Abstract
Morphological and genetic variation of organisms is generally lower in northern than in southern regions of the Palearctic. This ubiquitous geographical pattern has been associated with extinction and recolonization processes in the north versus persistence in the Mediterranean area, as governed by climate change in the Pleistocene. One area of differentiation and complexity is the Iberian Peninsula where two distinct sister-species of marbled newts possess adjoining and geographically sharply delimited ranges over the north (Triturus marmoratus) and the south of the peninsula (T. pygmaeus). Within the latter species two forms are here documented, with different colouration patterns and non-overlapping ranges. The southern form is striped, much as in T. marmoratus, and occurs to the south of the Guadalquivir River. The northern form, here described as a new subspecies, has a reticulated dorsal colouration. It occurs only north of the Guadalquivir and is thus sandwiched in between two striped marbled newt taxa. Both T. pygmaeus subspecies engage inside the Doñana National Park in a ca. 2,500 m narrow contact zone. Although an inferred substantial selection against hybrids goes a long way to support species status, a reanalysis of published genetic data is required to solve this issue, along with the wider investigation of the same type of (microsatelite) data for the southern, nominotypical subspecies.
Introduction
Biological systematics is the study of the variation and evolution of populations of organisms in relation to their taxonomic classification, with as a supplementary aim the reconstruction of the tree of life. A robust phylogeny for the group at hand allows the study of how characters evolve, along with the identification of the selective forces that eventually shaped the organisms over time and space. The insights obtained from such analyses bring us full circle to the enhanced study of biological systematics. Paramount to further progress in this discipline, and biodiversity research in general, is the identification and delimitation of species, here defined as separately evolving sets of populations that retain distinct evolutionary trajectories (Wiley, 1978; Burbrink & Ruane, 2021). The first step for identifying candidate species is to examine the structure of morphological and genetic discontinuities over the landscape and, obviously, the detail at which this exercise is to be carried out depends on how the variation is spatially arranged.
Within the Holarctic, northern regions regularly show low morphological and genetic variability at the population level, whereas southern population groups are often deeply differentiated, even over small geographical distances. This pattern of ‘northern purity and southern richness’ (Hewitt, 2000) is adequately explained by the history of climate change, in particular the repeated range contractions and northern extinctions resulting from the Pleistocene glaciations. Formerly glaciated or otherwise inhospitable regions in the north were most recently colonized in the Holocene, often via long-range dispersal out of one or few refugia. Fostered by founder effects, this resulted in decreasing genetic variability with increasing distance from the respective refugia. Conversely, southern populations survived the cold periods locally and preserved their ancient genetic diversity (Hewitt, 1996). Palearctic regions of high variation and complexity are the southern peninsulas, from west to east the Iberian, the Apennine and the Balkan peninsula (Schmitt, 2007). Here, the scope for discovery of as yet undescribed taxa is high, but calls for a dense sampling, in the case of ground-dwelling organisms with limited dispersal capability such as amphibians perhaps even more than in other groups.
One amphibian group in which the integrative analysis of comprehensively sampled morphological and genetic data resulted in species discovery and a well-supported phylogeny is the Old-World newt genus Triturus, with up to date one northern, two Iberian, two central European-Apennine and four Balkan-Anatolian species (Arntzen et al., 2018; Wielstra et al., 2019). The two species from the Iberian Peninsula, in casu the northern marbled newt T. marmoratus (Latreille, 1800) and the pygmy marbled newt, T. pygmaeus (Wolterstorff, 1905) are morphological and genetically well-differentiated sister-species that engage in a sharp, parapatric contact zone that runs across the western part of the Iberian Peninsula (Arntzen et al., 2014; Arntzen, 2018). Morphological variation is more pronounced within T. pygmaeus than within T. marmoratus. It is here shown that a northern form of T. pygmaeus with a reticulated dorsal colouration and a southern striped form are differentially distributed north and south of the Guadalquivir River, with ranges that only meet in the zone adjacent to the Atlantic Ocean (fig. 1). Published genetic data suggest that the colour forms are genetically isolated at close range within the Doñana National Park, but hypothesis testing is hampered by the unavailability of the constituent data. Even though an inferred lack of gene flow goes a long way in supporting species status of the newly recognized form, it is here conservatively described as a subspecies, in order to safeguard that it is properly labelled. Because the type locality of T. pygmaeus is the province of Cadiz (Wolterstorff, 1905), it is the northern taxon that requires recognition, for which the formal description is in Appendix A1. In the remainder of the paper Triturus p. pygmaeus and T. p. lusitanicus ssp. nov. may also be referred to as the southern and northern taxon of pygmy marbled newt, respectively.
Distribution of marbled newts in the Iberian Peninsula with Triturus marmoratus in blue and T. pygmaeus in red (after Wielstra et al., 2014). The arrow indicates the position of a T. pygmaeus subspecies border inside Doñana National Park (present paper).
