The Agabus bipustulatus complex includes one of Europe's most widely distributed and common diving beetles. This complex, which is known for its large morphological variation, has a complex demographic and altitudinal variation in elytral reticulation. The various depth of the reticulation imprint, both in smaller and larger meshes, results in both mat and shiny individuals, as well as intermediate forms. The West Palearctic lowland is inhabited by a sexually dimorphic form, with shiny males and mat females. In mountain regions, shiny individuals of both sexes are found intermixed with mat individuals or in pure populations in central and southern areas, whereas pure populations of mat individuals are exclusively found in the northern region at high altitude. Sexual selection is proposed as a driving force in shaping this variation. However, the occurrence of different types of reticulation in both sexes and disjunct geographical distribution patterns suggest an additional function of the reticulation. Here we investigate the phylogeographical history, genetic structure and reticulation variation of several named forms within the Agabus bipustulatus complex including A. nevadensis. The molecular analyses recognised several well-supported clades within the complex. Several of the named forms had two or more independent origins. Few south European populations were uniform in reticulation patterns, and the males were found to display large variation. Reticulation diversity and population genetic variability were clearly correlated to altitude, but no genetic differences were detected among populations with mixed or homogenous forms. Observed reduction in secondary reticulation in female and increased variance in male at high altitude in South Europe may be explained by the occurrence of an additional selective force, beside sexual selection. The combined effect of these selective processes is here demonstrated in an extreme case to generate isolation barriers between populations at high altitudes. Here we discuss this selective force in relation to thermal selection.
Citation: Drotz MK, Brodin T, Nilsson AN (2010) Multiple Origins of Elytral Reticulation Modifications in the West Palearctic Agabus bipustulatus Complex (Coleoptera, Dytiscidae). PLoS ONE 5(2): e9034. https://doi.org/10.1371/journal.pone.0009034
Editor: Robert DeSalle, American Museum of Natural History, United States of America
Received: August 12, 2009; Accepted: December 29, 2009; Published: February 3, 2010
Copyright: © 2010 Drotz et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research project was partially sponsored by Nilsson-Ehle foundation (material and analysis costs) within the Kungl. Fysiografiska Sällskapet (http://www.fysiografen.se/default.asp) and Lidköping municipality (salary cost for main author). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The elytral surface of most diving beetles is covered by a reticulation pattern impressed on the elytra, often consisting of small and large meshes – the primary and secondary reticulation –. The primary reticulation consists of a more or less regular fine meshwork, whereas the secondary reticulation is coarser with larger meshes of a more irregular, often oblong, shape. Sexual dimorphism is commonly found, especially in geographically disjunct areas at high altitude or gradually between longitudinal areas in e.g. the West Palaearctic region. Traditionally, these reticulation patterns have taxonomically been recognised as a set of characters used to identify and separate species, subspecies and varieties within several species complexes , –. However, little effort has been taken to understand the evolutionary significance of the sexual and altitudinal variation in elytral microsculpture and reticulation patterns occurring within a single species complex. This view changed when Bergsten and co-workers  nicely demonstrated an ongoing arms-race including female elytral sculpture and the number and size of adhesive male pro- and mesotarsal setae in several dytiscine genera. Miller  and subsequent studies – strengthened these observations and argued with support from the general model of sexual conflict – that the first step of a female response to increased mating cost is the occurrence of an increased female reticulation compared to males with tarsal modifications.
Accordingly, these findings provide an explanation of the sexual dimorphism and increased modification of female reticulation in dytiscids. However, the observation that both males and females in some water beetles display a large variation in primary and secondary reticulation indicates that the microsculpture most likely has evolved for some other reason. Moreover, the occurrence of similar reticulation patterns at high altitude in different mountain ranges across large geographical areas also indicates the presence of an additional selective agent or common evolutionary history –. Sexual selection could therefore be a selective mechanism that enhances or contributes secondary to the diversification of reticulation patterns at high altitude in some species –.
Clearly, several underlying evolutionary factors need to be considered in order to understand why different aberrant reticulation patterns, such as a more strongly impressed male secondary reticulation and more shiny females, occur at high altitude in contrast to the more homogeneous reticulation pattern normally observed at lower altitudes , –. Here we will focus on the phylogeographic history of different reticulation patterns and genetic differences between populations consisting of various reticulation forms in relation to altitude observed in the Agabus bipustulatus (Linnaeus, 1767) species complex.
This species complex is known to display a pronounced individual shape variation in combination with an elytral reticulation that varies both between and within sexes, especially at higher altitudes –, , –. The pronounced morphological, geographical and altitudinal variation is reflected taxonomically in the 23 available junior synonyms of uncertain taxonomic status . The montane form named Agabus solieri Aubé, 1837 has a body-shape, especially seen on the pronotum, which is more slender than in the widely distributed topotypic lowland form bipustulatus , . Most of the other named forms are montane, and found in populations imbedded in areas dominated by the solieri-like form. Main differences between the forms, which taxonomically often have been dealt with as subspecies or varities of either bipustulatus or solieri, are the shininess of the elytra and different patterns in their secondary reticulation (Fig. 1). In particular, the kiesenwetterii form stands out since both sexes have wider meshes in the secondary reticulation, and the elytra are more shining relative to the regular solieri form. The geographical distribution of these two forms is interesting since the kiesenwetterii form commonly is found at high altitude in central and south Europe imbedded in populations dominated by the duller solieri form, whereas solieri is the only form occurring in North Europe . These forms are suggested to be subjected to sexual selection which is seen in populations where males evolve wider tarsi and attached suckers in the presence of modified reticulated females . In other known forms, only the male displays reticulation variation, whereas the females are homogenous for a given pattern.
(1) Reticulation classified as type A and found within the topotypic male bipustulatus form, (2) as type B found within kiesenwetterii, (3) as type G found within the pyrenaeus form. Reticulation classes are coded as in table 2.
The Iberian endemic Agabus nevadensis Hå. Lindberg, 1939 is here of interest due to its geographically restricted high-altitude occurrence in the Sierra Nevada and morphological similarity to A. bipustulatus. This local endemic has a reticulation pattern identical to that of the kiesenwetterii form in both sexes . Its species status has been questioned and it is commonly included within the A. bipustulatus-complex , .
A total of 717 specimens representing six of the named forms of A. bipustulatus were collected from 37 sample sites across its West Palearctic distribution area in order to capture as large genetic and morphological variation as possible (Table 1, Fig. 2). All specimens used in the population genetic analysis were also included in the morphological analysis of the secondary reticulation.
Map codes and corresponding locality codes are given in table 1.
