Zootaxa, Allozyme differentiation among populations

Zootaxa 1945: 39–50 (2008) www.mapress.com / zootaxa/

ISSN 1175-5326 (print edition)

Copyright © 2008 · Magnolia Press

ISSN 1175-5334 (online edition)

ZOOTAXA

Allozyme differentiation among populations of the Pyrenean newt Calotriton asper (Amphibia: Caudata) does not mirror their morphological diversification ALBERT MONTORI1; GUSTAVO A. LLORENTE1 & MARIO GARCÍA-PARÍS2,3 1

Departament de Biologia Animal, Facultat de Biologia, Universitat de Barcelona, Av. Diagonal 645, 08028 Barcelona, Spain. E-mail: [email protected], [email protected] 2 Museo Nacional de Ciencias Naturales. CSIC. José Gutiérrez Abascal, 2. 28006 Madrid, Spain. E-mail: [email protected] 3 Corresponding author

Abstract The Pyrenean newt (Calotriton asper) is a morphologically diversified species endemic to the Pyrenean mountains (Western Europe) that inhabits fast running streams and mountain lakes. Given its high morphological diversity, the species has been subdivided into at least ten different taxa, subsequently treated by most authors as local forms. Herein, we examined the electrophoretic patterns produced by 20 presumptive allozyme loci in specimens of seven populations distributed over the entire geographic range of the species. Sixteen loci were monomorphic across the sampling area and only four loci were polymorphic. No diagnostic alleles of any population or population set were detected. The average number of alleles per locus was found to be extremely low, between 1.1 and 1.2 ± 0.1. Genetic divergence among populations was minimal, with a maximum divergence of Nei78 = 0.031. No correlation was shown between genetic and geographic distances (Mantel test: r = -0.29, t = -1.1, p = 0.13). Fst values were low, as would be expected for a nonfragmented population. Estimated gene flow among populations was high, with a Nm = 1.01. Cytochrome b mtDNA sequences from the two populations furthest apart only differed in a single position. According to these genetic/morphological discrepancies, we interpret the observed morphological diversification of C. asper as a product of rapid morphological change under local selection pressure, in response to population specific ecological conditions. The implication of our findings for conservation efforts is that we need to preserve the unique evolutionary processes occurring in single populations or small groups of populations, even if the populations involved cannot be taxonomically differentiated. Key words: Systematics, Taxonomy, Protein electrophoresis, Genetic differentiation, mtDNA, Morphology, Spain

Introduction The recently resurrected genus Calotriton (Caudata: Salamandridae) includes two especies of newts geogaphically restricted to the Pyrenean and Pre-Pyrenean mountains in southwestern Europe (Carranza & Amat 2005). Calotriton arnoldi Carranza & Amat 2005 is an endemic taxon of the Montseny Mountains of Catalonia (Spain), while Calotriton asper (Dugès 1852) is distributed over most of the Pyrenean Range (GarcíaParís et al. 2004). The Pyrenean newt, Calotriton asper, is present over most of the Pyrenees, a mountain range that forms the boundary between Spain and France and includes the small country of Andorra. The Pyrenees are formed by a main central backbone, that runs west-eastwards from the Cantabrian to the Mediterranean seas and whose highest peaks rise to 3404 m. There is also a series of smaller chains, the Pre-Pyrenean mountains, which run mostly parallel to the main range on its southern side, forming deep valleys that drain into the Ebro River towards the Mediterranean Sea. Calotriton asper occupies most of the streams of the central axis from altitudes of 360 m to 3300 m, and also inhabits sites that are isolated to a lesser or greater extent all along the Accepted by M. Vences: 23 Oct. 2008; published: 28 Nov. 2008

39

Pre-Pyrenean ranges (Montori et al. 2002). The Pre-Pyrenean mountains have locally Mediterranean-like climate conditions, with dry summers, during which most of the streams potentially occupied by these newts almost completely dry out (Martínez-Rica 1980). The high morphological divergence among populations of C. asper has been repeatedly acknowledged since its discovery (Dugès 1852) and this has resulted in numerous taxonomic descriptions: Hemitriton asper Dugès, 1852, H. cinereus Dugès 1852, H. rugosus Dugès 1852, H. punctulatus Dugès 1852, H. bibronii Dugès 1852, Euproctus pyrenaeus Duméril, Bibron et Duméril 1854, E. asper castelmouliensis Wolterstorff 1925, E. a. donceti Wolterstorff 1925, E. a. peyreladensis Wolterstorff 1925 and Molge bolivari Boscá 1918 (Dugès op cit., Duméril et al. 1854, Boscá 1918, Despax 1923, Wolterstorff 1925). Most of these local forms, although often described as full species, were subsequently treated as subspecies or individual variations within a single polymorphic species (Boulenger 1917, Mertens & Wermuth 1960, Salvador 1974). However, comprehensive morphological and ecological studies soon pointed out the existence of marked differences between a set of populations distributed across the Pre-Pyrenean ranges of Huesca (Spain), and the populations of the central Pyrenean axis (Clergue-Gazeau & Martínez-Rica 1978, Martínez-Rica 1980, Serra-Cobo et al. 2000). The Pre-Pyrenean populations have elongated epibranchial prominences, more developed cutaneous glands and a narrower neck, and show a characteristically low frequency of certain allelic variants at the serum transferrin protein locus. Both morphological and preliminary protein data suggest that these Pre-Pyrenean populations have become isolated from the central axis (Gasser and Clergue-Gazeau1981). Morphological diversification has also given rise to populations in which larval characters are retained by the adults (Campeny et al. 1984, 1986). The Pre-Pyrenean mountains of Huesca in Spain are geologically and climatically distinct (Serra-Cobo et al. 2000) and their C. asper populations are confined to a few shady streams running deep in canyons or covered by dense bush vegetation. In eastern Catalonia, the Pre-Pyrenean mountain chains run perpendicularly (N-S) to the Pyrenean axial chain and are connected to the Pre-Littoral mountain chain. This particular orography of the Catalonian Pre-Pyrenean gives rise to an ecological continuity, which takes the form of a gradient from high pasturelands to typical Mediterranean shrub landscapes at lower altitudes. Calotriton is able to colonize most of this gradient and only the southernmost populations (Montseny) corresponding to C. arnoldi are geographically isolated (Montori & Campeny 1992, Carranza & Amat 2005). The morphological singularility of the Pre-Pyrenean populations is suggestive of their genetic divergence from populations of the main Pyrenean axis. If so, their previous taxonomic subdivisions might have genetic support. Alternatively, their current morphological distinctiveness could be the result of rapid morphological change driven by active selection under intense ecological pressure, which might not be reflected at the genetic level since time elapsed is not enough to get a clear signal of genetic differentiation. Here, we performed an electrophoretic analysis of 13 enzyme systems (20 loci) in seven populations distributed across the entire geographical range of the species, and estimated gene flow and dispersal among populations.