Citation: Contributions to Zoology 93, 2 (2024) ; 10.1163/18759866-bja10055
Materials and methods
Adult newts were captured by dip-netting or with funnel traps in France, Portugal and Spain under licenses as appropriate (see acknowledgements). All material was investigated for the number of ‘links’ that they display (Nlinks). Links are dorso-lateral colour pattern character states as described and illustrated by Arntzen (2018). Briefly, links are counted as the number of uninterrupted connections over the green coloured surface, from the position of the crest in males or the mid-dorsal line in females (occasionally juveniles) to the belly, over the region in between the insertions of the fore- and hindlegs, with numbers summed for the left and right side of the body. Individuals with Nlinks ≤ 5 are likely to be T. marmoratus and those with Nlinks > 5 are likely to be T. pygmaeus. The model underlying this criterion is supported by an ‘area under the curve’ value of 0.925 (Arntzen, 2018). All marbled newt populations were assigned to one of five classes, depending on the percentage of inferred T. marmoratus (0‒20%, 20‒40%, 40‒60%, etc). With one exception, no special attention was given to museum collections, because in ethanol preserved material the green colour has frequently faded, with Nlinks difficult or impossible to discern. In the often brightly coloured juveniles Nlinks can usually be determined, but few juveniles are stored and it is currently not known if their colouration pattern is stable over lifetime.
Measurements were taken on live adults to investigate a possible morphological differentiation of the southern and northern forms. Body size was taken as the length from the tip of the snout up to and including the insertion of the hind limb (SVl1), or the posterior side of the cloaca (SVl2). Other morphometric characters are: ILd ‒ interlimb distance, FLl – forelimb length, TFl – third finger length, HLl – hindlimb length, FTl – fourth toe length, Hl ‒ head length and Hw – head width. Measurements were taken in mm with a plastic ruler or plasticized graph paper for SVl and ILd, and with Vernier callipers at 0.1 mm precision for the other characters. Extremities were measured at the right side of the body. Because adult size varies markedly over the study area, possibly as a direct response to the conditions of the environment (i.e., phenotypic plasticity) (García-París et al., 1993; Díaz-Paniagua et al., 1996), the standardized residuals of the regression of ln(character) to ln(SVl1) were used, with the aim to reduce the effect of overall size in the analysis and to increase the normality of the data. Because of a marked sexual dimorphism (see Results) the sexes were treated separately. Data were analysed by a principal component analysis (pca) with spss 26 (ibm spss, 2019), in which populations of the southern taxon (270, 271, 465 and 1003) were compared a posteriori to populations that represent the northern taxon (29, 65, 88, 93, 94, 436, 438 and 463) (fig. 2, for precise locality information, see table 1). The analysis was restricted to this set of southern Iberian populations to reduce the possible effects of gradual geographical variation, taxonomic uncertainty in the west of Portugal and admixture with T. marmoratus (Espregueiro Themudo et al., 2007; Arntzen & Espregueira Themudo, 2008; Arntzen, 2018; Arntzen et al., 2021; Lopez-Delgado et al., 2021). To facilitate interpretation and for comparability with data and analyses to come, also morphometric indices were calculated as (character)/SVl, i.e., from untransformed data. Values for these indices were compared between southern and northern groups with Mann-Whitney’s non-parametric test, also in spss.
The distribution of two species of marbled newts over the western part of the Iberian Peninsula, as inferred from the morphological character Nlinks. Triturus marmoratus is shown in two shades of blue, T. pygmaeus in two shades of red and four localities with both species are shown in grey (see colour legend). The spatial extrapolation does not exceed ca. 50 km. Note that southern Iberian populations are all T. pygmaeus, irrespective of high (cluster C1), low (C2), or bimodal Nlink counts (Doñana National Park). Populations that were studied morphometrically are highlighted by a box. The type locality for T. pygmaeus lusitanicus ssp. nov. is Granja, Portugal and is marked by an asterisk.
Citation: Contributions to Zoology 93, 2 (2024) ; 10.1163/18759866-bja10055
Microsatellite genetic data have recently been used to investigate population dynamic processes in T. pygmaeus in a southern section of the species range (Albert & García-Navas, 2022). Because detailed genetic data used in this paper (11 microsatellite loci) could not be obtained, the approximate reconstruction of the best fitting geographical cline of Structure Q-scores over Doñana National Park (dnp) was performed from printed results (Albert & García-Navas, 2022: Figure 4), with Hybrid Zone Analysis using R (hzar) software (Derryberry et al., 2014), following a protocol used earlier (Arntzen et al., 2017). The reference point for distance in this analysis is the northern bank of the Guadalquivir River with the coordinates 36.7956 N and 6.3700 W.
Results
Link counts were obtained for 2,137 adult and 15 juvenile newts from altogether 128 populations (table 1). The spatial distribution of morphological groups is tripartite (fig. 2). Inferred T. marmoratus are found in France and in the northwest of the Iberian Peninsula whereas inferred T. pygmaeus are found in the southwest of the Iberian Peninsula, which is in line with documented species ranges (fig. 1). Marbled newts from the south of Spain show (low) Nlinks values as typical for T. marmoratus, yet they unquestionably represent T. pygmaeus because the geographical distribution of this coherent group includes the type locality ‘Cadiz province’. In line with low (T. marmoratus and T. p. pygmaeus) and high Nlink values (T. p. lusitanicus ssp. nov.) individuals have either a dark, horizontally striped or a green, reticulated appearance (fig. 3).
