, A. J. and A. Rambaut. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol . 7: 214. Google Scholar CrossRef Search ADS PubMed  Drummond, A. J., S. Y. W. Ho, M. J. Phillips and A. Rambaut. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biol . 4: e88 Google Scholar CrossRef Search ADS PubMed  Edgar, R. C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res . 32: 1792– 1797. Google Scholar CrossRef Search ADS PubMed  Feder, J. L. 2017. Allopatric. In B.D. Roitberg (ed.) Ref. Modul. Life Sci . Elsevier Press, Amsterdam, Netherlands. Google Scholar CrossRef Search ADS   Feder, J. L., S. H. Berlocher, J. B. Roethele, H. Dambroski, J. J. Smith, W. L. Perry, V. Gavrilovic, K. E. Filchak, J. Rull, and M. Aluja. 2003. Allopatric genetic origins for sympatric host-plant shifts and race formation in Rhagoletis. Proc. Natl. Acad. Sci. U. S. A . 100: 10314– 10319. Google Scholar CrossRef Search ADS PubMed  Feder, J. L., X. Xie, J. Rull, S. Velez, A. Forbes, B. Leung, H. Dambroski, K. E. Filchak, and M. Aluja. 2005. Mayr, Dobzhansky, and Bush and the complexities of sympatric speciation in Rhagoletis. Proc. Natl. Acad. Sci. U. S. A . 102 ( Suppl): 6573– 6580. Google Scholar CrossRef Search ADS PubMed  Fjellstrom, R. G., and D. E. Parfitt. 1995. Phylogenetic analysis and evolution of the genus Juglans (Juglandaceae) as determined from nuclear genome RFLPs. Plant Syst. Evol . 197: 19– 32. Google Scholar CrossRef Search ADS   Guindon, S., and O. Gascuel. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst. Biol . 52: 696– 704. Google Scholar CrossRef Search ADS PubMed  Guindon, S., J. F. Dufayard, V. Lefort, M. Anisimova, W. Hordijk, and O. Gascuel. 2010. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst. Biol . 59: 307– 321. Google Scholar CrossRef Search ADS PubMed  Hasegawa, M., H. Kishino, and T. Yano. 1985. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J. Mol. Ecol. 22: 160– 174. Hebert, P. D. N., E. H. Penton, J. M. Burns, D. H. Janzen, and W. Hallwachs. 2004. Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator. Proc. Natl. Acad. Sci. U. S. A . 101: 14812– 14817. Google Scholar CrossRef Search ADS PubMed  Huelsenbeck, J. P., and F. Ronquist. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics . 17: 754– 755. Google Scholar CrossRef Search ADS PubMed  Jaramillo-Correa, J. P., E. Aguierre-Planter, D. P. Khasa, L. E. Eguiarte, D. Piñero, G. R. Furnier, and J. Bousquet. 2008. Ancestry and divergence of subtropical montane forest isolates: molecular biogeography of the genus Abies (Pinaceae) in southern México and Guatemala. Mol. Ecol . 17: 2476– 2490. Google Scholar CrossRef Search ADS PubMed  Klicka, J., and R. M. Zink. 1997. The importance of recent ice ages in speciation: a failed paradigm. Science . 277: 1666– 1669. Google Scholar CrossRef Search ADS   Knowles, L. L. 2001. Genealogical portraits of speciation in montane grasshoppers (genus Melanoplus) from the sky islands of the Rocky Mountains. Proc. R. Soc. B Biol. Sci . 268: 319– 324. Google Scholar CrossRef Search ADS   Lieberman, B. S. 2003. Paleobiogeography: the relevance of fossils to biogeography. Annu. Rev. Ecol. Evol. Syst . 34: 51– 69. Google Scholar CrossRef Search ADS   Lobo, J.M., and G. Hallffter. 2000. Biogeographical and Ecological Factors Affecting the Altitudinal Variation of Mountainous Communities of Coprophagous Beetles (Coleoptera: Scarabaeoidea): a Comparative Study. Ann. Entomol. Soc. Am. 93: 115– 126. Luna Vega, I., A. O. Alcántara, O. D. Espinosa, and J. J. Morrone. 1999. Historical relationships of the Mexican cloud forests: a preliminary vicariance model applying par- simony analysis of endemicity to vascular plant taxa. J Biogeogr . 26: 1299– 1305. Google Scholar CrossRef Search ADS   Mayr, E. 1963. Animal Species and Evolution . Harvard University Press, Harvard, MA. Google Scholar CrossRef Search ADS   McCormack, J. E., A. T. Peterson, E. Bonaccorso, and T. B. Smith. 2008. Speciation in the highlands of Mexico: genetic and phenotypic divergence in the Mexican jay (Aphelocoma ultramarina). Mol. Ecol . 17: 2505– 2521. Google Scholar CrossRef Search ADS PubMed  Michel, A. P., J. Rull, M. Aluja, and J. L. Feder. 