In France, 14 A. bipustulatus populations were sampled near Grenoble, Guillestre, and Montcenisio. The type locality of A. solieri lies within the Grenoble region, and Guignot  described A. solieri var. falcozi from Montcenisio. Seidlitz  described A. solieri var. kiesenwetterii from Illyria, Piemonte and the Pyrenees, and it is according to Guignot  now also found it at high altitudes in the French Alps. One Italian locality, from which A. bipustulatus dolomitanus Scholz, 1935 is known, was visited . Five populations were sampled in the Spanish Pyreenes near Vall de Boí in Catalunya, including the water system from which A. solieri pyrenaeus was described . A total of 413 A. nevadensis specimens from eight populations were sampled near its type locality in central Sierra Nevada  (Table 1, Fig. 2).
To analyse the evolutionary background of different reticulation forms we utilised two A. nevadensis specimens and 30 A. bipustulatus specimens representing the reticulation pattern of 16 bipustulatus, two kiesenwetterii, three falcozi, one dolomitanus, one pyrenaeus and seven solieri forms. Individual specimens were collected to represent different secondary reticulation forms from different geographical regions with the aim to maximise the haplotypic (mtDNA) variation. The optimal outgroup taxa for the A. tristis species group is A. bipustulatus and A. nevadensis, according to molecular analyses of the Agabini, species from the A. nebulosus group . Two A. nebulosus (Forster, 1771) specimens were therefore included, together with one A. affinis (Paykull, 1798) and one A. guttatus (Paykull, 1798). The former represents a basal clade of the subgenus Gaurodytes, whereas the latter belongs to the sister group of the A. nebulosus plus A. tristis groups . Seven specimens of the A. tristis group not belonging to the A. bipustulatus complex were also included in the phylogenetic analysis (Table 1): representing the European Madeiran endemic A. wollastoni Sharp, 1882 the North Palearctic A. melanarius Aubé, 1837 and the Holarctic A. tristis Aubé, 1838.
Collected specimens that were used in the population genetic and phylogenetic analyses were transported alive in damp moss to the laboratory, where they were sexed and frozen at −70°C. Morphologically analysed specimens were stored in 96% ethanol in a refrigerator at 4°C.
The reticulation patterns of the sampled forms are as follows; in North Europe, from the high mountains of Scandinavia to the lowland region below the Alps and the Pyrenees, the A. bipustulatus complex is represented by a sexually dimorphic form with a more or less constant elytral reticulation pattern. The males are shinier than the females and the secondary reticulation of both sexes consists of longitudinally stretched meshes. In the montane kiesenwetterii form both sexes have wide meshes in the secondary reticulation and the elytra are shinier than in topotypic A. bipustulatus from Sweden. Guignot  observed that falcozi males have wider secondary reticulation and are not as shining as those of kiesenwetterii, whereas the females are similar to the normal bipustulatus form. Only the males of pyrenaeus are shiny and have a more isomorphic secondary reticulation, and females have been described as similar to those of bipustulatus . The latter information is not fully correct, however, and we will here demonstrate that the pyrenaeus females also display a similar variation as the males. The dolomitanus form is similar to falcozi, but it has the pronotal shape of lowland bipustulatus . Franciscolo  also concluded, without any quantitative analysis, that the variation in secondary reticulation in kiesenwetterii and dolomitanus is so similar that they have to be conspecific.
Individual variation in the secondary elytral reticulation was classified after the total sample was screened for common patterns. Seven categories (A - G) ranging from tightly packed longitudinally elongated meshes to more or less isomorphic meshes were recognised (Fig. 1). Of these categories, the A, B, C and G patterns cover most of the total surface of the elytra (Table 2). The remaining categories D, E and F describe intermixed patterns of longitudinally elongated and isomorphic meshes. The A reticulation pattern is found in the bipustulatus and solieri forms. The B pattern is found in kiesenwetterii, falcozi and within the dolomitanus form. Within pyrenaeus we recognised three reticulation patterns: C, F and G.
The Simpson index (D), which measures the probability that two individuals randomly selected from a sample will belong to the same reticulation category, was used to calculate reticulation diversity per population .
A total of eleven enzyme systems coding for nine loci gave quantitatively reliable results: Hexokinase (Hk), Triosephosphate isomerase (Tpi), α-Glycerophosphate dehydrogenase (α-Gpdh), Isocitrate dehydrogenase (Idh), Malate dehydrogenase (Mdh), Malic enzyme (Me) and Esterase (Est-1 & Est-2). In addition, Alkaline phosphate (Alp) was screened in A. bipustulatus from France and Italy, and so were Phosphoglucoisomerase (Pgi), Phosphogluconate dehydrogenase (Pgd) and Xanthine dehydrogenase (Xdh) in A. bipustulatus and A. nevadensis from Spain. Staining recipes and gel buffers are modified from those of Shaw and Prasad . The beetles were prepared and the alleles named by migration distance to the most common allele as described in Drotz et al. .
Population structure was calculated as described by Weir and Cockerham  and genotypic linkage disequilibrium was tested for between loci per population. To evaluate if reticulation patterns influence population differentiation as assumed by sexual conflict in combination with assortative mating –, we tested for differences in gene diversity, observed heterozygosity and Fst among groups consisting of specimens within populations with mixed and homogeneous reticulation patterns. Populations were grouped according to their variability in secondary reticulation observed as described above. Significant levels were calculated via a permutation scheme of 5000 replicates where the total sample is allocated at random to groups. The P-value is based on a two-tailed probability test of the proportion of the randomised data sets that gives larger mean values than the observed. All analyses were conducted in the FSTAT v2.91 software . Pearson's correlation coefficient was estimated between altitude and mean heterozygosity and Fis.
DNA Amplification and Sequencing
DNA was extracted from one hind leg or the thoracic flight muscles of the frozen or alcohol preserved beetles, with a Qiagen DNeasy protocol for animal tissues. Partial cytochrome b (Cyt b) sequences were amplified with the two flanking primers CB-J-10933 and CB-N-11367, partial cytochrome c oxidase subunit I (COI) with CI-J-2195 and TL2-N-3014, and cytochrome oxidase II (COII) with C2-J-3279 and C2-N-3661 as described in Simon et al.  using the following polymerase chain reaction (PCR) program: denaturation 94°C (90 s), annealing at 50°C (30 s) and extension at 72°C (60 s). This cycle was repeated 30 times, followed by an extension period of 5 minutes. The PCR products obtained were single clear bands with no signs of non-specific amplification. The amplified product was approximately 430 bp in length for Cyt b, 730 bp for COI, and 400 bp for COII. The product was run on a 1.0 % agarose gel and then removed from the gel and purified with a Jetsorb DNA extraction kit. Sequencing reactions were performed with the DYEnamic ET terminator kit. Each sequence was sequenced from both its 3′ and 5′ ends. Corresponding accession number for the three partial mtDNA genes are AF439354-55, AF439359, AF439362, AF439365, AF439367-68, AF439372-77, and AY535285-400.