Nine marbled newts from three (sub)species each in dorsal and ventral view. Link counts for the left and right side of the body are given in parentheses. Top row – Triturus marmoratus from Jublains, France; left male (3, 2) and right female (1, 1). Middle row – T. pygmaeus lusitanicus ssp. nov. from Cardeña, Spain; from left to right female (2, 4), male (4, 3) and male (4, 5). Bottom row – T. p. pygmaeus from Salinas, Spain; from left to right male (2, 1), male (1, 2) male (2,0) and female (3, 1). Note that the colour pattern in T. marmoratus and T. p. pygmaeus is horizontally banded whereas in T. p. lusitanicus ssp. nov. it is reticulated.
Citation: Contributions to Zoology 93, 2 (2024) ; 10.1163/18759866-bja10055
photography l. a. van der laanSomewhat atypical link counts were found in northern Spain and France for T. marmoratus and in central Portugal for both species. Mixed species allocations were found in four populations at or close to the T. marmoratus ‒ T. pygmaeus contact zone in central Iberia (fig. 2). The differentiation of southern and northern T. pygmaeus has a sharp transition point that is located at the Atlantic coast, adjacent to the mouth of the Guadalquivir River. Link counts are significantly different among groups (Independent samples Mann-Whitney U-test, U = 128, P < 0.0001; fig. 4A). This analysis excludes the juveniles from the preserved Doñana material, that have lower Nlink counts than the adults from this metapopulation (Independent samples Mann-Whitney U-test, U = 41, P < 0.05). It is relevant to note that adults are locally exceptionally small, as observed earlier (García-París et al., 1993; Díaz-Paniagua et al., 1996) and this is illustrated by measurements on preserved material (mncn 9568, female, SVl1 = 40.5 mm, SVl2 = 44.2 mm; mncn 9569, male, SVl1 = 37.4 mm, SVl2 = 41.1 mm and mncn 12067, male, SVl1 = 27.5 mm, SVl2 = 30.0 mm; see also supplementary table S1).
Morphological and genetic variation in Triturus pygmaeus from in and around Doñana National Park (dnp). A – histograms of Nlink counts for four population groups with, from top to bottom: southwestern cluster of T. pygmaeus (localities 65, 436, 438 and 1004 in the south of Portugal, together marked C1 in fig. 2), populations in the northern section of dnp (er, locality 1005 and bs, locality 1006), population in the southern section of dnp (locality 271) and the Betic cluster of T. pygmaeus (localities 270, 463, 465 and 1003 in the very south of Spain, together marked C2 in fig. 1). Exceptionally, the northern Doñana museum material contains 15 juvenile specimens (open bars) along with three adults (shaded bars). B – geographical cline in microsatellite genetic variation. The horizontal axis shows ‘Distance from the Guadalquivir River’ in km and the vertical axis presents Structure Q-scores. The solid round symbols represent populations and the grey area shows the 95% credibility interval. The Structure Q-scores were extracted from a published figure (Albert & García-Navas, 2022: Figure 4) and are unavoidably imprecise. The position of the ‘Torre Carbonero’, a cultural landmark at the Doñana beach, is shown by an arrow (tc). Three populations with morphological data available, that can be associated to the transect are El Rocío (arrow marked er) and the biological station (bs) in northern Doñana and Zacallón C. de los Junqueros in southern Doñana (arrow marked 271). For details, see table 1 and supplementary table S1.
Citation: Contributions to Zoology 93, 2 (2024) ; 10.1163/18759866-bja10055
A principal component analysis on males had 42.6% of the total variance explained on the first axis and 21.6% on the second pc-axis and revealed limited morphometric differentiation among the southern and northern T. pygmaeus population groups (fig. 5A). Loadings (L) on the first pc-axis were high for extremity lengths (FLl, TFl, HLl and FTl, L > 0.78) and low for the other characters (0.26 < L < 0.32). On the second axis loadings were high for head size (Hl and Hw, L > 0.69) and for interlimb distance (L < ‒0.52) and low for the other characters (‒0.21 < L < 0.22). In females, morphometric differentiation was marginal (fig. 5B), with 71.6% of the total variance explained on the first axis and 9.4% on the second pc-axis. Loadings on the first axis were high for all characters (L > 0.77) whereas on the second axis loadings were differentiated for head size (Hl and Hw, L > 0.42) and digit length (TFl and FTl, L < ‒0.32) and not the other characters (‒0.14 < L < 0.14). In the univariate analyses of relative sizes, significantly shorter digits were observed in southern than in northern T. pygmaeus, but only in males (table 2).