2007. The genetic structure of hawthorn-infesting Rhagoletis pomonella populations in Mexico: implications for sympatric host race formation. Mol. Ecol . 16: 2867– 2878. Google Scholar CrossRef Search ADS PubMed  Nosil, P. 2012. Ecological Speciation . Oxford University Press, Oxford, UK. Google Scholar CrossRef Search ADS   Papadopoulou, A., I. Anastasiou, and A. P. Vogler. 2010. Revisiting the insect mitochondrial molecular clock: the Mid-Aegean trench calibration. Mol. Biol. Evol . 27: 1659– 1672. Google Scholar CrossRef Search ADS PubMed  Paradis, E., J. Claude, and K. Strimmer. 2004. APE: analyses of phylogenetics and evolution in R language. Bioinformatics . 20: 289– 290. Google Scholar CrossRef Search ADS PubMed  R Core Development Team. 2017. R: a language and environment for statistical computing . R Foundation for Statistical Compution, Vienna, Austria. https://www.R-project.org Ronquist, F., M. Teslenko, P. van der Mark, D. L. Ayres, A. Darling, S. Hohna, B. Larget, L. Liu, M. A. Suchard, and J. P. Huelsenbeck. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst. Biol . 61: 539– 542. Google Scholar CrossRef Search ADS PubMed  Rubinoff, D. and B. S. Holland. 2005. Between two extremes: mitochondrial DNA is neither the panacea nor the nemesis of phylogenetic and taxonomic inference. Syst. Biol . 54: 952– 961. Google Scholar CrossRef Search ADS PubMed  Ruiz-Sanchez, E., and C. D. Specht. 2013. Influence of the geological history of the Trans-Mexican volcanic belt on the diversification of Nolina parviflora (Asparagaceae : Nolinoideae). J. Biogeogr . 40: 1336– 1347. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, J. L. Feder, and S. Berlocher. 2006. Distribution and host range of hawthorn-infesting Rhagoletis (Diptera:Tephritidae) in Mexico. Ann. Entomol. Soc. Am . 99: 662– 672. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, and J. L. Feder. 2010. Evolution of intrinsic reproductive isolation among four North American populations of Rhagoletis pomonella (Diptera: Tephritidae). Biol. J. Linn. Soc . 100: 213– 223. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, and J. L. Feder. 2011. Distribution and basic biology of black cherry-infesting Rhagoletis (Diptera: Tephritidae) in México. Ann. Entomol. Soc. Am . 104: 202– 211. Google Scholar CrossRef Search ADS   Rull, J., E. Tadeo, M. Aluja, L. Guillen, S. P. Egan, and J. L. Feder. 2012. Hybridization and sequential components of reproductive isolation between parapatric walnut-infesting sister species Rhagoletis completa and Rhagoletis zoqui. Biol. J. Linn. Soc . 107: 886– 898. Google Scholar CrossRef Search ADS   Rull, J., M. Aluja, E. Tadeo, L. Guillen, S. Egan, M. Glover, and J. L. Feder. 2013. Distribution, host plant affiliation, phenology, and phylogeny of walnut-infesting Rhagoletis flies (Diptera: Tephritidae) in Mexico. Biol. J. Linn. Soc . 110: 765– 779. Google Scholar CrossRef Search ADS   Simons, C., F. Frati, A. Beckenbach, B. Crespi, H. Liu, and P. Flook. 1994. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann. Entomol. Soc. Am . 87: 651– 701. Google Scholar CrossRef Search ADS   Smith, J. J., and G. L. Bush. 1997. Phylogeny of the genus Rhagoletis (Diptera: Tephritidae) inferred from DNA sequences of mitochondrial cytochrome oxidase II. Mol. Phylogenet. Evol . 7: 33– 43. Google Scholar CrossRef Search ADS PubMed  Tadeo, E. 2014. Diferenciación y aislamiento reproductivo entre especies de moscas del grupo suavis en México . PhD Thesis, Universidad Veracruzana. Xie, X., J. Rull, A. P. Michel, S. Velez, A. A. Forbes, N. F. Lobo, M. Aluja, and J. L. Feder. 2007. Hawthorn-infesting populations of Rhagoletis pomonella in Mexico and speciation mode plurality. Evolution (N. Y) . 61: 1091– 1105. Xie, X., A. P. Michel, D. Schwarz, J. Rull, S. Velez, A. A. Forbes, M. Aluja, and J. L. Feder. 2008. Radiation and divergence in the Rhagoletis pomonella species complex: inferences from DNA sequence data. J. Evol. Biol . 21: 900– 913. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

Insect Systematics and DiversityOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial