Mitochondrial DNA Sequence Variation of the Mushroom Pest Flies, Lycoriella Mali (Diptera: Sciaridae) and Coboldia Fuscipes (Diptera: Scatopsidae), in Korea

Appl. Entomol. Zool. 36 (4): 451–457 (2001)

Mitochondrial DNA sequence variation of the mushroom pest ﬂies, Lycoriella mali (Diptera: Sciaridae) and Coboldia fuscipes (Diptera: Scatopsidae), in Korea

Jin-Sik Bae, Iksoo Kim, Seong-Ryul Kim, Byung-Rae Jin and Hung-Dae Sohn* College of Natural Resources and Life Science, Dong-A University, Busan 604–714, Korea (Received 12 March 2001; Accepted 14 May 2001)

Abstract We analyzed a portion of mitochondrial COI gene sequences (406 bp) to investigate the genetic diversity and geographic variation of the mushroom pest flies Lycoriella mali and Coboldia fuscipes in Korea. L. mali showed minimal sequence divergence (0.2%) in two mtDNA haplotypes, whereas C. fuscipes showed an intermediate level of sequence divergence (1.2% at maximum) compared with other relevant studies. While L. mali was fixed with one haplotype except for one population, C. fuscipes possessed a total of ten mtDNA haplotypes, and six of these occurred commonly in multiple populations. We ascribed the difference in the level of genetic variation between the two species to a difference in the degree to which they are dependent on cultured mushroom, which is a fluctuating food source. In C. fuscipes, as in other cosmopolitan insect pest species, a high rate of female migration (Nmϭ0.947–infinite) and little ϭ Ϫ genetic differentiation (FST 0.345– 0.094) between populations were estimated.

Key words: Mushroom ﬂy, Coboldia fuscipes, Lycoriella mali, mitochondrial DNA, genetic diversity

Coboldia fuscipes is another major oyster mush- INTRODUCTION room pest in Korea. The population grows in the Understanding the genetic population structure summer season (from July to August), and dimin- of insect pests may enable us to predict increases ishes somewhat in the fall season (from September and decreases in population size, as well as to November), in Korea (Kim et al., 1999; Choi et broaden our understanding of the mode of occur- al., 2000). This species occurs in diverse habitats rence and migration among localities (Roderick, such as decaying plants, animal materials, and 1996). Thus, such knowledge should provide an some flowering grasses (Shiraki, 1981). In Korea, important biological information for the control of mushroom flies recently received attention as pests, pest species. and taxonomy, geographic distribution, life cycle, Mushroom flies cause severe damage to mush- and some physiological aspects have been investi- room: larvae feed on the mycelium and fruitbody gated (Choi et al., 1997, 1999, 2000; Lee et al., of the mushroom and adult flies transport germs 1998, 1999; Kim et al., 1999, 2000). However, such as nematodes, mites and mold spores (Clift, mushroom flies have never been subjected to stud- 1979; Clancy, 1981; Wetzel, 1981; Kim and ies on genetic structure, gene flow, and genetic di- Hwang, 1996). Lycoriella mali is the most abun- versity. dant pest, occurring through at the year in most re- Mitochondrial DNA (mtDNA) has a high evolu- gions of Korea (Kim et al., 1999). This species is tionary rate compared to the functional counterpart distributed in North America, Europe, and Asia of nuclear DNA. In addition, it is inherited mater- (Lee et al., 1999). In North America, crop losses nally, lacking genetic recombination, and is easy to caused by L. mali were averaged to at least 17% handle (Brown et al., 1979, 1982; Cantatore and (Cantelo, 1979). L. mali inhabits mushrooms as Saccone, 1987; Harrison, 1989). These characteris- well as rotting wood, decaying potato, and rotting tics make the mtDNA molecule a particularly ap- vegetable (Stamets and Chilton, 1984; Lee et al., propriate marker for tracing the recent evolutionary 1999). history of animals, including colonization, intro-

* To whom correspondence should be addressed.

451 452 J.-S. Bae et al. duction, and population bottlenecks (Wilson et al., 1985). It is known that the cytochrome oxidase subunit I (COI) gene of mtDNA is highly variable in its DNA sequence, especially at the silent sites. We se- lected a portion of the COI gene that includes the membrane-spanning helices M3, M4, and M5, ex- ternal loops E2 and E3, and internal loop I2 (Lunt et al., 1996). This portion has been proven to be useful for the study of intraspeciﬁc genetic variation in insects (Simon et al., 1994; Kim et al., 2000a, b, c; Lee et al., 2000). We sequenced a portion of the COI gene from natural populations of L. mali and C. fuscipes in Korea to study the extent and nature of genetic variation, geographic structure, and gene ﬂow among populations.