Results of a principal component analysis on size corrected morphometric data for Triturus pygmaeus pygmaeus (solid lines) and T. p. lusitanicus ssp. nov. (interrupted lines). The ellipses show the 95% confidence interval of the mean. The left panel is for males and the right panel is for females. Percentages of total variance explained are indicated along the axes. For details, see text.
Citation: Contributions to Zoology 93, 2 (2024) ; 10.1163/18759866-bja10055
The analysis of Structure Q-scores over a densely sampled transect across dnp yields a sharp geographical cline (fig. 4B). The most relevant parameter estimates in the cline model are a width of 2.52 km (95% confidence interval, ci 0.015‒4.425) and a position of the cline centre at 16.93 km from the reference point (ci 16.03‒17.67 km). The nearest sampled localities at either side of the genetic transition are 3,870 m apart. For details, see supplementary table S2. The transition in microsatellite allelic profile takes place within the area over which the T. pygmaeus morphology changes also.
Discussion
The spatial interpretation of marbled newt species allocation by Nlinks corresponds to documented distributions with, however, several areas where morphological data are unconvincing or conflicting (figs. 1 and 2). First, T. marmoratus populations in the north of Spain and France show relatively high counts, but more data are needed to support a possible trend. Second, central Portugal displays a variety of values for populations that, from genetic information, are known to mostly be T. pygmaeus and few T. marmoratus or admixed (Espregueira Themudo & Arntzen, 2007; Arntzen & Espregueira Themudo, 2008; Arntzen et al., 2021; Lopez-Delgado et al., 2021). Third, and most relevant to the present study, is that the Betic region is occupied by T. pygmaeus with a horizontally banded pattern and low link counts, as alluded to earlier (García-París et al., 1993). From colour pattern alone all individuals in these populations would erroneously classify as T. marmoratus. A similar discrepancy may arise from ontogenetic variation. Triturus pygmaeus from dnp has been reported to start reproduction at exceptional small size (males svl1 ≥ 32.2 mm, females SVl1 ≥ 31.2 mm) and young age (1‒2 years) (Díaz-Paniagua et al., 1996; see also supplementary table S1). Because the museum material represents the same northern Doñana stock, it may be concluded that Nlinks increases over lifetime. This raises the question of how many links adults from northern Doñana would display if they were to reproduce at regular size (ca. 60 mm; García-París et al., 1993; Arntzen, 2018) and age (3‒4 years; Castanet & Caetano, 1993).
The southern taxon of T. pygmaeus is found in the Betic region whereas the northern taxon is found from north of the Guadalquivir River up to the range of T. marmoratus in central Iberia. Atlas data suggest that the T. pygmaeus subspecies are currently widely separated on account of the Guadalquivir and the associated low-quality habitats brought about by agricultural land (Pleguezuelos et al., 2002). This distribution is regionally similar to that of fire salamanders (S. (s.) longiristris Joger and Steinfartz, 1994 versus Salamandra salamandra (Linnaeus, 1758)), midwife toads (Alytes dickhilleni Arntzen and García-Paris, 1995 versus A. cisternasii Boscá, 1879 and A. obstetricans (Laurenti, 1768)) and parsley frogs (Pelodytes ibericus Sánchez-Herráiz, Barbadillo, Machordom and Sanchiz, 2000 versus P. punctatus Daudin, 1802 and P. atlanticus Diaz-Rodriguez, Gehara, Marquez, Vences, Goncalves, Sequeira, Martinez-Solano and Tejedo, 2017). Whereas these taxa are genetically more or less deeply differentiated (Martínez‐Solano et al., 2004; Diaz-Rodriguez et al., 2017; Burgon et al., 2021), the data on genetic differentiation of T. p. pygmaeus and T. p. lusitanicus ssp. nov. are ambiguous.
Substantial nuclear genetic variation has been documented within T. pygmaeus with, however, no different signature for Betic populations (Arntzen et al., 2014; Lopez-Delgado et al., 2020). Single populations standing out are from Los Barrios (Cadiz province) that was found to be genetically different from other T. pygmaeus (Arntzen et al., 2021) and from Doñana, possibly locality El Rocío, that is characterized by an aberrant chromosomal C-banding pattern (García-París et al., 1993). In the most detailed genetic study so far (Albert & García-Navas, 2022) different entities were found inside dnp. It is here shown that the data conform to a narrow cline positioned at ca. 17 km north of the Guadalquivir, with a centre that coincides with the position of the ‘Torre Carbonero’, a cultural hallmark at the adjacent beach (fig. 4B). Populations to the north and the south have different morphologies as well as different microsatellite genetic profiles and it is concluded that dnp harbours the southern as well as the northern subspecies. Mitochondrial dna variation within the species appears shallow and without a strong geographical component (Wielstra et al., 2013). No data are currently available to allocate the presumably isolated easternmost T. pygmaeus occurrences to the one or the other taxon (see fig. 1).