The mtDNA sequences were aligned with the ClustalW multiple alignment option in BioEdit, version 4.8.10 . No gaps were inserted within the alignment. Congruence among data sets was tested with the partition homogeneity test . Unweighted parsimony analysis was performed by applying heuristic search with tree bisection-reconnection branch swapping. A total of 3000 searches were done with 100 replicates and ten random-addition sequence iterations per search started from a random tree. Branches were collapsed if branch length was minimum zero. All characters were nonadditive, and uninformative characters were excluded before the analysis. To evaluate if the data set is subjected to ‘long branch attraction’ we compared the strict consensus tree topology between two phylogenetic analysis; the first including all sequences from all outgroups and the second analysis including only the sequences from the ingroup as described by Bergsten . The second dataset was also analysed with a maximum likelihood analysis. The Akaike Information Criterion in Modeltest  was used to find, the best evolutionary model given the data. Cladograms were rooted between the ingroup and outgroup according to Nixon and Carpenter . Nodal support within the phylogenetic trees within the parsimony analysis was assessed with bootstrap percentage after 1000 re-sampling steps  and within the maximum likelihood analysis with jackknife percentage after 1000 re-sampling steps with 30% character deletion . The number of extra steps required to collapse each clade was also calculated as described by Bremer . Measures of how well all individual character fit on a phylogenetic tree is measured by the consistency index (CI). The average value is calculated by dividing the minimum possible number of steps by the observed number of steps on the tree. The retention index (RI) measures the amount of synapomorphy expected from a data set that is retained as synapomorphy on a tree. The above analyses were run in both PAUP v.4.0b10  and WinClada ver. 0.9.99 . Tajima's Neutrality test for the mtDNA sequence data  was performed with DNAsp  and the assumption of a local or global molecular clock was tested with PHYHYP v. 1 . The Kishino and Hasegawa  test was used to test for two competing trees of equal length. To evaluate differences between tree topologies of unequal length we performed a permutation test where character states are randomly reassigned within characters of all taxa. Length differences between all permutated trees generate a null distribution that is used to compare the observed length differences of the original topologies. Significant level is gained in both analyses from a two-tailed probability test in PAUP v.4.0b10 .
Reticulation patterns were not optimised as a character on any of the parsimonious fundamental trees. Instead we here utilized the strict consensus to study the within and between clade distribution of reticulation patterns over all parsimonious fundamental trees. This method is well suited to distinguish if reticulation patterns have evolved multiple times within different monophyletic clades .
The secondary reticulation varies noticeably within almost all populations from France, Italy and Spain (Table 3). In three populations (BipFra4, 6 and 7), all at an altitude below 2000 m a.s.l near Grenoble, all specimens of the same sex had similar reticulation patterns. The other populations, sampled at higher altitudes, consisted of a mix of A and B patterns in various frequencies with one or several individuals representing the D, E or F patterns. The exceptional Spanish population BipEsp5 did not have any males of neither the A nor B type, and instead we here found the C, F and G patterns dominating in both sexes (Table 3). Some geographically very close populations, separated by altitude, as BipFra5 and BipFra6 or BipFra13 and BipFra14, had different dominating male elytral patterns (A and B, respectively). In males the variation is significantly correlated with altitude (p<0.001, adj-R2 = 60.2%, b = 0.79). Increased diversity of reticulation patterns is seen in several independently parallel areas in Spain and France (Fig. 3). In females, a similar pattern, however not significant, of high diversity at high altitude is also observed.
Reticulation diversity was estimated with the Simpson diversity index (D) based on the data given in table 3. Low index value (1-D) indicates small level of variation in the secondary reticulation. Geographically close populations are connected with lines. Altitude is given in meters. Locality codes follow table 1.
The within population variation in reticulation pattern made it difficult to assign a population into a given form. Populations were therefore divided into four groups based on the most frequent reticulation pattern in order to test for genetic differentiation between groups: (1) type A in all females and males (BipFra4, 6 and 7); (2) type A in all females and >50% type A in males (BipFra3, 8, 9 and 14); (3) type A in all females and >50% type B in males (BipFra9, 10, 12 and 13); (4) both sexes with variable reticulation patterns (BipFra1, 2, 5 and 11).
Population mean heterozygosity and the mean number of alleles per locus, within the total sample (BipFra1-14, BipIta1 and BipEsp3-7) of A. bipustulatus, ranged between 0.158–0.299 and 1.9 – 3.1, respectively in the French Alps populations, and between 0.148 – 0.293 and 1.8 – 2.3 in the Spanish populations. The single Italian dolomitanus population exhibits similar values to those observed in Spain (Table 4). In addition, the mean heterozygosity (r = -0477, p = 0.033, R2-adj = 18.5%) and inbreeding coefficient within population (FIS; r = 0.461, p = 0.041, R2-adj = 16.8%) were both significantly correlated to altitude (Fig. 4).
Estimated (•) observed mean heterozygosity (R2-adj = 0.185, P = 0.033); (○) estimated inbreeding coefficient (FIS) over populations (R2-adj = 0.168, P = 0.041). Altitude is given in meters.
No significant differences were detected between the above described reticulation groups in observed heterozygosity (p = 0.67), or genetic diversity (p = 0.44) or Fst (p = 0.45). Significant genotypic disequilibrium was only found between the Idh and Mdh loci in one French population (392 000 permutations, p-value 0.00001; adjusted P-value for 5% nominal level 0.000128).
Agabus nevadensis displays a significant deviating pattern in observed heterozygosity (p = 0.0054), genetic diversity (p = 0.0008) and allele frequencies in relation to A. bipustulatus. The mean heterozygosity of A. nevadensis is H = 0.134±0.044 and the mean number of alleles per locus is 1.9. Of the compared enzyme systems, the allele frequencies of Hk and Pgd stand out. The Pgd locus has three alleles in the Spanish A. bipustulatus sample, but only two in A. nevadensis. The allele that is missing in A. nevadensis is, interestingly, the most commonly found in A. bipustulatus in north Spain (Pgd 97), and the most commonly found allele in A. nevadensis (Pgd 102) is the least frequently found in A. bipustulatus in north Spain. A similar difference is seen in the allele frequency of the Hk locus. Throughout A. bipustulatus populations in Spain and France this locus consists of two alleles with a higher occurrence of allele Hk 100 in most populations, whereas in A. nevadensis this locus is more or less monomorphic for the Hk 102 allele in the eight studied populations. There is clearly a reproductive barrier between these two species.
