Logo of transbhomepageaboutsubmitalertseditorial board
Philos Trans R Soc Lond B Biol Sci. 2008 Oct 27; 363(1508): 3377–3390.
Published online 2008 Sep 2. doi:  10.1098/rstb.2008.0111
PMCID: PMC2607372

Diversification in a fluctuating island setting: rapid radiation of Ohomopterus ground beetles in the Japanese Islands


The Japanese Islands have been largely isolated from the East Asian mainland since the Early Pleistocene, allowing the diversification of endemic lineages. Here, we explore speciation rates and historical biogeography of the ground beetles of the subgenus Ohomopterus (genus Carabus) based on nuclear and mitochondrial gene sequences. Ohomopterus diverged into 15 species during the Pleistocene. The speciation rate was 1.92 Ma–1 and was particularly fast (2.37 Ma–1) in a group with highly divergent genitalia. Speciation occurred almost solely within Honshu, the largest island with complex geography. Species diversity is highest in central Honshu, where closely related species occur parapatrically and different-sized species co-occur. Range expansion of some species in the past has resulted in such species assemblages. Introgressive hybridization, at least for mitochondrial DNA, has occurred repeatedly between species in contact, but has not greatly disturbed species distinctness. Small-island populations of some species were separated from main-island populations only after the last glacial (or the last interglacial) period, indicating that island isolation had little role in speciation. Thus, the speciation and formation of the Ohomopterus assemblage occurred despite frequent opportunities for secondary contact and hybridization and the lack of persistent isolation. This radiation was achieved without substantial ecological differentiation, but with marked differentiation in mechanical agents of reproductive isolation (body size and genital morphology).

Keywords: beetles, Japan, divergence time, genital evolution, molecular phylogeny, speciation rate

1. Introduction

The Japanese Islands constitute a distinct system differing from highly isolated oceanic islands. Located close to the East Asian mainland, a rich biota has colonized the Japanese Islands, but many endemic lineages have evolved on the islands owing to the geographical complexity and wide range of climatic conditions along latitudinal and elevational gradients. The Japanese archipelago is based on an accretionary prism that formed at the edge of the East Asian mainland and existed as many fragmented islands in the Middle Miocene (Iijima & Tada 1990; Tada 1994; Yonekura et al. 2001). A land bridge connected eastern China and western Japan continuously from the Late Miocene to the end of the Pliocene (10–1.7 Ma). Thereafter, only a narrow land bridge between the Korean peninsula and western Japan occurred temporarily at glacial maxima during the Pleistocene (Tada 1994; Kitamura et al. 2001; Kitamura & Kimoto 2006). Thus, the current terrestrial fauna of the Japanese Islands originated from colonization events via the western connection at various times from the Late Miocene to the end of the Pleistocene, as evidenced by the fossil records of insects and mammals (Dobson & Kawamura 1998; Hayashi 2004). A large part of northern Japan (Hokkaido and northern Honshu) emerged after the Late Miocene, and Hokkaido and Honshu were separated by the deep Tsugaru Strait except during the Late Pleistocene glacial periods (Tada 1994; Yonekura et al. 2001). During the glacial period, most of Hokkaido and parts of northern Honshu were covered with boreal conifer forest and glaciers (Tsukada 1984), and the terrestrial biota was dominated by cool adapted immigrants from the north. In these areas, colonization by organisms adapted to a temperate climate might have occurred after the last glacial period.

Among endemic insects in the temperate zone of Japan, flightless ground beetles of the subgenus Ohomopterus of the genus Carabus exhibit marked diversity in body size and genital morphology (Ishikawa 1985, 1991; Takami 2000), as well as a typical assemblage pattern consisting of two or more different-sized species (Sota et al. 2000a). They are an intriguing group for evolutionary study. Since the 1980s, hybrid zones of various species pairs have been explored and the consequences of interspecific hybridization between parapatric species with divergent genitalia have been investigated (Kubota 1988; Kubota & Sota 1998; Sota et al. 2000c; Takami & Suzuki 2005; Ujiie et al. 2005). Interspecific body size differences in local species assemblages have been described and their possible effects on species coexistence investigated (Sota et al. 2000a). The role of genitalic divergence in mechanical reproductive isolation has also been documented (Sota & Kubota 1998; Usami et al. 2006; Takami et al. 2007).

Since the 1990s, molecular phylogenetic approaches have promoted further evolutionary studies of this group. The contradiction between a mitochondrial gene tree and morphological species (Su et al. 1996) prompted a test of incongruence between mitochondrial and nuclear gene trees, and morphological classification (Sota & Vogler 2001), phylogenetic reconstruction with multiple nuclear genes (Sota & Vogler 2003), and studies of interspecific mitochondrial introgression and intraspecific phylogeography based on mitochondrial data (Sota et al. 2001; Sota 2002; Nagata et al. 2007a,b), as well as allelic sequence diversity of nuclear genes (Sota & Sasabe 2006). Also, the function and evolution of the genital lock and key system have been explored in terms of sexual selection (Takami 2002, 2003, 2007; Takami & Sota 2007) and the genetic basis of exaggerated genital morphology has recently been investigated (Sasabe et al. 2007).

The Ohomopterus radiation lacks notable differentiation in food habits and associated morphology, a major aspect of adaptive radiation in other groups such as Galápagos finches and African lake cichlids (Schluter 2000). Instead, Ohomopterus represents a radiation driven by differentiation in mating traits (genital morphology and body size) due to local adaptation and sexual selection. Thus, studies of Ohomopterus will contribute to our understanding of various adaptive radiation pathways. In this paper, we give an overview of Ohomopterus speciation in the Japanese Islands and estimate speciation rates. We then describe genetic differentiation, diversity and introgression between species and among geographical populations within species using a large mitochondrial gene sequence dataset. We estimate divergence times for small-island from main-island populations. In doing so, we outline the historical processes of Ohomopterus diversification in the Japanese Islands.