Material and methods Specimens of seven populations were obtained from deep mountain valleys in the western, central and eastern Pyrenean axial chain, and from two isolated populations of the Pre-Pyrenean mountains of Huesca and Barcelona, respectively (Fig. 1). Since C. asper is described by Spanish legislation as an endangered species, only 5 to 9 specimens were collected from each population. Previous studies have shown that natural population sizes are generally large (estimated at 3637 individuals with a 95% interval of 3166-4336 individuals for 1.5 km of a representative Pyrenean stream using Chapman’s modification (Chapman, 21951) of the LincolnPetersen method (Montori et al., 2008), such that the numbers sampled are not likely to represent a threat for these populations.

40 · Zootaxa 1945 © 2008 Magnolia Press

MONTORI ET AL.

FIGURE 1. Map of the Pyrenees, comprising the complete geographic distribution of Calotriton asper (not delineated). The sampling sites for the allozyme study are indicated by numbers: 1.- Zuriza, 2.- Espelunciecha, 3.- Piedrafita, 4.- San Juan de la Peña, 5.- Fanlo, 6.- Pi, 7.- Susqueda. The solid circle represents Calotriton arnoldi populations.

Genetic variability among populations was quantified by allozyme electrophoresis. Specimens were processed in the field and samples of liver, heart and muscle were frozen in liquid nitrogen and later stored at –70º C. Homogenized tissues were analyzed by allozyme electrophoresis on 12% horizontal starch gels under three different buffer systems (tris-citrate pH6, tris-citrate pH8 and LiOH) (see Buckley et al. 1994 and Alcobendas et al. 1996 for details). We examined 13 enzyme systems representing the following 20 presumptive gene loci: aspartate amino transferase (AAT-1 and AAT-2, EC 2.6.1.1.), alcohol dehydrogenase (ADH-1 and ADH-2, EC 1.1.1.1), esterase (EST, EC 3.1.1.), glucose-6-phosphate dehydrogenase (G6PD, EC 1.1.1.49), isocitrate dehydrogenase (ICD-1 and ICD-2, EC 1.1.1.42), malate dehydrogenase (MDH-1 and MDH-2, EC 1.1.1.37), mannose phosphate isomerase (MPI-1 and MPI-2, EC 5.3.1.8), 6-phosphogluconatedehydrogenase (6PGD, EC 1.1.1.43), xanthine dehydrogenase (XDH-1 and XDH-2, EC 1.2.1.37), sorbitol dehydrogenase (SDH, EC 1.1.1.14), alpha glycerophosphate dehydrogenase (a-GLY, EC 1.1.1.8), leucine amino peptidase (LAP, EC 3.4.11.1), and malic enzyme (MOD-1 and MOD-2, EC 1.1.1.40). Genotype frequencies were obtained by direct counting and subsequently used to compute allele frequencies. Divergence among populations was estimated as Nei’s unbiased genetic distances (1978) calculated using BIOSYS-1 software (Swofford & Selander 1989). Allozyme polymorphism was given by percentages of polymorphic loci. We estimated both observed (Ho) and expected heterozygosity values (He). While a decrease in heterozygosity due to departure from the Hardy-Weinberg equilibrium can be indicative of some sort of non-random mating within the taxon studied, F-statistics can be used to explain a decrease in heterozygosity by the effects of both non-random mating and finite population sizes due to subdivision (Hartl & Clark 1989, Caballero 1994). Both He and Ho were used to test for departure from Hardy-Weinberg equilibrium, conducting X2 tests for goodness-of-fit with Levene’s (1949) correction for small sample sizes. As this test cannot be applied confidently when expected values are low, genotypes were pooled into three classes when there were more than two alleles at a locus. These classes were: (i) homozygotes for the most common allele, (ii) heterozygotes for the most common allele and one of the other alleles, (iii) all other genotypes. Chisquared tests were performed using exact probabilities. Departures from Hardy-Weinberg equilibrium were only considered significant if the three tests (STEP HDYWBG, OPTIONS LEVENE and EXACTP) implemented in the BIOSYS-1 program were significant. The genetic structure of the taxon was described using Wright’s hierarchical F-statistics (1978). Fis, Fit and Fst (STEP WRIGHT-78 in BIOSYS-1) were calculated and corrected for sample errors. Values given are averaged over all polymorphic loci and populations, although Fst values were also computed for pairs of populations. A test for homogeneity of allele frequencies among populations was also carried out by chi-squared testing using STEP HETXSQ (in BIOSYS-1). The jackknife procedure (Weir 1990) was applied to the loci and populations to obtain a more accurate value of Fst.