Combined, the three mtDNA data sets COI, COII and Cyt b included 1236 base pairs. A total of 214 sites were parsimony-informative over all taxa, 150 within the A. tristis group, and 33 within the A. bipustulatus complex. Of the 214 informative characters, 174 were from the third, 34 from the first, and 6 from the second position. Translation of nucleic acid to amino acid sequences revealed very few parsimony-informative characters; i.e. most of the variation seen represents silent mutations. The incongruence length difference test was not significant between any combination of the three data sets. Most of the character conflicts in the data are within the separate sets, and only 2.33–4.19 % increased homoplasy was found in the total data set including all taxa.
Tajima's test of the neutral mutation hypothesis of the A. nevadensis and the A. bipustulatus complex sequences showed no significant deviation (p>0.10) between the amount of segregating/polymorphic sites and the average number of nucleotide differences, number of segregating sites (S) = 52, Theta per sequence = 13.16036 and 0.01066 per site, nucleotide diversity, Pi = 0.00826 and Tajima's D = −0.83312.
Elytral Reticulation and Phylogeny
Parsimony analysis of the combined data set, including all outgroup taxa, resulted in 13 most parsimonious (MP) trees with a length of 400 steps (uninformative characters excluded), and CI and RI of 0.65 and 0.83, respectively. These fundamental trees are combined into a strict consensus (Fig. 5). The subsequent phylogenetic analysis only including the A. bipustulatus complex and A. nevadensis specimens resulted in four MP trees with a tree length of 48 steps, CI and RI of 0.90 and 0.97, respectively. Main differences between these fundamental trees were the position of three haplotypes (BipSwe2, BipFra3 and BipPor2) and the grouping of the two clades consisting of the BipIce1, BipIce2, BipFra12, BipFra13 and the A. nevadensis, BipFra5, BipIra1, and BipMor1 specimens. The topology and subclades of the strict consensus tree from both analyses were identical, which provides strong evidence that no long branch attraction is affecting the results. The Kishino and Hasegawa test of the four MP fundamental trees of the A. bipustulatus complex resulted in one phylogenetic hypothesis with the lowest likelihood value. This tree is hereafter referred to as the “best” tree (Fig. 6).
The combined unweighted parsimony analysis includes all three genes (cytochrome b, cytochrome c oxidase subunit I and cytochrome oxidase II) and the outgroup species Agabus nebulosus, A. affinis and A. guttatus. Number of fundamental trees = 13. Bootstrap values above 50% are reported above branches. Bremer support values are reported below branches. Locality codes follow table 1.
The “best” fundamental tree, according to Kishino and Hasegawa test, from combined unweighted parsimony analysis of all three genes (cytochrome b, cytochrome c oxidase subunit I and cytochrome oxidase II), rooted with Agabus bipustulatus specimen from BipRus1 in order to visualise the deep split within the complex (group I and II), and to display unambiguous unique character state transformations, marked with (•), and homoplasious character state transformations with (○). Bootstrap values above 50% are reported above branches.
According to the Akaike Information Criterion in Modeltest, the Translational model (TIM), with a proportion of invariable sites equal to 0.8199 and a gamma distribution shape parameter of 0.7877 is the best evolutionary model, within the maximum likelihood analysis, given the data only including the A. bipustulatus complex and A. nevadensis specimens (-lnL 2954.8911 and AIC score 4125.7822). Nucleotide frequencies; (A) 0.3368, (C) 0.1501, (G) 0.1214 and (T) 0.3917 and evolutionary model parameters; A-C: 1.0000, A-G: 11.6809, A-T: 0.1441, C-G: 0.1441, C-T: 6.3908 and G-T: 1.0000. The maximum likelihood analysis resulted in six trees after 75166 rearrangements tried from a starting Neigbour Joining tree. From these trees the Kishino-Hasegawa (KH) test selected significantly one tree (ln 348.43962) to be the most likely representing the evolutionary history of the data (Fig. 7).
Likelihood representation including all three genes (cytochrome b, cytochrome c oxidase subunit I and cytochrome oxidase II). The translational model (TIM) was used along with a proportion of invariable sites equal to 0.8199 and a gamma distribution shape parameter of 0.7877. The tree is rooted with the Agabus bipustulatus specimen from BipRus1 in order to visualise the deep split within the complex (group I and II). Jackknife values above 50% are reported above branches.
Two main evolutionary lineages are present within the “best” MP tree (Fig. 6) and the ML tree (Fig. 7), hereafter referred to as group I and II. These two groups are separated by 10 unambiguous unique nucleotide transformations, which result in both a 100% bootstrap (BS) and Jackknife support (JS). Group I includes specimens from two haplotypes from Russia, two from Sweden and one from France, whereas all other specimens from Sweden, France, Spain, Island, Italy, Morocco, and Iran belong to group II. The “best” MP tree (Fig. 6) and the ML tree (Fig. 7) differ chiefly in the position of the BipSwe2 haplotype and the presence or absence of the subclade including the haplotypes BipIce1, BipIce2, BipFra12 and BipFra13, and the subclade including BipEsp4 and BipEsp5. These differences do not change the evolutionary interpretation of the morphological results of this study, however. Both kinds of phylogenetic analyses of the A. bipustulatus complex clearly demonstrate a multiple independent origin of different reticulation patterns. This pattern is evident in the kiesenwetterii form (B type) that is found in several of the monophyletic groups in the MP strict consensus tree together with haplotypes of the bipustulatus form (A type) (Fig. 5 and Fig. 7). The falcozi form also displays a similar pattern and is found within two well-supported subclades (71 and 63% BS, respectively) within the MP analysis, and basally of group II and in one of its subclades in the ML (75% JS). The dolomitanus form (B type) from Italy (BipIta1) forms a strongly supported subclade (96% BS and 99 % JS) together with the bipustulatus form (A type) from France (BipFra6) and Spain (BipEsp3). Moreover, the grouping of bipustulatus, solieri and kiesenwetterii as sister forms within several well supported subclades (BS and JS 53–93%) strengthens the argument of adaptation to altitudinal environments.
Forcing the kiesenwetterii, falcozi or solieri forms on the “best” MP fundamental tree (Fig. 6) to be monophyletic resulted in a dramatic increase in three length of 71.1 % compared to the fundamental trees. Still an even larger increase (83.3%) is seen when we add an additional topological constraint assuming A. nevadensis as sistergroup.
Agabus nevadensis is deeply nested in group II, within the A. bipustulatus complex (Fig. 5 and Fig. 7). However, the allozyme differences seen in the population genetic analysis imply that the adaptation process can lead to assortative mating and the evolution of a reproductive barrier.
We have studied the phylogeographic history and genetic variation of different reticulation patterns across the A. bipustulatus complex and A. nevadensis distribution in the West Paleartic in order to understand the complex variation in elytral reticulation that occurs in both sexes at high altitudes in Central and South Europe in contrast to the more homogeneous reticulation pattern normally observed at lower altitudes across the total distribution area.