MATERIALS AND METHODS Insects. Adult L. mali and C. fuscipes were col- Fig. 1. Sampling locations of C. fuscipes and L. mali in leted using an aspirator at oyster mushroom houses Korea. General locality names are as follows: 2. Yeongwol- in eight localities in Korea from June 1999 to July gun, Gangwon-do; 3. Gyeongju-si, Gyeonsangbug-do; 4. Hwaseong-gun, Gyeonggi-do; 5. Busan-si; 6. Boseong-gun, 2000 (Fig. 1). One individual of each species (ani- Jeonlabuk-do; 7. Sacheon-si, Gyeongsangnam-do; 8. Chungju- mal numbers JB2 and JB3, respectively) was ob- si, Chungcheongbug-do; 9. Uiseong-gun, Gyeongsangbug-do. tained from indoor-rearing. Samples collected in the field were frozen at Ϫ70°C until molecular discovered (LM1 and LM2 in L. mali and CF1 and analysis. CF2 etc., in C. fuscipes). MtDNA. Total DNA was extracted by following Genetic diversity and distance. Haplotype di- the standard Proteinase K method (Kocher et al., versity and nucleotide diversity within local popu- 1989). A part of the COI gene was amplified by lations were estimated for C. fuscipes, using Arle- PCR using primers CI-J-1751 (5Ј-GGAGCTCCT- quin ver. 1.1 (Schneider et al., 1996). Genetic dis- GACATAGCATTCCC-3Ј) and CI-N-2191 (5Ј-CC- tance and female migration rate per-generation CGGTAAAATTAAAATATAAACTTC-3Ј) (Simon were estimated by subroutines in Arlequin ver. 1.1 et al., 1994). PCR conditions were as follows: an (Schneider et al., 1996). Genetic distance was initial denaturation step at 94°C for 5 min, 40 cy- based on pairwise FST indices (F ST) following the cles of 94°C for 30 s, 50°C for 40 s, and 72°C for approach described by Excoffier et al. (1992). Sta- 45 s, and a final extension step at 72°C for 7 min. tistical significance of the difference between a pair To confirm the successful DNA amplification, elec- of populations was tested by permutation (10,000 trophoresis was carried out using 0.5X TAE buffer bootstraps; Excoffier et al., 1992). Pairwise FST in 1% agarose gel. The PCR product was purified values were used to estimate per generation migra- using a PCR purification Kit (QIAGEN, Germany). tion rate, Nm (the product of the effective popula- DNA sequencing was performed using an ABI 377 tion size Ne and migration rate m), based upon the ϭ ϩ Genetic Analyzer (PE Applied Biosystems, USA). equilibrium relationship: FST 1/(2Nm 1). Sequence alignment was performed using an IBI MacVector (ver. 6.5). When homologous se- RESULTS quences from two individuals differed by at least one nucleotide site, the two sequences were consid- Lycoriella mali ered as different haplotypes. Haplotype designa- Sequence analysis of 55 L. mali individuals re- tions were applied to new sequences as they were sulted in only two haplotypes (LM1, LM2) for a Genetic Variation of Mushroom Flies 453

Table 1. A list of trapping localities, animal numbers, sex, mitochondrial COI haplotypes and GenBank accession numbers of C. fuscipes

Collecting locality COI GenBank accession Collection date Animal number Sex (no. of individuals) Haplotype number