ALLOZYME DIFFERENTIATION IN CALOTRITON ASPER

Zootaxa 1945 © 2008 Magnolia Press ·

41

We employed Mantel’s non-parametric test (Mantel & Valand 1970) using the NTSYS program (OPTION MXCOMP) (Rohlf 1993) to estimate the degree of correlation between Fst coefficients, Nei’s unbiased genetic distances and geographic distances. Geographic distances were defined as the shortest air distance between populations. Two indirect methods of calculating gene flow were employed to estimate the pattern of genetic exchange among newt populations (Wright 1943, Nei 1975, Slatkin 1985). The first method uses the relation between Fst and the number of migrants per generation (Nm) given by the formula Fst = 1/(4Nm+1) (Wright 1943). The second method estimates the maximum migration rate among populations (m) by means of the formula: I = m/(m+v), where I is Nei’s coefficient of genetic identity and v is the mutation rate (Nei 1975). Since the rate of mutation for proteins has not been calculated in newts, we opted for a value of 2 x 10-6 mutations per locus per generation as an approximation; this value has been used by other authors in similar calculations for human populations (Nei 1975) and for other urodeles (Larson et al. 1984, Ragghianti & Wake 1986, Crnobrnja & Kalezic 1990). A third method, known as the private alleles method (Slatkin 1985) gives estimates of Nm. However, its use was not considered feasible in our study due to the very low number of "private" alleles present. According to Slatkin (1985), 20 or more private alleles are needed to give a reliable estimate of Nm (we only found two in our populations). To obtain additional evidence of migration, we conducted a field study along the Pi stream valley, which is a typical habitat of this species. A preestablished 1500 m length reach of the stream was exhaustively explored, both visually and by removing stones upstream, by two researchers. All postmetamorphic newts were captured and marked by toe-clipping, their snout-vent lengths (SVL) were measured with a vernier caliper, and the newts then released on site. A second search was carried out 17 days after the first search following the same procedure. We compared the mean SVLs of the specimens found in the first search with those of the specimens found in the second search. A second statistical comparison including only recaptured specimens was used as a broad indication of the age structure (metamorphic, subadults, or adults) of the population fraction lost due to migration. We also assessed mitochondrial differentiation by examining short (385 base pairs) mitochondrial DNA (mtDNA) sequences of the cytochrome b (cyt b) gene of specimens from two distant populations (population 5 Fanlo and population 1 Zuriza) (Fig. 1). All DNA extraction, amplification and sequencing procedures were performed as indicated in García-París et al. (2003) (Genbank accession numbers FJ403325-326).

Results Through the electrophoretic analysis of 13 enzyme systems, we detected 20 presumptive loci. Sixteen loci were monomorphic across the individuals sampled. The four polymorphic loci observed were MPI-1, ADH-1, a-GLY and G6PD. Table 1 shows the population allele frequencies recorded for the four variable loci. No alleles were diagnostic for any population or population set (Table 1). The proportions of polymorphic loci per population were low, from 2 (10%) to 3 (15%) (Table 2). The average number of alleles per locus, 1.1 to 1.2 ± 0.1, was extremely low. Observed heterozygosity ranged from 0.022 to 0.12 (for the Fanlo and Susqueda populations, respectively) (Table 2). The mean heterozygosity observed across loci and populations was 0.063 (expected heterozygosity 0.060). No significant departures from Hardy-Weinberg equilibrium were detected among the polymorphic loci. Three of the polymorphic loci (MPI-1, ADH-1 and a-GLY) were heterogeneous across populations (p = 0.0002, 0.0015, and 0.012, respectively). The Cyt b sequences of the Fanlo (pop. 5) and Zuriza (pop. 1) populations showed a difference of only one base pair.


MONTORI ET AL.

TABLE 1. Allellic frequencies at polymorphic loci. Populations: 1. Zuriza, 2. Espelunciecha, 3. Piedrafita, 4. San Juan de la Peña, 5. Fanlo, 6. Pi, 7. Susqueda. Population Locus

1

2

3

4

5

6

7

(N)

5

6

6

6

9

6

5

A

.700

.583

.500

.333

1.000

.917

.500

B

.300

.083

.250

-

-

-

.400

C

-

.333

.250

.667

-

.083

.100

(N)

7

6

6

6

9

6

5

A

.571

.583

.833

.833

1.000

.917

.600

B

.429

.417

.167

-

-

.083

.400

C

-

-

-

.167

-

-

-

(N)

7

6

6

6

9

6

5

A

.643

1.000

.500

1.000

.556

.833

.700

B

.357

-

.500

-

.444

.167

.300

(N)