The contact zone of northern and southern T. pygmaeus is remarkably narrow. In the absence of selection, the width (w, in km) of a cline can be predicted from a diffusion model as a function of dispersal distance (d, in km per generation) and the length of time since contact (t, in generations), as w = 2.51*d*√t (Barton & Gale, 1993). Dispersal in Triturus newts has been estimated at ca. 1 km per year for T. cristatus (Laurenti, 1768) (Arntzen & Wallis, 1991; see also Smith & Green, 2005; Trochet et al., 2014). However, local populations of T. pygmaeus are charactered by small body size, possibly limiting dispersal capability. At a low dispersal of 100 m per generation, cline width would exceed the observed 2.5 km in ca. 100 generations, i.e., less (or much less) than 300 years. Arguably, the T. p. pygmaeus – T. p. lusitanicus ssp. nov. hybrid zone is much older than this (see discussion below) and its width appears to have been kept in check over thousands of years. In the absence of a strong environmental barrier to dispersal, this may point to a substantial selection against hybrids which, in turn, would assist to support species status for the taxa involved.
Alternatively, a ridge of dunes in between the subspecies has been interpreted as a strong environmental barrier (Albert & García-Navas, 2022), yet these authors report dispersal rates among southern and northern Doñana population groups comparable to, or higher than elsewhere. While I agree that (effective) population size is an important constituent to realized dispersal, I doubt that “The number of sampled individuals … constitutes a representative estimate of the population size (total number of individuals) of each pond” (Albert & García-Navas, 2022: 580). The claim is certainly unfounded for 13 populations for which I collected all material. To complicate matters further, I collected newt larvae in an ad hoc manner with a dip-net, therewith deviating from the standardized protocol with funnel traps reported by Albert and García-Navas (2022) for adults and juveniles. Accordingly, Albert’s inferences on gene flow, as far as depending on population size and life stage, are yet to be substantiated.
The T. pygmaeus subspecies ranges are connected along the Atlantic Ocean and the contact is probably restricted to Doñana National Park. I here suggest that the above-mentioned ridge of dunes marks a former position of the Guadalquivir River and that southern Doñana populations belong to the southern taxon (T. p. pygmaeus) that, with a redirection in the course of the river, became disconnected from its main distribution, to eventually connect with the northern taxon (T. p. lusitanicus ssp. nov.). Archaeological studies acknowledge that the Guadalquivir previously discharged into an inland sea, the Gulf of Tartessos, before reaching the Atlantic Ocean. Bathymetric reconstructions for the 4,000‒2,200 bp period suggest the past presence of a spit of land protruding northward from Sanlucar de Barrameda into the Tartessian gulf. This spit of land is not only a candidate location for the lost city of Tartessos (Abril et al., 2013), but may also have housed the southern taxon of T. pygmaeus.
Similar patterns of genetic variation within the dnp as for T. pygmaeus have been reported for the salamander Pleurodeles waltl Michahelles, 1830 and for the frog Pelophylax perezi (López-Seoane, 1885) (G. Sánchez-Montes, pers. comm., 2023). The occurrence of three microsatellite clines in parallel suggests that the three species involved have been subject to a common spatio-temporal regime, to which the most recent major change in the course of the Guadalquivir may have been a key event, i.e., by disrupting the overland connection to the south and establishing an overland connection to the north. A hierarchy in population genetic differentiation with a profound separation over the dnp transect and not the river would support the proposed vicariant scenario, but data for adjacent populations south of the lower Guadalquivir are yet to be gathered.
With T. pygmaeus as a prime candidate, it might be worthwhile to also collect morphological data over the wider dnp-transect, to investigate to what extent geographical clines from genetic and morphological markers are coincident and concordant. Furthermore, the existence of multiple contact zones within the Doñana National Park calls for studies on the relative contribution of ecological, behavioural and genetic incompatibilities of differentiated and possibly independently evolving, speciating lineages. A further hypothesis to be tested is that other organisms with a low dispersal capability have been subject to the same vicariant scenario. Fortuitously, the entire region is characterized by high levels of diversification among and within species (Hewitt, 2011; Molina-Venegas et al., 2017), so that the baseline genetic variation needed for this test will often be available. Finally, I amplify the call to journal editors to make data sharing prescriptive (Gabelica et al., 2022; Watson, 2022).
Editor: A. Ivanović
Acknowledgements
Licenses to observe and to collect were provided as follows: France – Prefecture de la Mayenne, by permit number 2003-A-2007; Spain – various provinces by permits numbers cn0010/12/aca, cn03/0085, cn04/0269, cn10/0030, dgmen/sen/avp_12_015_aut, dnp 27/2008, E.P-107/04 (mg), is/pa/epcyl/129/2012 and sgyb/foa/afr/cfs, and Portugal – Instituto da Conservação da Natureza, by letters dated 26/10/1998, 19/4/2000 and 19/3/2002 and by permit numbers 397/2007/capt, 102/2010/capt, 103/2010/capt and 107/2012/capt. Sampling in the Doñana National Park was carried out on the basis of the approved project proposal ‘Phylogeography of marbled newts (Triturus marmoratus and T. pygmaeus)’, with permit number 27/2008, allocated to jwa. I thank Marta Calvo Revuelta and Alberto Sánchez Vialas (mncn, Madrid) for access to the collection under their care.