2. Biology and biogeography of Ohomopterus

Ohomopterus is a subgenus of the genus Carabus (sensu stricto) of the holoarctic subtribe Carabina (=genus Carabus sensu lato) and consists of 15 species (Ishikawa 1985). Adult beetles are 18–35 mm in body length, flightless with degenerate hind wings and inhabit the forest floor and grasslands. They reproduce in spring and summer. The larvae develop during the summer and eclose as adults by the autumn, overwintering without reproduction (Sota 1985a,b). The adults are polyphagous, mainly feeding on earthworms, whereas the larvae are specialized predators of megascolecid earthworms (Sota 1985a,b). Among Megascolecidae in Japan, the genus Pheretima is the most abundant and diversified, with at least 124 described species (Ishizuka 2001). Thus, Ohomopterus populations depend on the presence of Pheretima.

Ohomopterus ranges from southern Kyushu to southern Hokkaido (31–44° N, 129–145° E; Sota et al. 2000a; figure 1a). The altitudinal range is 0–1750 m (occasionally reaching 2000 m, i.e. Carabus albrechti in central Honshu), with an annual mean temperature range of 5–15°C (occasionally 3°C for C. albrechti). Thus, Ohomopterus occurs widely in the temperate zones of Japan but cannot colonize subalpine coniferous forests and alpine zones. Among the four main islands, Honshu is the centre of diversity (14 species). Shikoku and Kyushu harbour four and two species, respectively, and only one species (C. albrechti) occurs in Hokkaido. Ohomopterus also occurs on other small islands (one to three species per island) but no species is endemic to such islands. Overall, species diversity is highest in central Honshu and low in both the southwest and northeast regions (figure 1b).

Figure 1
(a) The Japanese Islands and the distribution of Ohomopterus ground beetles. The coastline 20 000 years ago (the last glacial period) is based on Ohshima (1990) and Yonekura et al. (2001). (b) Number of species in each prefecture or island. The ...

In most of its range, two or more species with different body sizes co-occur (figure 1b; Sota et al. 2000a). The most species-rich assemblage, with five species, exists in a small mountain area of central Honshu. Carabus dehaanii is the largest species in an assemblage whenever it occurs, and Carabus japonicus, Carabus daisen and the species of the albrechti group are the smallest in assemblages with two or more species. Carabus yaconinus, Carabus tosanus and species of the iwawakianus–insulicola group are intermediate or the largest in an assemblage.

3. Origin and phylogeny of Ohomopterus

A molecular phylogeny of the subtribe Carabina based on nuclear gene sequences indicates that within Carabus (sensu stricto), Ohomopterus is sister to the subgenus Isiocarabus, which occurs in warm temperate regions of China and Korea (Sota & Ishikawa 2004). The ancestral Ohomopterus might have colonized Japan during the Late Miocene through to the Pliocene, when western Japan was connected to the East Asian mainland. A few fossilized elytra resembling Ohomopterus have been discovered from Late Miocene or Early Pliocene strata (8–6 Ma old; Hiura 1971; Hayashi 2001), but cannot be identified as extant Ohomopterus because their elytral sculpture is one of the common patterns of the present Carabina (i.e. triploid homodynamic). Fossils directly related to the present species have been discovered, mostly from Late Pleistocene strata (Hayashi 2001). After the beginning of the Pleistocene (1.7 Ma), the connection between Japan and the mainland was broken when the Korean (Tsushima) Strait opened (Kitamura et al. 2001; Kitamura & Kimoto 2006), so Ohomopterus probably diversified during the Pleistocene.

Reconstructing species relationships based on phylogenetic analyses of mitochondrial genes is difficult for Ohomopterus because mitochondrial genes show extensive trans-species polymorphisms. Nuclear genes provide phylogenetic information largely concordant with morphological species boundaries (Sota & Vogler 2001, 2003; Sota & Sasabe 2006). Here, we have expanded the nuclear data of Sota & Vogler (2003) to include 4164 base pairs (bp) from six genes (see table 1 for gene regions and primers). The specimens used were the same as those in table 1 of Sota & Vogler (2003) except for exclusion of a putative hybrid (Carabus insulicola pseudinsulicola) and inclusion of three Carabus maiyasanus specimens (sample codes MAI330, MAI332 and MAI349). For GenBank accession numbers of nuclear sequences see Sota & Vogler (2001, 2003); numbers for additional sequences are EU435018-EU435132.

Table 1
Nuclear genes used for phylogenetic analysis of Ohomopterus, with sequence length used in phylogenetic analyses and primers.

Although partly unresolved, the maximum-likelihood tree resulting from simultaneous analysis of these genes (figure 2; see legend for details of the analysis) reveals that Ohomopterus consists of three clades, the albrechti group (Ishikawa 1991), the iwawakianusinsulicola group (Sota & Vogler 2003) and a third group, named here the daisen group.

Figure 2
Species relationships in Ohomopterus with the ranges of the three species groups, male copulatory piece and gross morphology of male (left) and female (right) beetles. The tree is a maximum-likelihood (ML) tree resulting from simultaneous analysis of ...

The albrechti group consists of four small species distributed in central to northern Honshu and Hokkaido as well as Sado, Awa and Okushiri islands. The daisen group consists of five species distributed in Kyushu, Shikoku to central Honshu and on small islands in the western region. Body size is generally small in C. daisen and C. japonicus, intermediate in C. yaconinus and C. tosanus and the largest in C. dehaanii. The iwawakianusinsulicola group consists of six medium-sized species in central to northern Honshu. Ohomopterus exhibits high diversity in the morphologies of functional parts of the genitalia, especially the copulatory piece (a chitinized part of the endophallus) and its corresponding vaginal appendix. All the albrechti and daisen group species possess a small triangular copulatory piece (except C. yaconinus which has a pentagonal copulatory piece) and a correspondingly short vaginal appendix. The iwawakianus–insulicola group exhibits a small to highly elongate (hook like) copulatory piece with corresponding shapes of the vaginal appendix.