We found that the different secondary reticulation patterns in A. bipustulatus have evolved independently at several times. The phylogenetic pattern shown here within the “best” MP fundamental and ML tree describes the A. bipustulatus complex as a deep gene tree with two major lineages broadly sympatric, between specimens with different reticulation patterns (Fig. 6). This reflects a species with a large evolutionary effective population size and high gene flow, with a recent admixture of lineages that have diverged allopatrically . The geographic mix of sampled specimens within the recognised monophyletic group I and II, of A. bipustulatus indicates that the Pyrenees and Alps have not acted as dispersal barriers in the interglacial periods, which is argued to be a major potential isolation barrier for the Iberian water beetles . Instead specimens of A. bipustulatus have been able to disperse more or less randomly into different areas during the climatic oscillations that have occurred during the late Pleistocene . In addition, specimens from both groups I and II co-occur at the Stintbäcken locality in north Sweden. There is therefore no reason to believe that the deep gene tree represents two different monophyletic genetically isolated entities, since the genetic structure of the Stintbäcken population does not differ from other A. bipustulatus populations across its distribution area .
Our results also indicate the presence of different selective regimes at different altitudes. The lack of significant genetic differences among A. bipustulatus populations representing combinations of reticulation patterns indicates that the historical legacy of a population does not hinder gene flow –. However, decreasing mean heterozygosity and increased population substructure (FIS) at different altitudes regardless of reticulation patterns shows that the selective process, is strong enough to alter/change the genetic structure of populations (Fig. 4 and Table 4). This genetic pattern is congruent with earlier findings in north Scandinavia , . Individual response in secondary reticulation at high altitude is observed as widening of the longitudinally elongated secondary meshes and increased coverage of isomorphic meshes from the apex up to half of the elytral length. Males are more affected than females (Table 3). The occurrence of especially secondary reticulation type C and G specimens in France and Italy at very low frequencies together with other reticulation patterns outside the Vall de Boí in Catalunya (Table 3) implies that crosses between reticulation patterns may give rise to patterns not observed within parents seen as increased reticulation diversity at higher altitude. The geographical and altitudinal distribution of the reticulation types and the genetic variation within the bipustulatus-complex suggest a complex evolution. The enlarged isomorphic meshes of the secondary reticulation at high altitude in females and the evolution of wider male tarsi and attached suckers in the presence of modified reticulated females could indeed contribute to the arms-race between sexes proposed by Howe . An alternative explanation for the increased variance in female reticulation at high altitude could be that it arises as a side effect of the greater variation of elytral reticulation in males in these populations.
The high altitudinal populations of A. nevadensis stand out in comparison to the genetic variation found among populations of A. bipustulatus and their phylogenetic position deeply nested within the A. bipustulatus group II (Fig. 6 and 7). The large similarity of the primary and secondary reticulation in A. nevadensis to that of the kiesenwetterii form (type B, Table 2) suggests that the initial altitudinal adaptation follows the same selection process as the other reticulation patterns in A. bipustulatus (type A). However, the significant differences in allele frequencies between A. nevadensis in Sierra Nevada and A. bipustulatus populations in Spain, France, Italy and Sweden at different altitudes strongly indicate that A. nevadensis has relatively recently became reproductively isolated in situ –.
Drotz et al.  documented that selection acted on α-Glycerophosphate dehydrogenase (α-Gpdh) in A. bipustulatus between the valley floor and above the tree line in Scandinavia. This selective force is, however, not responsible for the observed reticulation variation seen in both A. nevadensis and A. bipustulatus since it occurs across the species total distribution area and affect all analysed forms of both species regarding their reticulation pattern . Studies of the leaf beetle Chrysomela lapponica could, however, give some clue to a possible selective candidate. This species displays a patchy distribution of elytral colour variation with large dark black spots in northern Europe and high mountainous regions in Western Europe. In central Europe, brightly red coloured beetles are more common at low altitude. Gross et al.  demonstrated empirically that the increase in melanic elytra in northern Europe and high altitude represent a selective thermal advantage at low temperatures in relation to the brightly coloured central European form. Here the size of the dark black spots regulates the body temperature. Similar demographic distribution and colour morph variation is seen in Coelophora inaequalis, Harmonia axyridis and Oreina sulcata –. The colour variation within these species is assumed to be driven by solar UV-β radiation and interestingly solar UV-β radiation normally increases with increasing altitude. This implies that the strongest selective effect within a geographical area is overlapping with our observed deviation in reticulation pattern of A. bipustulatus and A. nevadensis. The Alps, Pyrenees and Sierra Nevada display the highest UV levels in Europe as a consequence of their high altitude, snow-covered surface and low aerosol levels . In addition, the effect of UV radiation in aquatic environments is largest in shallow mountain lakes with high incident flux and deep penetration of UV radiation . This habitat is often associated with A. bipustulatus species at high altitude . Here different reticulation patterns may reflect UV radiation differently where more impressed primary reticulation can lead to increased body temperature .
Further research is needed in order to fully understand the interplay between the selective forces that clearly plays a part in forming the reticulation pattern observed within the A. bipustulatus complex and their combined importance for adaptive/sympatric speciation. One way to do this could be to address how the relative strength of sexual and natural selection changes at different altitudes, by analysing male reticulation and suction cup variation in relation to the female reticulation patterns at different altitudes. This, however, is beyond the scope of this paper.
The authors thanks Astrid Höglund how provided assistance with the electrophoresis, and Siv Sääf with mtDNA sequencing. Carmen E. Sainz Cantero, Ignacio Ribera, Stefan Ericsson, Bertil Andrén, Shidi O. Hosseinie kindly provided specimens, and Johannes Bergsten did fieldwork in Canada. Carmen E. Sainz Cantero helped with providing the collecting permission in Sierra Nevada from the Junta de Andalucia, and Per Sundberg allowed us to use his laboratory.
Conceived and designed the experiments: MKD. Performed the experiments: MKD. Analyzed the data: MKD TB. Contributed reagents/materials/analysis tools: MKD TB ANN. Wrote the paper: MKD.