1. Dong-A Univ., 1998. 12. 17 JB3 F CF1 AF319839 Busan-si (1) 2. Yeongwol-gun, 1999. 7. 28 C2 F CF2 AF319840 Gangwon-do (8) C3 F CF3 AF319841 C5 F CF4 AF319842 C6 M CF5 AF319843 C7 M CF6 AF319844 C8 M CF3 AF319845 C9 M CF6 AF319846 C11 F CF2 AF319847 5. Busan-si (8) 2000. 5. 9 C21 F CF2 AF319848 C22 F CF2 AF319849 C24 F CF4 AF319850 C25 F CF1 AF319851 C27 M CF1 AF319852 C28 M CF1 AF319853 C29 M CF7 AF319854 C30 M CF1 AF319855 6. Boseong-gun, 2000. 5. 12 C33 F CF4 AF319856 Jeonlanam-do (8) C34 F CF8 AF319857 C35 F CF6 AF319858 C36 F CF8 AF319859 C38 M CF4 AF319860 C39 M CF4 AF319861 C41 M CF3 AF319862 C42 M CF9 AF319863 7. Sacheon-si, 2000. 5. 12 C47 F CF1 AF319864 Gyeongsangnam-do (8) C48 F CF6 AF319865 C49 F CF2 AF319866 C50 F CF4 AF319867 C51 F CF1 AF319868 C59 M CF7 AF319869 C60 M CF1 AF319870 C61 M CF4 AF319871 9. Uiseong-gun, 2000. 7. 21 C62 F CF2 AF319872 Gyeongsangbug-do (8) C66 M CF4 AF319873 C67 M CF2 AF319874 C72 M CF6 AF319875 C73 M CF2 AF319876 C76 M CF3 AF320759 C78 M CF6 AF320760 C79 M CF10 AF320761 partial sequencing of the COI gene (406 bp). Se- AF319784–AF319838. quence alignment revealed only one variable nucleotide site (position 79), which was the 3rd posi- Coboldia fuscipes tion of the codon and did not substitute an amino A total of ten COI haplotypes (CF1–CF10) was acid. Except for two individuals in Busan (Fig. 1), found from 41 individuals of C. fuscipes (Table 1). all individuals possessed haplotype LM1. GenBank These haplotypes revealed eight variable nu- accession numbers of individual L. mali are cleotide sites, seven of which were the 3rd position 454 J.-S. Bae et al.

Table 2. Within-locality diversity estimates of C. fuscipes

Locality Na NHb Hc Sd p e

2. Yeongwol 8 5 0.89 5 0.006 5. Busan 8 4 0.75 6 0.005 6. Boseong 8 5 0.86 5 0.006 7. Sacheon 8 5 0.86 6 0.007 9. Uiseong 8 5 0.86 6 0.006

a Sample size. b Number of haplotypes. c Haplotype diversity. d Number of polymorphic sites. e Nucleotide diversity.

of the codon. Position 200 was the 1st position of the codon, replacing the corresponding amino acid of haplotype CF9 from Valine to Isoleucine (Fig. 2). Sequence divergence in a pairwise comparison of the ten haplotypes ranged from 0.2 to 1.2% (1–5 bp). The largest sequence divergence was found in two cases: comparisons between CF4 and CF1 and between CF4 and CF2. Among the ten haplotypes, six were found in multiple localities (Table 1). In particular, haplotype CF4 was found in all the localities sampled, and CF2 and CF6 were found in four localities. Four haplotypes (CF5, CF8, CF9 and CF10) were found in only one locality, and these restricted haplotypes amounted to 12.2%. Within-locality genetic diversity was estimated in terms of haplotype diversity (H) and nucleotide diversity (p) (Table 2). Within-locality H and p were high, ranging from 0.5 to 0.7 and from 0.75 to 0.89, respectively.

Genetic differentiation (FST) and per-generation female migration rates (Nm) are shown in Table 3.

The greatest genetic distance (FST) found was that between Busan and Boseong, and this distance was the only one with statistical signiﬁcance (pϽ0.05). In the analysis of per-generation female migration rate (Nm), a comparatively low Nm was obtained in a comparison between Busan and Boseong (Nmϭ 0.947), but a high gene ﬂow between other populations was characteristic and no obvious phylogeo- graphic structure was observed.

Fig. 2. Sequence alignment of ten mitochondrial haplotypes of C. fuscipes obtained from 406-bp COI gene se- DISCUSSION quences. Only positions that differ from haplotype CF1 are in- Sequence divergence in Lycoriella mali and dicated. Coboldia fuscipes The mtDNA sequence divergence was 0.2% in Genetic Variation of Mushroom Flies 455

Table 3. C. fuscipes mitochondrial COI sequence data of genetic distance (FST) and per-generation female migration rates (Nm)

2. Yeongwol 5. Busan 6. Boseong 7. Sachon

ϭ 5. Busan FST 0.177 Nmϭ2.327 ϭϪ ϭ 6. Boseong FST 0.002 FST 0.345* Nmϭinfinite Nmϭ0.947 ϭϪ ϭϪ ϭ 7. Sacheon FST 0.003 FST 0.056 FST 0.124 Nmϭinfinite Nmϭinfinite Nmϭ3.545 ϭϪ ϭ ϭ ϭ 9. Uiseong FST 0.094 FST 0.148 FST 0.086 FST 0.011 Nmϭinfinite Nmϭ2.880 Nmϭ5.333 Nmϭ45.538

* pϽ0.005.