Of 15 species, eight are confined to the main island of Honshu and one is endemic to Shikoku. Therefore, allopatric differentiation on different islands was not the major speciation process. Speciation within Honshu is likely for C. yaconinus, the four species of the albrechti group and the six species of the iwawakianus–insulicola group. These speciation events might have been allopatric or parapatric; speciation would have been facilitated by population fragmentation, with rivers and mountains acting as dispersal barriers. Climatic fluctuation during the Pleistocene might also have affected the differentiation because populations of Ohomopterus might have been restricted to coastal refugia during glacial periods.

4. Speciation rate

Speciation rates in Ohomopterus can be estimated by converting the phylogenetic tree to an ultrametric tree with a calibration for absolute time. In our previous study, the divergence between Isiocarabus and Ohomopterus was estimated to have occurred 2.14 Ma based on mitochondrial NADH dehydrogenase subunit 5 (ND5) sequences, assuming that these sister groups diverged following the separation between the East Asian mainland and the western region of Japan 3.5–1.7 Ma (Nagata et al. 2007a). We used this age to calibrate the nodes of the nuclear gene tree (figure 2).

With the age of the most recent common ancestor of Ohomopterus at 1.4 Ma, the speciation interval of Ohomopterus (i.e. millions of years divided by logarithm of species number) is 0.52 Ma (speciation rate, 1.92 Ma−1). The values were 0.63 Ma (1.59 Ma−1) for the albrechti group, 0.85 Ma (1.18 Ma−1) for the daisen group and 0.42 Ma (2.37 Ma−1) for the iwawakianus–insulicola group.

5. Genetic diversity in a mitochondrial gene

The mitochondrial gene ND5 has served as a marker to explore the evolution of Ohomopterus, including intraspecific phylogeography and interspecific introgression (Sota et al. 2001; Nagata et al. 2007a,b). For this study, we compiled a large ND5 dataset to investigate the overall genetic diversity and analyse the historical aspects of differentiation in Ohomopterus.

Figure 3 shows a ND5 gene tree of 1463 haplotypes detected from 6136 specimens (table 2), in which the major clades are named clades A–F and subclades A1, A2 and so on. There were 515 variable sites among the 1020 nucleotide sites, yielding 642 different amino acid sequences. The gene tree showed extensive trans-species polymorphisms. Clades A–C were mostly members of the iwawakianus–insulicola group, but C. dehaanii and C. yaconinus possessed many haplotypes from the A and B clades. Clades D and E were of the daisen group. Clade F consisted of distinct subclades of C. japonicus (F4) and C. daisen (F2); other subclades consisted of haplotypes of the albrechti group and of Carabus insulicola (with three other species of the same group). Clade G contained C. tosanus, C. japonicus and C. dehaanii within the daisen group. The pattern of sharing of haplotype lineages revealed introgressive hybridization and incomplete sorting of ancestral polymorphisms (Sota & Vogler 2001; Sota et al. 2001; Sota 2002; Nagata et al. 2007a,b).

Figure 3
Phylogeny of mitochondrial ND5 and distribution of species among lineages. A neighbour-joining tree was constructed using PAUP* v. 4.10b (Swofford 2002) with a substitution model (GTR+I+G) and parameters selected by Modeltest v. 3.07 (Posada & ...
Table 2
Sample sizes, regional and intra-population nucleotide diversity, calculated using Arlequin v. 3.0 (Excoffier et al. 2005) of the mitochondrial ND5 gene sequence.

Figure 4 shows the pattern of sharing of haplotype clades among species. For a mitochondrial gene with a fast evolutionary rate, sharing of a haplotype between species in sympatry or parapatry probably indicates recent introgression. Also, the geographical distribution of haplotype clades within a species (figure 5) suggests that an introgressed haplotype or lineage probably occurs close to the contact zone with other species that share the haplotype or lineage (Sota 2002). Haplotype sharing was common in clade A and B haplotypes that appear to have been extensively introgressed into C. dehaanii and C. yaconinus. Although A4 is exclusively made up of C. dehaanii, it may have originated from an ancient introgression from the iwawakianus–insulicola group; A4, and A1 and A2 occurred only in eastern populations of C. dehaanii and are unlikely to be an original clade. F1a haplotypes, mostly from C. insulicola and occurring throughout its range, might have originated from ancient introgression from the albrechti group (see Sota et al. 2001). In fact, clade F1a includes a haplotype of C. albrechti. At the western margin of the C. insulicola range, F1a is replaced with another introgressant clade (B1) from C. arrowianus (Sota et al. 2001). Thus, C. insulicola may not have retained any original haplotypes.

Figure 4
Haplotype clade composition of each species based on the number of ND5 haplotypes (large circles) and the pattern of sharing of clades and haplotypes between sympatric/parapatric species (lines connecting species: dashed lines, no clade shared; thin solid ...
Figure 5
Distribution of ND5 lineages by prefecture and island. Coloured areas indicate distribution ranges and small open circles indicate sampling localities. The pie graph for each prefecture/island shows the haplotype clade composition among individuals sampled, ...

6. Dispersal of widely distributed species

For species with wide ranges, the direction of range expansion may be inferred from the distribution of mitochondrial haplotypes using the framework of nested clade phylogeographic analysis (Templeton 1998). Here we simply examined the geographical distribution of interior (ancestral) and tip (derived) clades within haplotype networks constructed with statistical parsimony analysis (Templeton et al. 1992; Clement et al. 2000).

Clade D (D1 and D2) haplotypes of C. japonicus dispersed from Kyushu to Honshu (Chugoku) and Shikoku (figures 5 and and66a). This species also possessed clade G3 haplotypes in Honshu (Chugoku) and Awaji Island, and a unique clade F4 haplotype from Tsushima Island. Thus, C. japonicus seems to have retained highly diverged original mitochondrial lineages. Clade D1 haplotypes of C. dehaanii have expanded eastwards from western Honshu or Kyushu, with clade D3 haplotypes confined to Kyushu (figures 5 and and66b); the abrupt replacement with clade A haplotypes in the east could have been caused by introgressive hybridization. The populations on Fukue Island and east Kyushu possess haplotypes of the internal clade within clade D, and these populations may retain ancestral haplotypes. Carabus dehaanii is the largest species in this subgenus and its range expansion resulted in formation of species assemblages with small- and medium-sized species; this species co-occurs with all but one species.