7

6

6

6

9

6

5

A

-

-

-

-

.111

-

-

B

1.000

1.000

1.000

1.000

.889

1.000

1.000

MPI-1

ADH-1

?-GLY

G6PD

TABLE 2. Genetic variability at 20 loci in all populations. Ni: Sample size. Na: Mean of number of alleles per locus. P: Percentage of loci polymorphic (.95). SE: standard errors. Mean heterozygosity Population

Ni

Na

P

Observed (±SE)

Estimated (±SE)

Zuriza (1)

7

1.1±0.1

15

0.0460.026

0.074±0.041

Espelunciecha (2)

6

1.1±0.1

10

0.067±0.047

0.056±0.039

Piedrafita (3)

6

1.2±0.1

15

0.1±0.06

0.077±0.044

S.J. de la Pea (4)

6

1.1±0.1

10

0.05±0.036

0.039±0.028

Fanlo (5)

9

1.1±0.1

10

0.022±0.022

0.037±0.028

Pi (6)

6

1.1±0.1

15

0.033±0.019

0.032±0.018

Susqueda (7)

5

1.2±0.1

15

0.12±0.067

0.082±0.045

Genetic divergence among populations was minimal, with a maximum divergence of Nei78 = 0.031 estimated for the populations of San Juan de la Peña (pop. 4) and Fanlo (pop. 5). There was no correlation between genetic and geographic distances (Mantel test: r = -0.29, t = -1.1, p = 0.13). Fst values were low (Table 3), as expected for a non-fragmented population. A Mantel test indicated no correlation between the Fst values and geographic distances (r = -0.27, t = -0.98, p = 0.16). Fis values indicated an excess of heterozygotes. Estimated gene flow among populations was high (Nm=1.01). Indirect evidence of gene flow as a result of migration was provided by the data obtained during the field searches in the Pi Valley. We captured 724 specimens, which were marked and released on the day of the



43

search. In the second search 17 days later, 546 specimens were captured, 108 of which were recaptured specimens. Snout-vent lengths did not differ between the two sets of samples obtained. Snout-vent-lengths, however, did show a significant difference when we compared captured and recaptured specimens (Fig. 2) such that there was a clear reduction in the number of smaller individuals in the set of recaptured newts (MannWhitney Z = -2.77, p = 0.005).

FIGURE 2. Snout-vent-length distribution of C. asper in the Pi Valley. Above the axis, SVL of specimens from a first capture; below the axis, SVL distribution of recaptured specimens. Arrows indicate the low proportion of small adults in the recaptured sample.

Discussion Morphological versus genetic differentiation Calotriton asper inhabits fast running mountain streams and high mountain lakes across the Pyrenees. Its pronounced morphological variation has led to the description of several taxonomic units, based on morphological differences including color, size and shape along with the differential development of bone structures (Clergue-Gazeau & Martínez-Rica 1978, Martínez-Rica 1980, Serra-Cobo et al. 2000, Carranza & Amat 2005). Morphological variation in the species is greatest when we compare populations of C. asper inhabiting streams of the main backbone of the Pyrennees to isolated populations found in the Pre-Pyrenean ranges (Serra-Cobo et al. 2000). This difference cannot, however, be interpreted as the result of clinal variation (Martínez-Rica 1980). Morphological diversification can be the consequence of processes of gradual accumulation of changes through long periods of time or rapid morphological change mediated by phenotypic plasticity under local


MONTORI ET AL.

selection pressure in response to specific ecological conditions. Our results indicate both minimal allozyme differentiation across populations of Calotriton asper and genetic similarity among geographically distant populations. At the allozyme level, isolated populations from the Pre-Pyrenees, San Juan de la Peña (population 4), Pi (population 6), Susqueda (population 7) are almost identical to populations inhabiting the central Pyrenean range. In fact, the level of polymorphism for the whole data set is low compared to the degree of polymorphism shown by other salamander species with a similarly fragmented habitat such as S. salamandra (Alcobendas et al. 1996). TABLE 3. Geographic distance in km (above diagonal) and genetic distance -DNei- (below diagonal), between studied populations. Piedrafita

Espelunciecha

Zuriza

S.J. de la Peña

Fanlo

Pi

Susqueda

Piedrafita

-------

22.5

47.5

17.5

55

167.5

237.5

Espelunciecha

0.011

-------

35

40

65

177.5

252.5

Zuriza

0.001

0.005

------

42.5

97.5

210

285

S.J. de la Peña

0.016

0.007

0.025

------

82.5

190

262.5

Fanlo

0.007

0.023

0.011

0.031

------

112.5

187.5

Pi

0.008

0.008

0.007

0.016

0.003

------

77.5

Susqueda

0

0.002

0

0.018

0.015

0.009

------

TABLE 4. F-statistics at polymorphic loci. Locus

F(is)

F(it)

F(st)

MPI-1

-0.399

-0.069

0.236

ADH-1

-0.118

0.072

0.169

?-GLY

-0.150

0.064

0.186

G6PD-1

1.000

1.000

0.097

Mean

-0.204

0.036

0.199

The genetic patterns observed in C. asper, favor the hypothesis of morphological diversification as the result of phenotypic plasticity in response to different ecological conditions as the most likely explanation. Morphological variation was greater across than within populations, and local morphotypes were found to be characteristic of single populations. The populations from Fanlo (population 5) and Susqueda (population 7) are an example (Fig. 3). Mature newts from Fanlo were characterized by their small size (average SVL = 50.92), narrow heads, and smooth skin, while the newts from Susqueda were large (average SVL = 77.82), with wide heads and warty skin, and both populations lacked the typical mid-dorsal light stripe (Fig. 3). Fanlo and Susqueda are located 187 km apart (air distance) and specimens are unmistakably ascribable to either population (F = 1.12, p = 0.0000), while genetic divergence between the two populations is extremely low (DNei = 0.015). The morphological differentiation observed among populations of C. asper (García-París et al. 2004, Carranza & Amat 2005) is likely the product of recent processes. Isolated populations, sufficiently recent to show signs of molecular differentiation (with the markers used in this study), diverged morphologically to such an extent that they have been considered different taxa. This diversification is probably the consequence of diverse selective pressures associated with different environmental conditions, including conditions far from optimal for the species across the southern Pre-Pyrenean region.