L. mali and a maximum of 1.2% in C. fuscipes. In growth of L. mali, resulting in a decrease of in- previous studies that utilized the COI gene, the traspecific genetic divergence to the minimum esti- maximal sequence divergences within species were mate. 0.2% for the domestic silkworm (Kim et al., Another question is how C. fuscipes has sus- 2000c), –0.23% and 0.12% for two species of rice tained a moderate level of nucleotide divergence planthopper (Mun et al., 1999), 0.4% for spruce (1.2%) haplotype number (ten haplotypes) under budworm species (Sperling and Hickey, 1994), such a fluctuating food-resource. In contrast, C. 0.5% for Heliconius butterflies (Brower, 1994), fuscipes has often been found in greenhouses for 1.2% for firefly species (Lee et al., 2000), 1.4% for flower cultivation (Shiraki, 1981), suggesting that the diamondback moth (Kim et al., 2000a), and C. fuscipes is less dependant upon mushroom than 3.8% for leaf beetle species (Funk et al., 1995). is L. mali. It appears that C. fuscipes can sustain a Compared with other data, the sequence diver- population by utilizing a diverse food source if the gence in L. mali is extremely low, while it is mod- proper temperature is maintained. Therefore, the erate in C. fuscipes. Also, haplotype diversity in C. larger genetic diversity in C. fuscipes compared to fuscipes was higher (ten) than it was in L. mali L. mali may be relevant to the differential host (two). preferences. Maruyama and Kimura (1980) found that fre- Although we tentatively concluded the above, it quent extinction and colonization of populations is difficult to reach a convincing conclusion, be- decrease the genetic diversity within species and cause no equivalent data is available for species that is particularly likely when the migration rate outside Korea. Furthermore, we can not rule out among subpopulations is higher than the extinction the possibility that the low nucleotide divergence in rate of subpopulations. Applying this theory to our L. mali may have stemmed from a founder effect, data, it appears that L. mali may have undergone because the species is distributed throughout the population fluctuations more frequently than C. world. fuscipes. For L. mali, the availability of mushroom is the most important factor for oviposition, feed- Gene flow in C. fuscipes ing, and population propagation, although some Most C. fuscipes populations showed genetic other minor habitats have been reported (Stamets similarity with a substantially high gene flow rate, and Chilton, 1984; Lee et al., 1999). In Korea, oys- and this resultantly appears to show overall similar ter mushrooms are cultivated mostly for the spring genetic differentiation estimates (FST) among the and autumn seasons, and summer is avoided be- C. fuscipes populations (Table 3). High genetic re- cause of the difficulty in temperature control. In latedness in C. fuscipes populations was also sup- addition, cultivation is often discontinued due to ported by co-occurrence of CF4 in all localities, infection of fungi (i.e. Trichoderma, Trichurus spi- extending over a distance of 300 km (Fig. 1 and ralis, Neurospora sp., Hypocrea rufa) (Mushroom Table 1). Such genetic structure is a characteristic Society of Korea, 2000). Such discontinuation of of insect species that travel with human beings and cultivation may directly influence the population crop pest species that occur throughout the world. 456 J.-S. Bae et al.