Figure 6
Statistical parsimony networks of ND5 haplotypes showing the geographical relationships between internal and tip clades. We inferred directions of dispersal as from the geographical ranges of interior clades to those of tip clades. (a) japonicus, (b) ...

Carabus albrechti and C. insulicola expanded northwards from central Honshu (figures 5 and and66c,d). In C. albrechti, clade F1 haplotypes expanded from Chubu to Tohoku from the Japan Sea side of Honshu, and from Chubu to Kanto and Tohoku from the Pacific side of Honshu. In C. insulicola, clade F (F1a) haplotypes spread from Chubu to central Kanto, and afterwards to Tohoku and the Japan Sea side of central Honshu. Clade B1, introgressed from C. arrowianus in central Chubu and spread northward and southeastward.

7. Divergence times of small-island populations

Ohomopterus populations inhabiting small islands adjacent to the main islands are often discriminated as subspecies. We estimated the divergence time from main-island populations of nine populations on six islands using ND5 (table 3). The time of the last connection between a small island and a main island is not always clear, based on the geological evidence, as is the case for Tsushima and Sado Islands and Hokkaido, which are separated by deep straits. However, Iki Island was probably separated from Kyushu at the last postglacial transgression approximately 15 000 years ago (Ohshima 1990; Machida et al. 2001). Iki is a small volcanic island and the last eruption occurred 0.9–0.6 Ma; it is only 21 km from Kyushu and they were connected during the last glacial period. Therefore, it is reasonable to assume that the divergence of C. japonicus (the sole species on Iki) between Iki and Kyushu started 15 000 years ago.

Table 3
Estimated times of the last separation of small-island populations from main-island populations. Geographical distance and geological separation time (Ohshima 1990) are given in square brackets.

In each of the nine cases, the divergence time between the small island and the main island was estimated by a coalescent simulation analysis using IMa (Hey & Nielsen 2007). Although this program allows migration between populations after the split, this was assumed not to have occurred between populations that are currently separated by the sea. The time calibration was performed assuming that the estimated divergence time in C. japonicus between Iki and Kyushu was 15 000 years. To avoid the effect of introgressed haplotypes on the estimation, the main-island population used in the calculation consisted of populations adjacent to the small island and not including putative introgressed haplotypes.

Divergence times estimated by the IMa analysis fell in the Late Pleistocene to Holocene (75–1.4 ka ago; table 3), suggesting that small-island (except Tsushima Island) populations separated after the Last Glacial Maximum. Around Kyushu, the Goto Islands, including Fukue Island, were last connected to Kyushu during the last glacial period; divergence of C. japonicus and C. dehaanii was 15 and 8 ka ago, respectively. Tsushima is far (88 km) from Kyushu and may have been separated 100 ka ago (Ohshima 1990); the divergence of C. japonicus 75 ka ago (the last interglacial) is congruent with geohistory. The Oki Islands were connected to Honshu (Chugoku district) during the last glacial period and divergence of C. yaconinus (9 ka ago) was consistent with this. However, divergence of another Oki Island inhabitant, C. daisen, is earlier (23 ka ago), the difference perhaps attributable to different habitat uses; C. yaconinus inhabits both floodplains and mountains and could probably cross the plain between Oki and Honshu, whereas C. daisen is confined to mountain areas and its dispersal was limited. Note that the haplotype lineages of Oki and Honshu populations of C. daisen differed completely and the haplotype composition of Honshu populations was highly heterogeneous, possibly causing the uncertainty in divergence time estimates.

Sado Island has been separated from Honshu since the Middle Pleistocene according to geological inference (Ohshima 1990). However, divergence of C. albrechti dates approximately to the Last Glacial Maximum (18 ka ago), suggesting that gene flow occurred at the glacial maximum. Also, the estimated time of divergence for C. insulicola is quite recent at 1.4 ka ago. This suggests colonization as a result of flushing from the big river opposite to the island on Honshu. The occurrence of C. insulicola on Izu-Ohshima Island off the Pacific coast of Honshu might result from a similar cause, but two recently discovered populations in Hokkaido are probably due to anthropogenic accidental transportation.

The divergence time for C. albrechti between Honshu and Hokkaido was also recent (6 ka ago; in the Holocene). Although a land bridge might not have appeared over the deep Tsugaru Strait during the last glacial period (Ohshima 1990; Koaze et al. 2003), the valley between Honshu and Hokkaido was narrow and might have allowed exchange of terrestrial fauna. Our estimate suggests that the Hokkaido population was established at or after the last glacial period. The C. albrechti population in Hokkaido possesses a common haplotype that also occurs in northern Honshu, and from which all other haplotypes are derived. Hokkaido was mostly covered with boreal conifer forest (otherwise tundra and glacier) during the last glacial period (Tsukada 1984), a unsuitable habitat for Ohomopterus. Therefore, the present Hokkaido population probably originated from a small refugium in southern Hokkaido or northern Honshu.

Overall, gene flow between small islands and main islands occurred until recently (later than the early Late Pleistocene). However, the single species populations on Iki Island and Hokkaido exhibit larger body sizes than main-island populations coexisting with larger species, possibly related to competitive release (Sota et al. 2000a). The short duration of segregation was therefore sufficient for ecologically significant evolution.