45

FIGURE 3. An example of differentiation among populations of C. asper at the morphological level. Both specimens are representative of full grown adults from Fanlo (smaller, SVL 47.15 mm) and Susqueda (larger specimen, SVL 81.35 mm). Their size and shape, as in this case, together with coloration differences, have been used as indications for taxonomic subdivisions in C. asper.

Gene flow and dispersal Geographic isolation among populations of C. asper has always been described as high for the entire species range (Thorn 1968, Clergue-Gazeau & Martinez-Rica 1978, Serra-Cobo et al. 2000). Streams in deep canyons and valleys are not directly connected and migration over mountain tops was thought rather unlikely. Connections via streams joining the main river systems is possible, but fish (trout) predation and a lack of suitable habitats impairs present day movements via this route, and there are no records of C. asper inhabiting large rivers (Montori 1997, Montori & Herrero 2004, Montori et al. 2006). Specimens might have been driven downstream only by unusual storms, forcing accidental migration, although our observations indicate that, as soon as a storm starts, the newts abandon the stream climbing to the shores (Montori, unpublished data). Parameters of genetic variability found across populations are within the range of intrapopulational variability in salamandrids (Ragghianti & Wake 1986, Arano et al. 1991, Alcobendas et al. 1996). In addition to dispersal data (see below), this observation prompts the question of whether the populations of C. asper are really as isolated as they appear on morphological grounds or they form part of a large metapopulation system. Apparently there is direct (genetic) and indirect (population structure) evidence for relatively high gene exchange between populations. The possibility of migration across ridges separating adjacent valleys was indirectly inferred in an analysis of the population structure of this species in the Pi Valley region of the Catalonian Pyrenees (Montori 1988, 1990). Capture and recapture studies indicated that individual migration takes place by land rather than along the stream bed (Montori 1988). Skeletochronological analyses of a sample taken from the stream revealed a deficit of small sized adults and immature individuals in the brook (Montori


MONTORI ET AL.

1990). This is not surprising given that most of the activity of juvenile specimens occurs on land (Llorente et al. 1995, García-París et al. 2004). Newly metamorphosed newts leave the streams and remain on land until they reach sexual maturity (4-5 years [Montori 1990]). These specimens could colonize adjacent streams as suggested by a typical contagious distribution (in which the probability of finding the species in adjacent streams is higher than the probability of finding the species in any two randomly selected streams across the distribution range) across the main Pyrenean range (Montori 1988). Young adults have even been identified at the top of ridges between streams (A. Montori unpublished data). We interpret the lack of small specimens in our capture-recapture study as a result of this dispersal behavior of small adults and juveniles (under 9-11 years of age). Paleobiogeography The lineage leading to Calotriton has been independent from other salamandrids lineages since at least the last 25 million years (Sbordoni et al. 1982, Accordi et al. 1984, Caccone et al. 1996, Carranza & Amat 2005). Although divergence between the sister species, C. asper and C. arnoldi, is relatively low (Carranza & Amat 2005), we cannot evoke a recent origin of the lineage to account for such genetic uniformity. On the contrary, given the complex geological history of the area inhabited by C. asper, which is likely to have undergone dramatic climatic changes through time (Barbadillo et al. 1997), we would expect the isolation of peripheral populations as a result of range expansions and contractions, as shown for other salamander species inhabiting the same region (García-París et al. 2003). Many of the behavioral and morphological traits of C. asper have been considered adaptations to life in fast running streams (e.g., specialized courtship in which the male grasps the female during spermatophore transfer, lack of larval balancers, reduced lungs, digit keratinization etc.) (Thorn 1968) such that, consistent also with the ecological data, we could describe this species as ecologically constrained. Expected isolation due to geological and climatic changes combined with an ecological restriction to mountain habitats, makes C. asper a good candidate for multiple genetically differentiated populations, or for high overall genetic diversity as a consequence of possible contact among previously differentiated populations. Our data indicate, however, that populations of C. asper are almost undifferentiated at the allozyme level (when compared to other amphibians inhabiting the same geographic region) and that overall allozymic diversity (e.g., the number of alleles per locus) is extremely low. Homogenization due to extensive gene flow among populations could account for the observed lack of genetic differentiation, but not for the reduced allelic diversity. Hence, we propose a process of large-scale extinction of the old potentially differentiated populations followed by relatively recent recolonization, as an explanation for the genetic pattern found. Extinction has to be evoked as a major force shaping the current genetic structure of C. asper. This hypothesis also fits in with the paleogeographic evidence, since the species appears in the record of the Upper Pleistocene of the Cantabrian Mountains (Northern Spain) (Martín 1986), where it no longer exists. We suggest that late Pliocene or Pleistocene climatic changes drove Iberian populations of Calotriton to almost complete extinction, wiping out the Cantabrian Mountain populations and confining the Pyrenean ones to two southern refugia, leading to bottle-neck processes. One of these bottlenecked populations remained confined to the Montseny mountains originating C. arnoldi, while the second, corresponding to C. asper would then have been able to recolonize the entire Pyrenean range during the warmer and dryer climate periods of the last few thousands years (after the last Glacial period, no earlier than 8.000 years ago). Concurrently, the original population source would have disappeared from the warmer southern refugium, also as a consequence of warming. This hypothesis is consistent with the proposal by Gasser & Clergue-Gazeau (1981), who argued that there must be an extra-Pyrenean source for the current populations. Along its geographic expansion, it is likely that C. asper contacted with isolated peripheral populations of C. arnoldi. Levels of past nuclear introgression and reciprocal monophyly of the two lineages (García-París & Jockusch, 1999), should be addressed across populations of C. asper located near the isolated population of C. arnoldi.