- 1. Sharp D (1882) On aquatic carnivorous Coleoptera or Dytiscidae. The Scientific Transactions of the Royal Dublin Society Se 2 Z: 17–1003.D. Sharp1882On aquatic carnivorous Coleoptera or Dytiscidae.The Scientific Transactions of the Royal Dublin Society Se 2 Z171003
- 2. Balfour-Browne F (1950) British Water Beetles Vol 2. 394. London: The Ray Society. F. Balfour-Browne1950British Water Beetles Vol 2.394LondonThe Ray Society
- 3. Khalaf KT (1980) Micromorphology of beetle elytra, using simple replicas. Florida Entomologist 63: 307–340.KT Khalaf1980Micromorphology of beetle elytra, using simple replicas.Florida Entomologist63307340
- 4. Zimmermann A (1919) Die Schwimmkäfer des Deutschen entom. Mus. zur Berlin-Dahlem. Archiv für Naturgeschicht 83: 107–249.A. Zimmermann1919Die Schwimmkäfer des Deutschen entom. Mus. zur Berlin-Dahlem.Archiv für Naturgeschicht83107249
- 5. Zimmermann A (1920) Dytiscidae, Haliplidae, Hygrobiidae, Amphizoidae. In: Schenkling S, editor. Coleopterorum Catalogus. Berlin: W. Junk. A. Zimmermann1920Dytiscidae, Haliplidae, Hygrobiidae, Amphizoidae.S. SchenklingColeopterorum CatalogusBerlinW. Junk
- 6. Zimmermann A (1934) Monographie der paläarktischen Dytiscidae. V. Colymbetinae (1. Teil). Koleopterologische Rundscha 20: 138–214.A. Zimmermann1934Monographie der paläarktischen Dytiscidae. V. Colymbetinae (1. Teil).Koleopterologische Rundscha20138214
- 7. Guignot F (1931) Les Hydrocanthares de France. Toulouse: Les Frères Douladoure. F. Guignot1931Les Hydrocanthares de France.ToulouseLes Frères Douladoure
- 8. Nilsson AN (2001) Dytiscidae (Coleoptera) world catalogue of insects 3. Stenstrup: Apollo Books. AN Nilsson2001Dytiscidae (Coleoptera) world catalogue of insects 3.StenstrupApollo Books
- 9. Bergsten J, Töyra A, Nilsson AN (2001) Intraspecific variation and intersexual correlation in secondary sexual characters of three diving beetles (Coleoptera, Dytiscidae). Biological Journal of the Linnean Societ 73: 221–232.J. BergstenA. TöyraAN Nilsson2001Intraspecific variation and intersexual correlation in secondary sexual characters of three diving beetles (Coleoptera, Dytiscidae).Biological Journal of the Linnean Societ73221232
- 10. Miller KB (2003) The phylogeny of diving beetles (Coleoptera: Dytiscidae) and the evolution of sexual conflict. Biological Journal of the Linnaean Societ 79: 359–388.KB Miller2003The phylogeny of diving beetles (Coleoptera: Dytiscidae) and the evolution of sexual conflict.Biological Journal of the Linnaean Societ79359388
- 11. Bergsten J, Miller KB (2007) Phylogeny of diving beetles reveals a coevolutionary arms race between the sexes. PLoS ONE 2(6): e522.J. BergstenKB Miller2007Phylogeny of diving beetles reveals a coevolutionary arms race between the sexes.PLoS ONE2(6)e522
- 12. Bergsten J (2005) J. Bergsten2005Taxonomy, phylogeny, and secondary sexual character evolution of diving beetles, focusing on the genus Acilius. Umeå: Print & Media. 36. Taxonomy, phylogeny, and secondary sexual character evolution of diving beetles, focusing on the genus Acilius. Umeå: Print & Media. 36.
- 13. Howe JW (2008) Variation in both male and female secondary sexual characters in the diving beetle, Agabus bipustulatus (L.) (Coleoptera: Dytiscidae). The Plymouth Student Scientist 1 (2): 56–86.JW Howe2008Variation in both male and female secondary sexual characters in the diving beetle, Agabus bipustulatus (L.) (Coleoptera: Dytiscidae).The Plymouth Student Scientist 1(2)5686
- 14. Bilton DT, Thompson A, Foster G (2008) Inter- and intra-sexual dimorphism in the diving beetle Hydroporus memnonius Nicolai (Coleoptera:Dytiscidae). Biological Journal of the Linnean Societ 94: 685–697.DT BiltonA. ThompsonG. Foster2008Inter- and intra-sexual dimorphism in the diving beetle Hydroporus memnonius Nicolai (Coleoptera:Dytiscidae).Biological Journal of the Linnean Societ94685697
- 15. Parker GA (1979) Sexual selection and sexual conflicts. In: Blum MS, Blum NA, editors. Sexual selection and sexual conflicts. New York: Academic Press. pp. 123–166.GA Parker1979Sexual selection and sexual conflicts.MS BlumNA BlumSexual selection and sexual conflictsNew YorkAcademic Press123166
- 16. Rowe L, Arnqvist G, Sih A, Krupa JJ (1994) Sexual conflict and the evolutionary ecology of mating patterns: water striders as a model system. Trends in Ecology & Evolution 9: 289–293.L. RoweG. ArnqvistA. SihJJ Krupa1994Sexual conflict and the evolutionary ecology of mating patterns: water striders as a model system.Trends in Ecology & Evolution9289293
- 17. Schluter D (2001) Ecology and the origin of species. Trends in Ecology & Evolutio 16: 372–380.D. Schluter2001Ecology and the origin of species.Trends in Ecology & Evolutio16372380
- 18. Via S (2001) Sympatric speciation in animals: the ugly duckling grows up. Trends in Ecology & Evolutio 16: 381–390.S. Via2001Sympatric speciation in animals: the ugly duckling grows up.Trends in Ecology & Evolutio16381390
- 19. Via S (2002) The ecological genetics of speciation. The American Naturalist 159: S1–S7.S. Via2002The ecological genetics of speciation.The American Naturalist159S1S7
- 20. Doebeli M, Dieckmann U (2000) Evolutionary branching and sympatric speciation caused by different types of ecological interactions. The American Naturalist 156: 77–S101.M. DoebeliU. Dieckmann2000Evolutionary branching and sympatric speciation caused by different types of ecological interactions.The American Naturalist15677–S101
- 21. Bolnick DI, Doebeli M (2003) Sexual dimorphism and adaptive speciation: Two sides of the same ecological coin. Evolution 57(11): 2433–2449.DI BolnickM. Doebeli2003Sexual dimorphism and adaptive speciation: Two sides of the same ecological coin.Evolution57(11)24332449
- 22. Kirkpatrick M, Nuismer SL (2004) Sexual selection can constrain sympatric speciation. Proc R Soc Lond B 271: 687–693.M. KirkpatrickSL Nuismer2004Sexual selection can constrain sympatric speciation.Proc R Soc LondB 271687693
- 23. Grahame WJ, Wilding CS, Butlin RK (2006) Adaptation to a steep environmental gradient and an associated barrier to gene exchange in Littorina saxatilis. Evolution 60: 268–278.WJ GrahameCS WildingRK Butlin2006Adaptation to a steep environmental gradient and an associated barrier to gene exchange in Littorina saxatilis.Evolution60268278
- 24. Knowles LL, Carstens CB (2006) Estimating a geographically explicit model of population divergence. Evolution 43: 477–493.LL KnowlesCB Carstens2006Estimating a geographically explicit model of population divergence.Evolution43477493