8. Discussion

We examined a large dataset of mitochondrial ND5 gene sequences to determine the pattern of diversification of Ohomopterus ground beetles endemic to the Japanese Islands. Haplotype diversity was high and there was extensive trans-species polymorphism. Our study sets out an entire scheme of mitochondrial diversification in Ohomopterus that was partially reported previously (Sota et al. 2001; Sota 2002; Nagata et al. 2007a,b). Initial diversification of Ohomopterus occurred in the central region of Honshu, which harbours all but one of the 15 species. Shikoku and Kyushu are separated from Honshu only by narrow, shallow channels. Separation of the Shikoku Ohomopterus population resulted in the origin of C. tosanus, and the populations of C. dehaanii and C. japonicus on Kyushu may be ancestral to those on Honshu because eastward range expansion was inferred for these species. Except for these species, speciation probably occurred only in central Honshu. Our divergence time estimates suggest that differentiation of small-island populations occurred recently, from the Late Pleistocene through the Holocene, resulting in minor morphological differentiation at the most. Thus, isolation on small islands or peripheral differentiation seems to have played only a minor role in speciation in this subgenus.

In central Honshu, large populations of Ohomopterus have lived in favourable temperate habitats (warm and wet) separated by rivers and mountains; allopatric or parapatric differentiation has occurred repeatedly, resulting in parapatry of several species. The parapatric species can hybridize at their contact zones, sometimes resulting in the formation of hybrid swarms, and mitochondrial introgression has occurred repeatedly. Nevertheless, interspecific gene flow is limited, and species boundaries are maintained by geographical barriers. Introgressive hybridization has also occurred between sympatric species but is largely prevented by differences in the genital and body sizes (Nagata et al. 2007b). By contrast, divergence among populations within species is ongoing, for instance, in polytypic C. arrowianus in central Honshu (N. Nagata, K. Kubota, Y. Takami & T. Sota 2008, unpublished data). In this species, allopatric divergence of both body and genital length occurred in the latest Middle Pleistocene with limited gene flow between separate populations. Thus, diversification and speciation of Ohomopterus has occurred without persistent isolation and despite frequent opportunities for secondary contact and hybridization under the fluctuating geographical conditions during the Pleistocene.

The estimated speciation rates of Ohomopterus are among the fastest of any animal group (Coyne & Orr 2004), with a speciation interval of less than 1 Ma. A high speciation rate was reported in the Hawaiian cricket genus Laupala (Mendelson & Shaw 2005) that diverged into 21 species in 3.7 Ma (0.82 species Ma–1). Laupala are dietary generalists and thus exhibit little ecological differentiation; they are differentiated primarily by secondary sexual traits, suggesting rapid speciation by sexual selection. The highest known speciation rate was in a monophyletic clade on the island of Hawaii (six species in 0.43 Ma; 4.17 species Ma–1). The case of Ohomopterus parallels that of Laupala, in that the diversification of traits for reproductive isolation (genital morphology) due to local sexual selection might have resulted in rapid speciation. Speciation might be particularly fast in the iwawakianus–insulicola group on Honshu, as it exhibits marked divergence in genital morphology (figure 2). The Ohomopterus species used here are as defined morphologically by Ishikawa (1985, 1991). However, some consist of distinct populations (subspecies) with different body sizes or genital morphologies, which may be reproductively incompatible with each other. For instance, C. tosanus has been divided into two species by Imura & Mizusawa (1996) because there are contact zones between distinct populations with divergent body sizes. Thus, the current species may be further divided into additional biological species and our estimates of speciation rates are regarded as minimum rates.

Speciation of Ohomopterus was achieved without major ecological differentiation such as diet, but with divergence of body size and genital morphology, which facilitates mechanical isolation (Sota & Kubota 1998; Nagata et al. 2007b). It is tempting to conclude that body size differences among species would result in resource partitioning in terms of prey size and facilitate species coexistence, implying the possibility of ecological speciation as a by-product of resource partitioning. In fact, the body size of larvae, which is correlated with that of adults, is related to the efficiency of predation on different-sized earthworms (Y. Okuzaki & T. Sota 2008, unpublished data), and hence different-sized species may depend primarily on different prey size classes. However, owing to the wide, overlapping range of prey sizes that different-sized larvae can attack and consume, effective resource partitioning in prey size is not likely to occur. Compared to the density-dependent strength of resource competition, sexual competition (via interspecific mating) is frequency dependent and has a strong effect on demography (i.e. competitive exclusion; Ribeiro & Spielman 1986; Kuno 1992; Yoshimura & Clark 1994). Therefore, whereas vague resource partitioning might result from body size differences, sexual isolation due to body size differences will have more direct and hence ubiquitous effects on species coexistence.

The cause and process of divergence in body size and genital morphology need to be clarified. Because intraspecific body size variation in Ohomopterus is clinal on a temperature gradient (i.e. the converse of Bergmann's rule; Masaki 1967; Roff 1980), climatic adaptation must be involved in body size variation. However, the differences among sympatric species cannot be explained by climatic adaptation. The fact that island individuals of C. japonicus and C. albrechti in the absence of co-occurring species are larger than those in populations with sympatric species suggests an effect of interspecific interactions (Sota et al. 2000a). A possible scenario is that the body size difference originated from local adaptation. Rapid dispersal of different-sized species during a glacial/interglacial cycle then resulted in secondary contact; and the difference has been maintained (or reinforced) by selection related to sexual (and resource) competition (i.e. this process implies parapatric speciation along a climatic gradient; see also Konuma & Chiba 2007). This explanation is similar to that of Sota et al. (2000b) for a geographical cline of interspecific body size of Leptocarabus on Japanese Islands, mostly on Honshu. This hypothetical process could be tested using intraspecific geographical variation in body size; species such as C. tosanus show steep altitudinal clines in body size and populations of different body size exist in an area. Selection on body size is not simple, and disentangling the components of adaptation to seasonality in temperature and food availability and intra- and interspecific sexual interactions would be necessary.