For example, Kim et al. (2000a, b) reported that the of Australia. Mushroom Sci. X (part II): 367–383. diamondback moth, Plutella xylostella, showed Excoffier, L., P. E. Smouse and J. M. Quattro (1992) Analysis only one nucleotide difference between mtDNA of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochon- haplotypes in Korean and Hawaiian populations, drial DNA restriction data. Genetics 131: 479–491. and suggested that populations throughout the Ko- Funk, D. J., D. J. Futuyma, G. Orti and A. Meyer (1995) Mito- rean peninsula are panmictic. For Nilaparvata lu- chondrial DNA sequences and multiple data sets: a phy- gens and Sogatella furcifera, COI haplotypes col- logenetic study of phytophagous beetles (Chrysomelidae: lected from 11 localities of seven Asian countries Ophraella). Mol. Biol. Evol. 12: 627–640. Harrison, R. G. (1989) Animal mitochondrial DNA as a ge- also showed little nucleotide divergence (0.23% in netic marker in population and evolutionary biology. N. lugens and 0.12% in S. furcifera), and most Trends Ecol. Evol. 4: 6–11. dominant haplotypes were distributed throughout Kim, I., J. S. Bae, K. H. Choi, B. R. Jin, K. R. Lee and H. D. all the localities surveyed (Mun et al., 1999). It ap- Sohn (2000a) Haplotype diversity and gene flow of the pears that C. fuscipes migrates to distant mush- diamondback moth, Plutella xylostella (L.) (Lepidoptera: room houses via refugees in neighboring agricul- Yponomeutidae), in Korea. Korean J. Appl. Entomol. 39: 43–52. tural areas and/or through harvested mushrooms in Kim, I., J. S. Bae, K. H. Choi, S. R. Kim, B. R. Jin, K. R. Lee the form of larvae. and H. D. Sohn (2000b) Mitochondrial DNA polymor- phism and population genetic structure of diamondback ACKNOWLEDGEMENTS moths, Plutella xylostella (Lepidoptera: Yponomeutidae), This research was supported by grants from the Korean in Korea. Korean J. Entomol. 30: 21–32. Ministry of Agriculture and Forestry. Kim, I., J. S. Bae, H. D. Sohn, P. D. Kang, K. S. Ryu, B. H. Sohn, W. B. Jeong and B. R. Jin (2000c) Genetic homo- REFERENCES geneity in the domestic silkworm, Bombyx mori, and phylogenetic relationship between B. mori and wild silk- Brower, A. V. Z. (1994) Phylogeny of Heliconius butterflies in- worm, B. mandarina using mitochondrial COI gene se- ferred from mitochondrial DNA sequences (Lepidoptera: quences. Int. J. Indust. Entomol. 1: 9–17. Nymphalidae). Mol. Phyl. Evol. 3: 159–174. Kim, K. C. and C. Y. Hwang (1996) An investigation of insect Brown, W. M., M. George, Jr. and A. C. Wilson (1979) Rapid pest on the mushroom (Lentinus edode, Pleurotus ostrea- evolution of animal mitochondral DNA. Proc. Natl. Acad. tus) in south region of Korea. Korean J. Appl. Entomol. Sci. USA 76: 1967–1971. 35: 45–51 Brown, W. M., E. M. Prager, A. Wang and A. C. Wilson Kim, S. R., K. H. Choi, E. S. Cho, W. J. Yang, B. R. Jin and H. (1982) Mitochondrial DNA sequence of primates: tempo D. Sohn (1999) An investigation of the major Dipteran and mode of evolution. J. Mol. Evol. 18: 225–239. pest on the oyster mushroom (Pleurotus ostreatus) in Cantatore, P. and C. Scacone (1987) Organization, structure, Korea. Korean J. Appl. Entomol. 38: 41–46. and evolution of mammalian mitochondrial genes. Int. Kim, S. R., K. H Choi, W. J. Yang, I. Kim, B. R. Jin and H. D. Rev. Cytol. 108: 149–208. Sohn (2000) Ovarian development and vitellin of mush- Cantelo, W. W. (1979) Lycoriella mali: control in mushroom room fly, Lycoriella mali (Diptera: Sciaridae). J. Asia-Pa- compost by incorporation of insecticides into compost. J. cific Entomol. 3: 11–18. Econ. Entomol. 72: 703–705. Kocher, T. D., W. K. Thomas, A. Meyer, S. V. Edwards, S. Choi, K. H., S. R. Kim, E. S. Cho, J. S. Bae, B. R. Jin, W. J. Pääbo, F. X. Villablanca and A. C. Wilson (1989) Dy- Yang and H. D. Sohn (1999) A study on respiratory horns namics of mitochondrial DNA evolution in animals: am- on pupa of mushroom pest, Coboldia fuscipes (Diptera: plification and sequencing with conserved primers. Proc. Scatopsidae). Korean J. Appl. Entomol. 38: 241–247. Natl. Acad. Sci. USA 86: 6196–6200. Choi, K. H., S. R. Kim, E. S. Cho, W. J. Yang, B. R. Jin, M. Lee, H. S., K. J. Kim and H. U. Lee (1998) Effect of tempera- Takada and H. D. Sohn (2000) Developmental and life ture on the development of sciarid fly Bradysia sp. history characteristics of the oyster mushroom fly, (Diptera : Sciaridae). Korean J. Appl. Entomol. 37: 171– Coboldia fuscipes (Diptera: Scatopsidae). Appl. Entomol. 178. Zool. 35: 495–498. Lee, H. S., K. C. Kim, C. G. Park and W. K. Shin (1999) De- Choi, K. H., H. C. Park, P. D. Kang, S. K. Kang and H. D. scription of fungus gnat, Lycoriella mali Fitch (Diptera: Sohn (1997) Development characteristics and life cycle Sciaridae) from Korea. Korean J. Appl. Entomol. 38: of a sciarid fly (Lycoriella sp.) in indoor rearing. Korean 209–212. J. Appl. Entomol. 36: 77–82. Lee, S. C., I. Kim, J. S. Bae, B. R. Jin, S. E. Kim, J. K. Kim, Clancy, G. (1981) Observations of mites associated with the H. J. Yoon, S. R. Yang, S. H. Lim and H. D. Sohn (2000) low yielding crops of cultivated Agaricus bisporus in Mitochondrial DNA sequence variation of the firefly, Py- Australia. Mushroom Sci. XI: 233–244. rocoelia rufa (Coleoptera: Lampyridae), in Korea. Korean Clift, A. D. (1979) The identity, economic importance and J. Appl. Entomol. 39: 181–191. control of insect pests of mushroom in New South Wales Lunt, D. H., D.-X. Zhang, J. M. Szymura and G. M. Hewitt Genetic Variation of Mushroom Flies 457