- 25. Falkenström G (1940) Beitrag zur festen Begründung einiger Dytisciden- Arten, bzw. Zur Absonderung aus dem System anderer dergleichen, unter denen mehrere nur Kreuzungs-Produkte darstellen. Arkiv för Zoologi 32: 1–47.G. Falkenström1940Beitrag zur festen Begründung einiger Dytisciden- Arten, bzw. Zur Absonderung aus dem System anderer dergleichen, unter denen mehrere nur Kreuzungs-Produkte darstellen.Arkiv för Zoologi32147
- 26. Franciscolo ME (1974) Idrocantari di alcune località della Alpi occidentali. Bollettino del Museo Civico di Storia Naturale 1: 345–357.ME Franciscolo1974Idrocantari di alcune località della Alpi occidentali.Bollettino del Museo Civico di Storia Naturale1345357
- 27. Drotz MK, Nilsson AN, Saura A (2001) The species delimitation problem applied to the Agabus bipustulatus complex (Coleoptera, Dytiscidae) in north Scandinavia. Biological Journal of the Linnean Society 73: 11–22.MK DrotzAN NilssonA. Saura2001The species delimitation problem applied to the Agabus bipustulatus complex (Coleoptera, Dytiscidae) in north Scandinavia.Biological Journal of the Linnean Society731122
- 28. Lindberg H (1939) Inventa entomologica itineris Hispanici et Maroccani, quod a. 1926 fecerunt Harald et Håkan Lindberg. XXIII In Spanien und Marokko gefundene Coleoptera Adephaga. Commentationes Biologicae 7: 1–35.H. Lindberg1939Inventa entomologica itineris Hispanici et Maroccani, quod a. 1926 fecerunt Harald et Håkan Lindberg. XXIII In Spanien und Marokko gefundene Coleoptera Adephaga.Commentationes Biologicae7135
- 29. Seidlitz G (1887) Bestimmungs-Tabelle der Dytiscidae und Gyrinidae des europäischen Faunengebietes. Verhandlungen des Naturforschenden Vereines in Brünn 25: 3–136.G. Seidlitz1887Bestimmungs-Tabelle der Dytiscidae und Gyrinidae des europäischen Faunengebietes.Verhandlungen des Naturforschenden Vereines in Brünn253136
- 30. Fresneda J, Hernando C (1988) Los Hydradephaga de la Alta Ribagorza y Valle de Arán. Eos; Revista Española de Entomología 64: 17–55.J. FresnedaC. Hernando1988Los Hydradephaga de la Alta Ribagorza y Valle de Arán.Eos; Revista Española de Entomología641755
- 31. Ribera I, Nilsson AN, Vogler AP (2004) Phylogeny and historical biogeography of Agabinae diving beetles (Coleoptera) inferred from mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 30: 545–562.I. RiberaAN NilssonAP Vogler2004Phylogeny and historical biogeography of Agabinae diving beetles (Coleoptera) inferred from mitochondrial DNA sequences.Molecular Phylogenetics and Evolution30545562
- 32. Franciscolo ME (1979) Coleoptera. Haliplidae, Hygrobiidae, Gyrinidae, Dytiscidae. Fauna d'Italia 14: 11–804.ME Franciscolo1979Coleoptera. Haliplidae, Hygrobiidae, Gyrinidae, Dytiscidae.Fauna d'Italia1411804
- 33. Simpson HE (1949) Measurement of Diversity. Nature 163: 688.HE Simpson1949Measurement of Diversity.Nature163688
- 34. Shaw RC, Prasad R (1970) Starch gel electrophoresis of enzymes - A compilation of recipes. Biochemical Genetics 4: 297–320.RC ShawR. Prasad1970Starch gel electrophoresis of enzymes - A compilation of recipes.Biochemical Genetics4297320
- 35. Weir SB, Cockerham CC (1984) Estimating F-statistics for the analysis of population structure. Evolution 38: 1358–1370.SB WeirCC Cockerham1984Estimating F-statistics for the analysis of population structure.Evolution3813581370
- 36. Gavrilets S (2000) Rapid evolution of reproductive barriers driven by sexual conflict. Nature 403: 886–889.S. Gavrilets2000Rapid evolution of reproductive barriers driven by sexual conflict.Nature403886889
- 37. Gavrilets S, Waxman D (2002) Sympatric speciation by sexual conflict. Proceedings of the National Academy of Sciences 99: 10533–10538.S. GavriletsD. Waxman2002Sympatric speciation by sexual conflict.Proceedings of the National Academy of Sciences991053310538
- 38. Goudet J (2001) FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3). J. Goudet2001FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3).Available from Http://wwwunilch/izea/softwares/fstathtml. Available from Http://wwwunilch/izea/softwares/fstathtml.
- 39. Simon C, Frati F, Beckenbach A, Crespi B, Liu H, et al. (1994) Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Annals of the Entomological Society of America 87: 686–701.C. SimonF. FratiA. BeckenbachB. CrespiH. Liu1994Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers.Annals of the Entomological Society of America87686701
- 40. Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95–98.TA Hall1999BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT.Nucleic Acids Symposium Series419598
- 41. Farris SJ, Källersjö M, Kluge GA, Bult C (1995) Testing significance of incongruence. Cladistics 10: 315–319.SJ FarrisM. KällersjöGA KlugeC. Bult1995Testing significance of incongruence.Cladistics10315319
- 42. Bergsten J (2005) A review of long-branch attraction. Cladistics 21: 163–193.J. Bergsten2005A review of long-branch attraction.Cladistics21163193
- 43. Posada D, Crandall KA (1998) Modeltest: testing the model of DNA substitution. Bioinformatics 14(9): 817–818.D. PosadaKA Crandall1998Modeltest: testing the model of DNA substitution.Bioinformatics14(9)817818
- 44. Nixon CK, Carpenter MJ (1993) On outgroups. Cladistics 9: 413–426.CK NixonMJ Carpenter1993On outgroups.Cladistics9413426
- 45. Felsenstein J (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783–791.J. Felsenstein1985Confidence limits on phylogenies: An approach using the bootstrap.Evolution39783791
- 46. Farris JS, Albert VA, Källersjö M, Lipscomb D, Kluge AG (1996) Parsimony Jackknifing outperforms Neighbor-Joining. Cladistics 12: 99–124.JS FarrisVA AlbertM. KällersjöD. LipscombAG Kluge1996Parsimony Jackknifing outperforms Neighbor-Joining.Cladistics1299124
- 47. Bremer K (1994) Branch support and tree stability. Cladistics 10: 295–304.K. Bremer1994Branch support and tree stability.Cladistics10295304
- 48. Swofford DL (2000) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4.0b4. Sunderland, USA: Sinauer Associates. DL Swofford2000PAUP*Phylogenetic Analysis Using Parsimony (*and Other Methods)Version 4.0b4. Sunderland, USASinauer Associates
- 49. Nixon K (1999) WinClada ver. 0.9.99m24 (Beta). K. Nixon1999WinClada ver. 0.9.99m24 (Beta).Available from http://www.cladistics.com. Available from http://www.cladistics.com.