The evolution of genital morphology, however, is independent of environmental gradients and governed by within-population selection, most likely sexual selection (Eberhard 1985; Arnqvist 1998; Hosken & Stockley 2004). Notable changes in genital morphology have occurred mostly within the iwawakianusinsulicola group. A quantitative genetic analysis suggests that species differences in the dimensions of male and female genital parts (i.e. copulatory piece and vaginal appendix) are determined by a limited number of loci (Sasabe et al. 2007). The selective advantage of the genital lock and key system and that of the exaggerated genital parts is not fully understood. Strict matching of genital parts between sexes ensures insemination (Takami 2003) and hence is beneficial to males in particular. Elongation of male genital parts can improve a male's ability to remove a rival's spermatophore and be advantageous in sperm competition (Takami 2007), although other factors such as sexual conflict might explain the arms race of male and female genital evolution (Sota 2002).

In conclusion, the divergence of Ohomopterus beetles occurred with varying fragmentation of favourable habitats by geographical barriers, evolution of body size and genital morphology in local populations, and secondary contact and interactions between diverging populations. Resolving the details of the whole process and the genetic architecture of the key traits for reproductive isolation will greatly improve our understanding of the origin of species and local species richness.


We thank R. Ishikawa, K. Kubota, M. Ujiie and Y. Takami for their long-lasting collaboration in the study of Ohomopterus. We also thank many people who helped us with the sample collection, especially K. Yahiro, T. Dejima and H. Nishi; and R. H. Cowie and two anonymous reviewers for their helpful comments on the manuscript. This study was partly supported by grants-in-aid from the Japan Society for the Promotion of Science (nos. 15207004, 17405007 and 2037011), the Japan Science Society (Sasakawa Scientific Research Grant 16-274), Lake Biwa Museum, and Ministry of Education, Culture, Sports, Science and Technology, Japan (Global Centre of Excellence Programme ‘Formation of a Strategic Base for Biodiversity and Evolutionary Research: from Genome to Ecosystem’).


One contribution of 15 to a Theme Issue ‘Evolution on Pacific islands: Darwin's legacy’.