Version of Record
The electronic version of this article is considered a published work according to the International Commission on Zoological Nomenclature (iczn). Therefore, the new name contained in the electronic version is officially published under the iczn code. The work has also been registered in ZooBank (http://zoobank.org/) and can be accessed under http://zoobank.org/urn:lsid:zoobank.org:pub: A6340CA8-57CD-4F2F-81A0-0E0C6422BE90.
Supplementary material
Supplementary material is available online at: https://doi.org/10.6084/m9.figshare.24420589
References
Abril, J. M., Periáñez, R., Escacena, J. L. (2013). Modeling tides and tsunami propagation in the former Gulf of Tartessos, as a tool for Archaeological Science. Journal of Archaeological Science 40: 4499–4508. https://doi.org/10.1016/j.jas.2013.06.030.
Albert, E. M., García-Navas, V. (2022). Population structure and genetic diversity of the threatened pygmy newt Triturus pygmaeus in a network of natural and artificial ponds. Conservation Genetics 230: 575–588. https://doi.org/10.1007/s10592-022-01437-7.
Arntzen, J. W. (2018). Morphological and molecular characters to describe a marbled newt hybrid zone in the Iberian peninsula. Contributions to Zoology 87: 167–185. https://doi.org/10.1163/18759866-08703003.
Arntzen, J. W., Espregueira Themudo, G. (2008). Environmental parameters that determine species geographical range limits as a matter of time and space. Journal of Biogeography 35: 1177–1186. https://doi.org/10.1111/j.1365-2699.2007.01875.x.
Arntzen, J. W., Wallis, G. P. (1991). Restricted gene flow in a moving hybrid zone of the newts Triturus cristatus and T. marmoratus in western France. Evolution 45: 805–826. https://doi.org/10.2307/2409691.
Arntzen, J. W., Wielstra, B., Wallis, G. P. (2014). The modality of nine Triturus newt hybrid zones assessed with nuclear, mitochondrial and morphological data. Biological Journal of the Linnean Society 113: 604–622. https://doi.org/10.1111/bij.12358.
Arntzen, J. W., de Vries, W., Canestrelli, D., Martínez‐Solano, I. (2017). Hybrid zone formation and contrasting outcomes of secondary contact over transects in common toads. Molecular Ecology 26: 5663–5675. https://doi.org/10.1111/mec.14273.
Arntzen, J. W., López‐Delgado, J., van Riemsdijk, I., Wielstra, B. (2021). A genomic footprint of a moving hybrid zone in marbled newts. Journal of Zoological Systematics and Evolutionary Research 59: 459–465. https://doi.org/10.1111/jzs.1243.
Barton, N. H., Gale, K. S. (1993). Genetic analysis of hybrid zones. Pp. 13–45 in: Harrison, R. G. (ed.) Hybrid Zones and the Evolutionary Process. Oxford University Press, Oxford, UK.
Burbrink, F. T., Ruane, S. (2021). Contemporary philosophy and methods for studying speciation and delimiting species. Ichthyology and Herpetology 109: 874–894. https://doi.org/10.1643/h2020073.
Burgon, J. D., Vences, M., Steinfartz, S., Bogaerts, S., Bonato, L., Donaire-Barroso, D., Martínez-Solano, I., Velo-Antón, G., Vieites, D. R., Mable, B. K., Elmer, K. R. (2021). Phylogenomic inference of species and subspecies diversity in the Palearctic salamander genus Salamandra. Molecular Phylogenetics and Evolution 157: 107063. https://doi.org/10.1016/j.ympev.2020.107063.
Castanet, J., Caetano, M. H. (1993). Variability and microevolutionary patterns in Triturus marmoratus from Portugal: age, size, longevity and individual growth. Amphibia-Reptilia 14: 117–129. https://doi.org/10.1163/156853893X00291.
Derryberry, E. P., Derryberry, G. E., Maley, J. M., Brumfield, R. T. (2014). hzar: hybrid zone analysis using an R software package. Molecular Ecology Resources 14: 652–663. https://doi.org/10.1111/1755-0998.12209.
Díaz-Paniagua, C., Mateo, J. A., Andreu, A. C. (1996). Age and size structure of populations of small marbled newts (Triturus marmoratus pygmaeus) from Doñana National Park (sw Spain). A case of dwarfism among dwarfs. Journal of Zoology 239: 83–92. https://doi.org/10.1111/j.1469-7998.1996.tb05438.x.
Diaz-Rodriguez, J., Gehara, M., Marquez, R., Vences, M., Goncalves, H., Sequeira, F., Martinez-Solano, I., Tejedo, M. (2017). Integration of molecular, bioacoustical and morphological data reveals two new cryptic species of Pelodytes (Anura, Pelodytidae) from the Iberian Peninsula. Zootaxa 4243 (1): 1–41. https://doi.org/10.11646/zootaxa.4243.1.1.