- 50. Tajima F (1989) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123: 585–595.F. Tajima1989Statistical method for testing the neutral mutation hypothesis by DNA polymorphism.Genetics123585595
- 51. Rozas J, Sánchez-DelBarrio JC, Messeguer X, Rozas R (2003) DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19: 2496–2497.J. RozasJC Sánchez-DelBarrioX. MesseguerR. Rozas2003DnaSP, DNA polymorphism analyses by the coalescent and other methods.Bioinformatics1924962497
- 52. Pond LS, Frost DS, Muse VS (2005) HyPhy: hypothesis testing using phylogenies. Bioinformatics 21: 676–679.LS PondDS FrostVS Muse2005HyPhy: hypothesis testing using phylogenies.Bioinformatics21676679
- 53. Kishino H, Hasegawa M (1989) Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea. Journal of Molecular Evolution 29: 170–179.H. KishinoM. Hasegawa1989Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea.Journal of Molecular Evolution29170179
- 54. Avise JC (2000) Phylogeography: the history and formation of species. JC Avise2000Phylogeography: the history and formation of species.Harvard University Press, London. Harvard University Press, London.
- 55. Dansgaard W, Johnsen SJ, Clausen HB, Dahl-Jensen D, Gundestrup SN, et al. (1993) Evidence for general instability of past climate from a 250-kyr ice-core record. Nature 364: 218–220.W. DansgaardSJ JohnsenHB ClausenD. Dahl-JensenSN Gundestrup1993Evidence for general instability of past climate from a 250-kyr ice-core record.Nature364218220
- 56. Drotz MK (2003) MK Drotz2003Population genetics and phylogeography of the West Paleartic Agabus bipustulatus diving beetle complex. Doctoral Dissertation. Department of Molecular Biology, University of Umeå: Umeå. Population genetics and phylogeography of the West Paleartic Agabus bipustulatus diving beetle complex. Doctoral Dissertation. Department of Molecular Biology, University of Umeå: Umeå.
- 57. Slatkin M (1985) Gene flow in natural populations. Annual review of ecology and systematics 16: 393–430.M. Slatkin1985Gene flow in natural populations.Annual review of ecology and systematics16393430
- 58. Slatkin M (1987) Gene flow and geographic structure of natural populations. Science 236: 787–792.M. Slatkin1987Gene flow and geographic structure of natural populations.Science236787792
- 59. Barton NH (2001) The evolutionary consequences of gene flow and local adaptation: future approches. In: Clobert J, Danchin E, Dhondt AA, Nichols DJ, editors. Dispersal. New York: Oxford University Press. pp. 330–340.NH Barton2001The evolutionary consequences of gene flow and local adaptation: future approches.J. ClobertE. DanchinAA DhondtDJ NicholsDispersalNew YorkOxford University Press330340
- 60. Lenormand T (2002) Gene flow and the limits to natural selection. Trends in Ecology & Evolution 17: 183–189.T. Lenormand2002Gene flow and the limits to natural selection.Trends in Ecology & Evolution17183189
- 61. Nosil P, Vines TH, Funk DJ (2005) Perspective: Reproductive isolation caused by natural selection against immigrants from divergent habitats. Evolution 59: 705–719.P. NosilTH VinesDJ Funk2005Perspective: Reproductive isolation caused by natural selection against immigrants from divergent habitats.Evolution59705719
- 62. Gross J, Schmolz E, Hilker M (2004) Thermal Adaptations of the Leaf Beetle Chrysomela lapponica (Coleoptera: Chrysomelidae) to Different Climes of Central and Northern Europe. Environmental Entomology 33: 799–806.J. GrossE. SchmolzM. Hilker2004Thermal Adaptations of the Leaf Beetle Chrysomela lapponica (Coleoptera: Chrysomelidae) to Different Climes of Central and Northern Europe.Environmental Entomology33799806
- 63. Houston KJ (1979) Mosaic dominance in the inheritance of the colour patterns of Coelophora inaequalis (F.) (Coleoptera: Coccinellidae). Journal of the Australian Entomological Society 18: 45–51.KJ Houston1979Mosaic dominance in the inheritance of the colour patterns of Coelophora inaequalis (F.) (Coleoptera: Coccinellidae).Journal of the Australian Entomological Society184551
- 64. Mikhailov YE (2001) Significance of color polymorphism in mountain populations of abundant leaf beetles (Coleoptera: Chrysomelidae). Pirineos 156: 57–68.YE Mikhailov2001Significance of color polymorphism in mountain populations of abundant leaf beetles (Coleoptera: Chrysomelidae).Pirineos1565768
- 65. Crow JF (2003) C. C. Tan: A Life of Peaks and Valleys. Genetics 164: 1–4.JF Crow2003C. C. Tan: A Life of Peaks and Valleys.Genetics16414
- 66. Schmucki D, Philipona R (2002) Ultraviolet radiation in the Alps: the altitude effect. Optical Engineering 41: 3090–3095.D. SchmuckiR. Philipona2002Ultraviolet radiation in the Alps: the altitude effect.Optical Engineering4130903095
- 67. Laurion I, Marc Ventura M, Catalan J, Psenner R, Sommaruga R (2000) Attenuation of ultraviolet radiation in mountain lakes: Factors controlling the among- and within-lake variability. The American Society of Limnology and Oceanography 45: 1274–1288.I. LaurionM. Marc VenturaJ. CatalanR. PsennerR. Sommaruga2000Attenuation of ultraviolet radiation in mountain lakes: Factors controlling the among- and within-lake variability.The American Society of Limnology and Oceanography4512741288
- 68. Nilsson AN, Holmen M (1995) The aquatic Adephaga (Coleoptera) of Fennoscandia and Denmark. II. Dytiscidae. Fauna Entomologica Scandinavica 32: 1–192.AN NilssonM. Holmen1995The aquatic Adephaga (Coleoptera) of Fennoscandia and Denmark. II. Dytiscidae.Fauna Entomologica Scandinavica321192
- 69. Pope RD, Hinton HE (1977) A preliminary survey of ultraviolet reflectance in beetles. Biological Journal of the Linnean Society 9: 331–348.RD PopeHE Hinton1977A preliminary survey of ultraviolet reflectance in beetles.Biological Journal of the Linnean Society9331348