  • Arnqvist G. Comparative evidence for the evolution of genitalia by sexual selection. Nature. 1998;393:784–786. doi:10.1038/31689
  • Britton T, Anderson C.L, Jacquet D, Lundqvist S, Bremer K. Estimating divergence times in large phylogenetic trees. Syst. Biol. 2007;56:741–752. doi:10.1080/10635150701613783 [PubMed]
  • Clement M, Posada D, Crandall K.A. TCS: a computer program to estimate gene genealogies. Mol. Ecol. 2000;9:1657–1660. doi:10.1046/j.1365-294x.2000.01020.x [PubMed]
  • Coyne J.A, Orr H.A. Sinauer; Sunderland, MA: 2004. Speciation.
  • Dobson M, Kawamura Y. Origin of the Japanese land mammal fauna: allocaton of extant species to historically-based categories. Quat. Res. (Daiyonki-kenkyu) 1998;37:385–395.
  • Eberhard W.G. Harvard University Press; Cambridge, MA: 1985. Sexual selection and animal genitalia.
  • Excoffier L, Laval G, Schneider S. Arlequin v. 3.0: an integrated software package for population genetics data analysis. Evol. Bioinform. Online. 2005;1:47–50. [PMC free article] [PubMed]
  • Iijima A, Tada R. Evolution and Tertiary sedimentary basins of Japan in reference to opening of the Japan Sea. J. Facult. Sci. Univ. Tokyo Sect. II. 1990;22:121–171.
  • Ishizuka K. Taxonomic study of the genus Pheretima s. lat. in Japan. Bull. Seikei Univ. 2001;33:1–125. [In Japanese.]
  • Hayashi M. Late Cenzoic Coleoptera of Japan (Part 1): Carabidae. Gekkan-mushi. 2001;369:16–19. [In Japanese.]
  • Hayashi M. Faunal changes in Donaciinae during the Quaternary in central Japan (Coleoptera, Chrysomelidae) In: Jolivet P, Santiago-Blay J.A, Schmitt M, editors. New development in the biology of Chrysomelidae. SPB Academic Publishing; The Hague, The Netherlands: 2004. pp. 263–274.
  • Hey J, Nielsen N. Integration within the Felsenstein equation for improved Markov chain Monte Carlo methods in population genetics. Proc. Natl Acad. Sci. USA. 2007;104:2785–2790. doi:10.1073/pnas.0611164104 [PMC free article] [PubMed]
  • Hiura I. Discovery of Apotomopterus from the Japanese Neogene (Coleoptera: Carabidae) Shizenshi-kenkyu Occas. Pap. Osaka Mus. Nat. Hist. 1971;1:45–50.
  • Hosken D.J, Stockley P. Sexual selection and genital evolution. Trends Ecol. Evol. 2004;19:87–93. doi:10.1016/j.tree.2003.11.012 [PubMed]
  • Huelsenbeck J.P, Ronquist F. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics. 2001;17:754–755. doi:10.1093/bioinformatics/17.8.754 [PubMed]
  • Imura O, Mizusawa K. Mushi-sha; Tokyo, Japan: 1996. The Carabus of the World.
  • Ishikawa R. Carabidae: Carabinae. In: Ueno S, Kurosawa Y, Sato M, editors. The Coleoptera of Japan in color. vol. II. Hoikusha; Osaka, Japan: 1985. pp. 14–54.
  • Ishikawa R. Yasaka-shobo; Tokyo, Japan: 1991. The evolution of Carabus.
  • Kim C.G, Zhou H.Z, Imura Y, Tominaga O, Su Z.H, Osawa S. Pattern of morphological diversification in the Leptocarabus ground beetles (Coleoptera, Carabidae) as deduced from mitochondrial ND5 and nuclear 28S rDNA sequences. Mol. Biol. Evol. 2000;17:137–145. [PubMed]
  • Kitamura A, Kimoto K. History of the inflow of the warm Tsushima Current into the Sea of Japan between 3.5 and 0.8 Ma. Palaeogeogr. Palaeoclimatol. Palaeoecol. 2006;236:355–366. doi:10.1016/j.palaeo.2005.11.015
  • Kitamura A, Takano O, Takata H, Omote H. Late Pliocene–Early Pleistocene paleoceanographic evolution of the Sea of Japan. Palaeogeogr. Palaeoclimatol. Palaeoecol. 2001;172:81–98. doi:10.1016/S0031-0182(01)00272-3
  • Koaze T, Nogami M, Ono Y, Hirakawa K. Geomorphology of Hokkaido. vol. 2. University of Tokyo Press; Tokyo, Japan: 2003. Regional geomorphology of the Japanese Islands.
  • Konuma J, Chiba S. Ecological character displacement caused by reproductive interference. J. Theor. Biol. 2007;21:354–364. doi:10.1016/j.jtbi.2007.03.013 [PubMed]
  • Kubota K. Natural hybridization between Carabus (Ohomopterus) maiyasanus and C (O.) iwawakianus (Coleoptera, Carabidae) Kontyu Tokyo. 1988;53:370–380.
  • Kubota K, Sota T. Hybridization and speciation in the carabid beetles of the subgenus Ohomopterus (Coleoptera, Carabidae, genus Carabus) Res. Popul. Ecol. 1998;40:213–222. doi:10.1007/BF02763406
  • Kuno E. Competitive exclusion through reproductive interference. Res. Popul. Ecol. 1992;34:275–284. doi:10.1007/BF02514797
  • Machida H, Ota Y, Kawana T, Moriwaki H, Nagaoka S. Regional geomorphology of the Japanese islands. vol. 7. University of Tokyo Press; Tokyo, Japan: 2001. Geomorphology of Kyushu and the Ryukyus.
  • Masaki S. Geographic variation and climatic adaptation in a field cricket (Orthoptera: Gryllidae) Evolution. 1967;21:725–741. doi:10.2307/2406770
  • Mendelson T.C, Shaw K.L. Rapid speciation in an arthropod. Nature. 2005;433:375–376. doi:10.1038/433375a [PubMed]
  • Morse G.E, Farrell B.D. Ecological and evolutionary diversification of the seed beetle genus Stator (Coleoptera: Chrysomelidae: Bruchinae) Evolution. 2005;59:1315–1333. doi:10.1554/04-140 [PubMed]
  • Nagata N, Kubota K, Sota T. Phylogeography and introgressive hybridization of the ground beetle Carabus yamato in Japan based on mitochondrial gene sequences. Zool. Sci. 2007a;24:465–474. doi:10.2108/zsj.24.465 [PubMed]
  • Nagata N, Kubota K, Yahiro K, Sota T. Mechanical barriers to introgressive hybridization revealed by mitochondrial introgression patterns in Ohomopterus ground beetle assemblages. Mol. Ecol. 2007b;16:4822–4836. doi:10.1111/j.1365-294X.2007.03569.x [PubMed]
  • Normark B.B, Jordal B.H, Farrell B.D. Origin of a haplodiploid beetle lineage. Proc. R. Soc. B. 1999;266:2253–2259. doi:10.1098/rspb.1999.0916
  • Ohshima K. The history of straits around the Japanese Islands in the Late-Quarternary. Quat. Res. Daiyonki-Kenkyu. 1990;29:193–208. [In Japanese.]
  • Palumbi S.R. Nucleotide acids II: the polymerase chain reaction. In: Hillis D.M, Moritz C, Mable B.K, editors. Molecular systematics. Sinauer Associates; Sunderland, MA: 1996. pp. 205–247.
  • Posada D, Crandall K.A. Modeltest: testing the model of DNA substitution. Bioinformatics. 1998;14:817–818. doi:10.1093/bioinformatics/14.9.817 [PubMed]
  • Ribeiro J.M.C, Spielman A. The satyr effect: a model predicting parapatry and species extinction. Am. Nat. 1986;128:513–528. doi:10.1086/284584
  • Roff D. Optimizing development time in a seasonal environment: the ‘ups and downs’ of clinal variation. Oecologia. 1980;45:202–208. doi:10.1007/BF00346461
  • Sasabe M, Takami Y, Sota T. The genetic basis of interspecific differences in genital morphology of closely related carabid beetles. Heredity. 2007;98:385–391. doi:10.1093/jhered/esm076 [PubMed]