Espregueira Themudo, G., Arntzen, J. W. (2007). Newts under siege: range expansion of Triturus pygmaeus isolates populations of its sister species. Diversity and Distributions 13: 580–586. https://doi.org/10.1111/j.1472-4642.2007.00373.x.
Gabelica, M., Bojčić, R., Puljak, L. (2022). Many researchers were not compliant with their published data sharing statement: a mixed-methods study. Journal of Clinical Epidemiology 150: 533–41. https://doi.org/10.1016/j.jclinepi.2022.05.019.
García-París, M., Herrero, P., Martín, C., Dorda, J., Esteban, M., Arano, B. (1993). Morphological characterization, cytogenetic analysis, and geographical distribution of the Pygmy marbled newt Triturus marmoratus pygmaeus (Wolterstorff, 1905) (Caudata: Salamandridae). Contributions to Zoology 63: 3–14. https://doi.org/10.1163/26660644-06301001.
Hewitt, G. M. (1996). Some genetic consequences of ice ages, and their role in divergence and speciation. Biological Journal of the Linnean Society 58: 247–276. https://doi.org/10.1006/bijl.1996.0035.
Hewitt, G. M. (2000). The genetic legacy of the quaternary ice ages. Nature 405: 907–913. https://doi.org/10.1038/35016000.
Hewitt, G. M. (2011). Mediterranean peninsulas: the evolution of hotspots. Pp. 123–147 in Zachos, F. E., Habel, J. C. (Eds.). Biodiversity Hotspots: Distribution and Protection of Conservation Priority Areas. Springer, Heidelberg, Germany. https://doi.org/10.1007/978-3-642-20992-5_7.
ibm spss (2019). ibm spssStatistics for Windows. ibm Corporation, Armonk, New York, USA.
López‐Delgado, J., van Riemsdijk, I., Arntzen, J. W. (2021). Tracing species replacement in Iberian marbled newts. Ecology and Evolution 11: 402–414. https://doi.org/10.1002/ece3.7060.
Martínez‐Solano, I., Gonçalves, H. A., Arntzen, J. W., García‐París, M. (2004). Phylogenetic relationships and biogeography of midwife toads (Discoglossidae: Alytes). Journal of Biogeography 31: 603–618. https://doi.org/10.1046/j.1365-2699.2003.01033.x.
Molina-Venegas, R., Aparicio, A., Lavergne, S., Arroyo, J. (2017). Climatic and topographical correlates of plant palaeo-and neoendemism in a Mediterranean biodiversity hotspot. Annals of Botany, 119: 229–238. https://doi.org/10.1093/aob/mcw093.
Pleguezuelos, J. M., Márquez, R., Lizana, M. (Eds.) (2002). Atlas y Libro Rojo de los Anfibios y Reptiles de España. Spain: Dirección General de Conservación de la Naturaleza, Madrid, Spain.
Schmitt, T. (2007). Molecular biogeography of Europe: Pleistocene cycles and postglacial trends. Frontiers in Zoology 4: 1–13. https://doi.org/10.1186/1742-9994-4-11.
Smith, M. A., Green, D. M. (2005). Dispersal and the metapopulation paradigm in amphibian ecology and conservation: are all amphibian populations metapopulations? Ecography 28: 110–128. https://doi.org/10.1111/j.0906-7590.2005.04042.x.
Trochet, A., Moulherat, S., Calvez, O., Stevens, V., Clobert, J., Schmeller, D. (2014). A database of life‐history traits of European amphibians. Biodiversity Data Journal 2, e4123. https://doi.org/10.3897/BDJ.2.e4123.
Watson, C. (2022). Many researchers say they’ll share data-but don’t. Nature 606 (7916): 853. https://doi.org/10.1038/d41586-022-01692-1.
Wielstra, B., Crnobrnja-Isailović, J., Litvinchuk, S. N., Reijnen, B. T., Skidmore, A. K., Sotiropoulos, K., Toxopeus, A. G., Tzankov, N., Vukov, T., Arntzen, J. W. (2013). Tracing glacial refugia of Triturus newts based on mitochondrial dna phylogeography and species distribution modeling. Frontiers in Zoology 10: 1–14. https://doi.org/10.1186/1742-9994-10-13.
Wielstra, B., Sillero, N., Vörös, J., Arntzen, J. W. (2014). The distribution of the crested and marbled newt species (Amphibia: Salamandridae: Triturus) – an addition to the New Atlas of Amphibians and Reptiles of Europe. Amphibia-Reptilia 35: 376–381. https://doi.org/10.1163/15685381-00002960.
Wielstra, B., McCartney-Melstad, E., Arntzen, J. W., Butlin, R. K., Shaffer, H. B. (2019). Phylogenomics of the adaptive radiation of Triturus newts supports gradual ecological niche expansion towards an incrementally aquatic lifestyle. Molecular Phylogenetics and Evolution 133: 120–127. https://doi.org/10.1016/j.ympev.2018.12.032.
Wiley, E. O. (1978). The evolutionary species concept reconsidered. Systematic Zoology 27: 17–26. https://doi.org/10.2307/2412809.