47

Morphological diversity and conservation Our data, as well as the results of previous analyses (Gasser 1973, Gasser & Clergue-Gazeau 1981) do not provide support for maintaining the previously proposed taxonomic units within C. asper. However, we believe that the extreme morphological diversity of the species deserves special consideration from both an evolutionary and conservation standpoint. Conservation categories are generally based upon taxonomic units (species or subspecies). The problem of defining or delimiting subspecies is complicated, and often generates harsh debate (Ryder 1986). A subspecies can be roughly defined as a population or a set of populations in which all the specimens (or most of them) share a character or a series of characters that can be identified and used as diagnostic features (morphological traits, mtDNA haplotypes, nuclear markers etc.). In practice, this definition is eventually subjected to the criteria of the researcher, particularly with regard to character choice or to the relevance of diagnostic features. Many of the evolutionary significant units defined for conservation purposes have been based on genetic data (Fraser & Bernatchez 2001), but these units rarely take into account evolutionary processes beyond genetic differentiation (e.g., heterochrony leading to neotenic populations) (see Moritz 2002, Martínez-Solano et al. 2005). We believe that taxonomic and genetically defined units should not be considered as the only basis for conservation management, and that single populations or small groups of populations, in which unique evolutionary processes are taking place, especially warrant preservation for their protection, even though the populations involved are not taxonomically differentiated. Whether these processes are adaptive (Crandall et al. 2000) or not is not an issue, since heterochronic changes early in development might have important evolutionary consequences in salamanders (Buckley et al. 2007), so in this respect any evolutionary process, besides genetic differentiation, should be considered when defining units for conservation. Here, we show that on the scale of our study, the large morphological diversity found in C. asper does not mirror its genetic diversity, and thus any argument for defining taxa within the species will surely be subjective and debatable. Consequently, according to general conservation practices, none of the extremely differentiated populations of C. asper would be included in specific conservation plans. However, C. asper constitutes an extraordinary case of morphological plasticity. Each morphologically differentiated population of C. asper represents a remarkable evolutionary process, which is most likely the consequence of environmentally mediated differential selection pressures acting locally upon a common genetic background, coupled to, for example, developmental heterochronic processes. Accordingly, loosing any differentiated population would imply the loss of the evolutionary process leading to that particular morphology. We believe we should document the existence of locally differentiated populations and advocate the need for their protection for the conservation of biodiversity. This is particularly relevant for species with reduced geographic ranges, such as C. asper, for which the disappearance of a single population represents an important loss of intraspecific diversity.

Acknowledgments The authors thank Mercedes París and Begoña Arano for technical support during the allozyme work and Luis Serra, Iñigo Martínez-Solano, David Buckley and Gabriela Parra-Olea for their comments on the manuscript. We also thank the Environmental Agencies of Aragón and Catalonia for providing the collection permits. Grant CGL2007-64621/BOS of the Spanish Ministerio de Educación y Ciencia provided support for this study.

References Accordi, F., Grassi-Milano, E. & Gallo, V.P. (1984) The adrenal gland of Euproctus (Urodela, Salamandridae), comparison of three species and phylogenetic inferences. Journal of Anatomy, 139, 209–214.


MONTORI ET AL.