  • Schluter D. Oxford University Press; New York, NY: 2000. The ecology of adaptive radiation.
  • Sota T. Life history patterns of carabid beetles belonging to the subtribe Carabina (Coleoptera, Carabidae) in the Kinki District, western Japan. Kontyu Tokyo. 1985a;53:370–378.
  • Sota T. Activity patterns, diets and interspecific interactions of coexisting spring and autumn breeding carabids: Carabus yaconinus and Leptocarabus kumagaii (Coleoptera: Carabidae) Ecol. Entomol. 1985b;10:315–324. doi:10.1111/j.1365-2311.1985.tb00728.x
  • Sota T. Radiation and reticulation: extensive introgressive hybridization in the carabid beetles Ohomopterus inferred from mitochondrial gene genealogy. Popul. Ecol. 2002;44:145–156. doi:10.1007/s101440200018
  • Sota T, Ishikawa R. Phylogeny and life-history evolution of Carabus (the subtribe Carabina: Coleoptera, Carabidae) based on sequences of two nuclear genes. Biol. J. Linn. Soc. 2004;81:135–149. doi:10.1111/j.1095-8312.2004.00277.x
  • Sota T, Kubota K. Genital lock-and-key as a selective agent against hybridization. Evolution. 1998;52:1507–1513. doi:10.2307/2411321
  • Sota T, Sasabe M. Utility of nuclear allele networks for the analysis of closely-related species in the genus Carabus, subgenus Ohomopterus. Syst. Biol. 2006;55:329–344. doi:10.1080/10635150500541607 [PubMed]
  • Sota T, Vogler A.P. Incongruence of mitochondrial and nuclear gene trees in the carabid beetles Ohomopterus. Syst. Biol. 2001;50:39–59. [PubMed]
  • Sota T, Vogler A.P. Reconstructing species phylogeny of the carabid beetles Ohomopterus using multiple nuclear DNA sequences: heterogeneous information content and the performance of simultaneous analysis. Mol. Phylogenet. Evol. 2003;26:139–154. doi:10.1016/S1055-7903(02)00311-1 [PubMed]
  • Sota T, Takami Y, Kubota K, Ujiie M, Ishikawa R. Interspecific body size differentiation in species assemblages of the carabid beetles Ohomopterus in Japan. Popul. Ecol. 2000a;42:279–291. doi:10.1007/PL00012006
  • Sota T, Takami Y, Kubota K, Ishikawa R. Geographic variation in the body size of some Japanese Leptocarabus species (Coleoptera, Carabidae): the “toppled-domino pattern” in species along a geographic cline. Entomol. Sci. 2000b;3:309–320.
  • Sota T, Kusumoto F, Kubota K. Consequences of hybridization between Ohomopterus insulicola and O. arrowianus (Coleoptera, Carabidae) in a segmented river basin: parallel formation of hybrid swarms. Biol. J. Linn. Soc. 2000c;71:297–313.
  • Sota T, Ishikawa R, Ujiie M, Kusumoto F, Vogler A.P. Extensive trans-species mitochondrial polymorphisms in the carabid beetles Carabus subgenus Ohomopterus caused by repeated introgressive hybridization. Mol. Ecol. 2001;10:2833–2847. [PubMed]
  • Su Z.H, Tominaga O, Ohama T, Kajiwara E, Ishikawa R, Okada T.S, Nakamura K, Osawa S. Parallel evolution in radiation of Ohomopterus ground beetles inferred from mitochondrial ND5 gene sequences. J. Mol. Evol. 1996;43:662–671. doi:10.1007/BF02202114 [PubMed]
  • Swofford, D. L. 2002 PAUP* Phylogenetic analysis using parsimony (*and other methods). v. 4.0. Sunderland, MA: Sinauer Associates.
  • Tada R. Paleoceanographic evolution of the Japan Sea. Palaeogeogr. Palaeoclimatol. Palaeoecol. 1994;108:487–508. doi:10.1016/0031-0182(94)90248-8
  • Takami Y. Phylogeny of the subgenus Ohomopterus (Coleoptera, Carabidae, genus Carabus): a morphological aspect. Tokyo Metropol. Univ. Bull. Nat. Hist. 2000;4:1–32.
  • Takami Y. Mating behavior, insemination and sperm transfer in the ground beetle Carabus insulicola. Zool. Sci. 2002;19:1067–1073. doi:10.2108/zsj.19.1067 [PubMed]
  • Takami Y. Experimental analysis of the effect of genital morphology on insemination success in the ground beetle Carabus insulicola (Coleoptera: Carabidae) Ethol. Ecol. Evol. 2003;15:51–61.
  • Takami Y. Spermatophore displacement and male fertilization success in the ground beetle Carabus insulicola. Behav. Ecol. 2007;18:628–634. doi:10.1093/beheco/arm022
  • Takami Y, Sota T. Rapid diversification of male genitalia and mating strategies in Ohomopterus ground beetles. J. Evol. Biol. 2007;20:1385–1395. doi:10.1111/j.1420-9101.2007.01338.x [PubMed]
  • Takami Y, Suzuki H. Morphological, genetic and behavioral analyses of a hybrid zone between the ground beetles Carabus lewisianus and C. albrechti (Coleoptera, Carabidae): asymmetrical introgression caused by movement of the zone? Biol. J. Linn. Soc. 2005;86:79–94. doi:10.1111/j.1095-8312.2005.00527.x
  • Takami Y, Nagata N, Sasabe M, Sota T. Asymmetry in reproductive isolation and its effect on directional mitochondrial introgression between the parapatric ground beetles Carabus yamato and C. albrechti. Popul. Ecol. 2007;49:337–346. doi:10.1007/s10144-007-0052-6
  • Templeton A.R. Nested clade analyses of phylogeographic data: testing hypotheses about gene flow and population history. Mol. Ecol. 1998;7:381–397. doi:10.1046/j.1365-294x.1998.00308.x [PubMed]
  • Templeton A.R, Crandall K.A, Sing C.F. A cladistic analysis of phenotypic association with haplotypes inferred from restriction endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics. 1992;132:619–633. [PMC free article] [PubMed]
  • Tsukada M. A vegetation map in the Japanese archipelago approximately 20,000 years B.P. Jpn J. Ecol. 1984;34:203–208.
  • Ujiie M, Kubota K, Sota T, Ishikawa R. Parallel formation of hybrid swarms of ground beetles in the genus Carabus (Coleoptera: Carabidae) in adjacent river basins. Entomol. Sci. 2005;8:429–437. doi:10.1111/j.1479-8298.2005.00142.x
  • Usami T, Yokoyama J, Kubota K, Kawata M. Genital lock-and-key system and premating isolation by mate preference in carabid beetles (Carabus subgenus Ohomopterus) Biol. J. Linn. Soc. 2006;87:145–154. doi:10.1111/j.1095-8312.2006.00562.x
  • Yonekura N, Kaizuka S, Nogami M, Chinzei K. Introduction to Japanese geomorphology. vol. 1. University of Tokyo Press; Tokyo, Japan: 2001. Regional geomorphology of the Japanese Islands.
  • Yoshimura J, Clark C.W. Population dynamics of sexual and resource competition. Theor. Popul. Biol. 1994;45:121–131. doi:10.1006/tpbi.1994.1007 [PubMed]
  • Zwickl, D. J. 2006 Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. PhD dissertation, the University of Texas at Austin. See www.bio.utexas.edu/faculty/antisense/garli/Garli.html

Articles from Philosophical Transactions of the Royal Society B: Biological Sciences are provided here courtesy of The Royal Society
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...