(1996) The insect cytochrome oxidase I gene: evolution- Shiraki, T. (1981) Classification of Insects. 4th ed. Hoku- ary patterns and conserved primers for phylogenetic stud- ryukan Co. Press, Tokyo. 668 pp. ies. Insect Mol. Biol. 5: 153–165. Simon, C., F. Frati, A. Beckenbach, B. Crespi, H. Liu and P. Maruyama, T. and M. Kimura (1980) Genetic variability and Flook (1994) Evolution, weighting, and phylogenetic util- effective population size when local extinction and recol- ity of mitochondrial gene sequences and a compilation of onization of subpopulations are frequent. Proc. Natl. conserved polymerase chain reaction primers. Ann. Ento- Acad. Sci. USA 77: 6710–6714. mol. Soc. Am. 87: 651–701. Mun, J. H., Y. H. Song, K. L. Heong and G. K. Roderick Sperling, F. A. H. and D. A. Hickey (1994) Mitochondrial (1999) Genetic variation among Asian populations of rice DNA sequence variation in the spruce budworm species planthoppers, Nilaparvata lugens and Sogatella furcifera complex (Choristoneura: Lepidoptera). Mol. Biol. Evol. (Hemiptera: Delphacidae): mitochondrial DNA se- 11: 656–665. quences. Bull. Entomol. Res. 89: 245–253. Stamets, P. and J. S. Chilton (1984) The Mushroom Cultivator: Mushroom Society of Korea (2000) Diagnois and Control of A Practical Guide to Growing Mushrooms at Home. Disease and Pest in Mushrooms. Research Information Agarikon Press, Washington, D.C. 415 pp. Center, Seoul. 782 pp. Wetzel, H. A. (1981) Integrated pest management. Mushroom Roderick, G. K. (1996) Geographic structure of insect popula- News 29 (8): 29–33. tion: gene flow, phylogeography, and their uses. Annu. Wilson, A. C., R. L. Cann, S. M. Carr, M. George, U. B. Gyl- Rev. Entomol. 41: 325–352. lenstein, K. M. Helm-Bychowski, R. G. Highchi, S. R. Schneider, S., J.-M. Kueffer, D. Roessli and L. Excoffier Palumbi, E. M. Prager, R. D. Sage and M. Stoneking (1996) Arlequin: a software package for population ge- (1985) Mitochondrial DNA and two perspectives on evo- netics. Genetics and Biometry Lab., Dept. Anthropology, lutionary genetics. Biol. J. Linn. Soc. 26: 375–400. Univ. of Geneva.