Alcobendas, M., Dopazo, H. & Alberch, P. (1996) Geographic variation in allozymes of populations of Salamandra salamandra exhibiting distinc reproductive modes. Journal of Evolutionary Biology, 9, 83–102. Arano, B., Arntzen, J.W., Herrero, P. & García-París, M. (1991) Genetic differentiation among Iberian populations of the Alpine newt, Triturus alpestris. Amphibia-Reptilia, 12, 409–421. Barbadillo, L.J., García-París, M. & Sanchiz, B. (1997) Orígenes y relaciones evolutivas de la herpetofauna ibérica. In: Pleguezuelos, J.M. (Ed.). Distribución y biogeografía de los anfibios y reptiles en España y Portugal. Monografías de Herpetología Vol 3. Universidad de Granada-Asociación Herpetológica Española, pp. 47–100. Boscá, E. (1918) Una nueva forma de Anfibio urodelo (Molge bolivari). Boletín de la Real Sociedad Española de Historia Natural, 18, 58–61. Boulenger, G. (1917) Les batraciens Urodèles rapportés au genre Euproctus, leurs rapports ethologiques et phylogeniques. Comptes rendus de l'Académie des Sciences, 164, 709–712, 801–805. Buckley, D., Alcobendas, M., García-París, M. & Wake, M. (2007) Heterochrony, cannibalism, and the evolution of viviparity in Salamandra salamandra. Evolution and Development, 9(1), 105–115. Buckley, D., Arano, B., Herrero, P., Llorente, G.A. & Esteban, M. (1994) Moroccan water frogs vs. Rana perezi: allozymes studies show up their differences. Zoologica Poloniae, 39(3-4), 393–401. Caballero, A. (1994) Developments in the prediction of effective population size. Heredity, 73, 657–679. Caccone A., Milinkovitch, M.C., Sbordoni, V. & Powell, J.R. (1996) Mitochondrial DNA rates and biogeography in European newts (Genus Euproctus). Systematic Biology, 46(1), 126–144. Campeny, R., Montori, A. & Llorente, G.A. (1984) Nota preliminar sobre la permanencia de caracters larvaris en individus adults d'una població d'Euproctus asper. Bulletin de la Societat Catalana de Ictiologia i Herpetologia, 9, 43–45. Campeny, R., Montori, A. & Llorente, G.A. (1986) Nuevos datos sobre la permanencia de caracteres larvarios en individuos adultos de una población de tritón pirenaico (Euproctus asper) en el Valle de Arán. Doñana Acta Vertebrata, 13, 170–174. Carranza, S. & Amat, F. (2005) Taxonomy, biogeography and evolution of Euproctus (Amphibia: Salamandridae), with the resurrection of the genus Calotriton and the description of a new endemic species from the Iberian Peninsula. Zoological Journal of the Linnean Society, 145, 555–582. Chapman, D.G. (1951) Some properties of the hypergeometric distribution with applications to zoological censuses. University of California Publications on Statistics. 1. 131-160.

Clergue-Gazeau, M. & Martínez-Rica, J.P. (1978) Les différents biotopes de l'urodèle pyrénéen, Euproctus asper. Bulletin de la Société d’ Histoire Naturelle de Toulouse, 114, 461–471. Crandall, K.A., Bininda-Emonds, O.R.P., Mace, G.M. & Wayne, R.K. (2000) Considering evolutionary processes in conservation biology: returning to the original meaning of "evolutionarily significant units". Trends in Ecology and Evolution, 15(7), 290–295. Crnobrnja, J. & Kalezic, M.L. (1990) The organization of the genetic variation in the Crested newt (Triturus cristatus complex) from Yugoslavia. Genetika, 22, 133–146. Despax, R. (1923) Contribution à l'étude anatomique et biologique des Batraciens Urodeles du groupe des Euproctes et spécialement de l'Euprocte des pyrénées Triton (Euproctus) asper. Bulletin de la Société d’ Histoire Naturelle de Toulouse, 51, 185–440. Dugès, A. (1852) Recherches zoologiques sur les urodeles de France. Annales de Sciences Naturelles, Zoologie, Série 3, 17, 253–272. Duméril, A.M.C. Bibron, G. & Duméril, A.H. (1854) Erpétologie générale ou Histoire Naturelle complète des Reptiles. 9, 1–440, atlas, 1–24, pls., 1–108. Roret. París. Fraser, D.J. & Bernatchez, L. (2001) Adaptative evolutionary conservation: towards a unified concept for defining conservation units. Molecular Ecology, 10, 2741–2752. García-París, M., Alcobendas, M., Buckley, D. & Wake, D.B. (2003) Dispersal of viviparity across contact zones in Iberian populations of fire salamanders (Salamandra) inferred from discordance of genetic and morphological traits. Evolution, 57, 129–143. García-París M. & Jockusch E.L. (1999) A mitochondrial DNA perspective on the evolution of Iberian Discoglossus (Amphibia: Anura). Journal of Zoology, 248, 209–218. García-París, M., Montori, M. & Herrero, P. (2004) Amphibia, Lissamphibia. Fauna Ibérica, vol. 24, Museo Nacional de Ciencias Naturales, CSIC, Madrid, pp. 1–640. Gasser, F. (1973) Cartographie des proteines sèriques de l'Urodèle pyrénéen. Euproctus asper (Dugès). Stabilité de l'aire de répartition de l'espèce. Comparative Biochemistry and Physiology, 45B, 419–443. Gasser, F. & Clergue-Gazeau, M. (1981) Les proteines sèriques de l'urodèle Euproctus asper (Dugès). Elements de différentiation génétique dans les prepyrénées espagnols. Vie et Milieu, 31, 297–302. Hartl, D.L. & Clark, A.G. (1989) Principles of Population Genetics. 2nd Ed. Sinauer Associates Inc., Sunderland, MA. Larson, A., Wake, D.B. & Yanev, K.P. (1984) Measuring gene flow among populations having high levels of genetic fragmentations. Genetics, 106, 293–308. Levene, H. (1949) On a matching problem arising in genetics. Annals of Mathematical Statistics, 20, 91–94.



49

Llorente, G.A., Montori, A., Santos, X. & Carretero, M.A. (1995) Atlas dels amfibis i rèptils de Catalunya i Andorra. El Brau, Figueres, pp. 1–191. Mantel, N. & Valand, R.S. (1970) A technique for nonparametric multivariate analysis. Biometrics, 26, 547–558. Martín, C. (1986) Morfología y registro fósil de Euproctus asper. Actas del Primer Congreso Nacional de Herpetología Segunda Reunió General de la SCIH, Benicassim., SCIH, Barcelona, pp. 72. 114pp. Martínez-Rica, J.P. (1980) Algunos datos sobre las poblaciones meridionales del tritón pirenaico Euproctus asper Dugés. Studia Oecologica, 2, 135–154. Martínez-Solano, I. Alcobendas, M., Buckley, D. & García-París, M. (2005) Molecular characterization of the endangered Salamandra salamandra almanzoris (Caudata, Salamandridae). Annales Zoologici Fennici, 42, 57–68. Mertens, R. & Wermuth, H. (1960) Die Amphibien und Reptilien Europas. Waldemar Kramer, Frankfurt am Main. Montori, A. (1988) Estudio sobre la biología y ecología del tritón pirenaico Euproctus asper (Dugès, 1852) en la Cerdanya. Tesis doctoral, Universitat de Barcelona, Barcelona. Montori, A. (1990) Skeletochronological results in the pyrenean newt Euproctus asper (Dugès, 1852) from one prepyrenean population. Annales des Sciences Naturelles, Zoologie, Série 13, 11, 209–211. Montori, A. (1997) Trophic segregation between the Brown Trout (Salmo trutta) and the Pyrenean newt (Euproctus asper). In: Herpetologia Bonnensis, 1997. Bohme, W. Bischoff, W. & Ziegler, T. (Eds). Bonn, pp. 273–278. Montori, A. & Campeny, R. (1992) Situación actual de las poblaciones de tritón pirenaico Euproctus asper en el macizo del Montseny. Boletín de la Asociación Herpetológica Española, 2, 10–12. Montori, A. & Herrero, P. (2004) Caudata. In: García-París, M., Montori, A., Herrero, P. (Eds.) Amphibia, Lissamphibia. Fauna Ibérica, vol. 24, Museo Nacional de Ciencias Naturales, CSIC, Madrid, pp. 43–275 Montori, A., Llorente, G.A., Santos, X. & Carretero, M.A. (2002) Euproctus asper. In: Pleguezuelos, J.M., Márquez, R. & Lizana, M. (Eds.) Atlas y libro rojo de los anfibios y reptiles de España. Dirección General de la Conservación de la Naturaleza-Asociación Herpetológica Española, Madrid, pp. 48–50. Montori, A., Tierno De Figueroa, J.M. & Santos, X. (2006) The diet of the Brown Trout Salmo trutta (L.) during the reproductive period: size-related and sexual effects. International Review of Hydrobiology, 91(5), 438–450. Montori, A., Llorente, G.A. & Richter-Boix, A. (2008) Habitat features affecting the small-scale distribution and longitudinal migration patterns of Calotriton asper in a Pre-Pyrenean population. Amphibia-Reptolia, 29. 371-381. Moritz, C. (2002) Strategies to protect biological diversity and the processes that sustain it. Systematic Biology, 51, 238– 254. Nei, M. (1975) Molecular Population Genetics and Evolution. North-Holland Publishers, Amsterdam. Nei, M. (1978) Estimation of average heterozygosity and genetic distance from small number of individuals. Genetics, 89, 583–590. Ragghianti, M. & Wake, D.B. (1986) Genetic variation and its evolutionary implications in the Italian newt, Triturus italicus. Herpetologica, 42, 206–214. Rohlf, R.-J. (1993) NTSYS-pc. Numerical Taxonomy and Multivariate Analysis System, Version 1.8. Exeter Software, Setauket, NY. Ryder, O.A. (1986) Species conservation and systematics: the dilemma of subespecies. Trends in Ecology and Evolution, 1, 9–10. Salvador, A. (1974) Guía de los anfibios y reptiles españoles. Instituto Nacional para la Conservación de la Naturaleza (ICONA), Madrid. pp 1–282. Sbordoni, V., Cobolli, M., De Mattaeis, E., Allegrucci, G., Cesaroni, D., Caccone, A. & Carpareto, G.M. (1982) Orologi molecolari e paleogeografia: congruenza ta stime geocronologiche e datazioni elettroforetiche della divergenza nelle specie del genere Euproctus (Caudata, Salamandridae). Bolletino di Zoologia, 49, suppl., p. 170. Serra-Cobo, J., Uiblein, F.T. & Martínez-Rica, J.P. (2000) Variation in sexual dimorphism between two populations of the Pyrenean salamander Euproctus asper from ecologically different mountain sites. Belgian Journal of Zoology, 130(1), 39–45. Slatkin, M. (1985) Rare alleles as indicators of gene flow. Evolution, 39, 53–65. Swofford, D.L. & Selander, R.B. (1989) Biosys-1: A Computer Program for the Analysis of Allelic Variation in Population Genetics and Biochemical Systematics, Release 1.7, Illinois Natural History Survey, Chicago. Thorn, R. (1968) Les salamandres d'Europe, d'Asie et d'Afrique du Nord. Paul Lechevalier. Paris. pp. 1–373. Weir, B.S. (1990) Intraespecific differentiation. In: Hillis, D.M. & Moritz, C. (Eds.). Molecular Systematics. Sinauer Associates Inc., Sunderland, MA Wolterstorff, W. (1925) Über mehrere Lokalformen des Pyreneenmolches Euproctus asper Dugès. Abhandlungen und Berichte für Naturkunde und Vorgeschichte, 4, 61–76. Wright, S. (1943) Isolation by distance. Genetics, 28, 114–138. Wright, S. (1978) Evolution and the Genetics of Populations. Vol. 4. Variability In and Among Natural Populations. University of Chicago Press, Chicago.


MONTORI ET AL.