Wolterstorff, W. (1905). Zwergformen der Palearktischen Urodelen. Comptes Rendus 6 Congrès International de Zoologie, Berne: 258–263.
Appendix A1
Triturus pygmaeus
The description of the species’ external morphology below largely follows García-París et al. (1993: 5), albeit that in my terminology marbled newts have a dark dorsal side with a green colouration pattern. The male crest is lowered over the pelvic area and males from the northern part of the species range (with the reticulated colour pattern) were found to have not shorter but longer digits that those in the south.
Large head, longer than wide, gradually sharpened, ending abruptly in a narrow, blunt nose. Small eyes placed in a forward position. Gular fold moderately conspicuous. Elongated body with long, thin legs. Digits long, especially in males. Toe webbing absent. Relatively short tail, its cross-section being almost cylindrical at the base and progressively compressed towards the tip. Finely granulated skin, ridged in the gular region, with large granules unevenly spread over the dorsal surface. A general north to south decrease in overall size, be it that individuals of extremely small size have been reported in and around the Doñana National Park, irrespective of intraspecific taxonomic affiliation.
During the terrestrial phase the dorsal colour pattern consists of an irregular pattern of olive-green spots on a dark background. These spots can create a reticulum covering the side of the body or be sparse or absent in the middle, to the effect of a dark, more or less wide band running along the side of the body (fig. 3). In females, the dorsal midline appears as a continuous orange stripe whereas males have a line with alternating orange-yellow and black narrow bands. The dorsal part of the tail shows the same colour as the dorsum. Ventrally, the colouration is that of an off-white, crème to grey background, sometimes with an orange hue towards the cloaca, on which large blackish spots and small white dots may be found. These white dots may show a higher density on the flanks and occasionally completely cover the background colouration. White dots are also present on the base of the flanks, spreading over the sides of the neck and the cheeks, here alternating with black spots. The ventral black spots are usually large and roundish in males and small and more elliptically shaped in females. The ventral background colour continues over the gular region, often at a lighter shade of pale. The distribution of black and white dots does not seem to follow any obvious geographical trend (fig. 3, Appendix fig. A1). Juveniles show a brighter colour than adults, frequently with a conspicuous, fluorescent-like orange dorsal line and fluorescent-like green dorso-lateral colouration, and a yellowish ventral colouration.
Variation in ventral colouration of Triturus pygmaeus. Top row – T. p. lusitanicus ssp. nov.: male (with dark tail underside) and two females from Villalba and male from Cardeña. Middle row – T. p. pygmaeus: male from Salinas, male and female from Los Barrios and male from Rio Alberite. Bottow row – T. p. pygmaeus: two males and two females from Rio Alberite.
Citation: Contributions to Zoology 93, 2 (2024) ; 10.1163/18759866-bja10055
photography l. a. van der laanDuring the breeding period the males have a moderately high dorsal crest, which continues along the tail with an inflexion in the pelvic area. Male crest blotched by alternating black and white vertical bands. In the female, the mid-dorsal line usually shows orange-yellowish, or greyish tones. Male cloaca exposed, swollen and deep black. Female cloaca walnut-shaped, bordered with papillae. Well-developed labial folds during the aquatic phase, especially in females. While the intensity of the dorso-lateral colouration pattern may change during the aquatic phase, especially in males, the reticulation itself remains. Some individuals may become very dark and melanic in appearance, with the colouration pattern difficult to discern. Post-metamorphic juveniles are rarely if ever found in the water.
Description of Triturus pygmaeus lusitanicus ssp. nov.
Description of the type specimen – adult male in breeding condition. Preserved on ethanol in excellent storage condition at the Museo Nacional de Ciencias Naturales, Madrid, under catalogue number mncn 51799. Measurements are SVl1 51 mm, SVl2 57 mm, ILd 25 mm, Hw 10.3 mm, Hl 14.9 mm, FLl 20.0 mm, TFl 6.0 mm, HLl 20.8 mm and FTl 7.7 mm. Total length 105 mm. Relative finger lengths 1 < 4 < 2 < 3. Relative toe lengths 5 < 1 < 2 < 3 = 4. Nlinks left 5, right 6, total 11; colour pattern reticulated. Thirteen crème-white bands over the head and body part of the mid-dorsal crest. Light grey ventral colouration, with an orange hue towards the swollen, deep black cloaca. Several roundish but vague ventral dark spots and many irregularly shaped medium large white spots. The type specimen is shown in Appendix fig. A2.
Holotype of Triturus pygmaeus lusitanicus ssp. nov. in live condition, at the left (top) and right dorso-lateral view (middle), and in ventral view (bottom). The scale bar equals 10 mm.
Citation: Contributions to Zoology 93, 2 (2024) ; 10.1163/18759866-bja10055
Description of paratypes
First paratype – mncn 51800 Female. Excellent storage condition. SVl1 52 mm, SVl2 55 mm, ILd 27 mm, Hw 10.0 mm, Hl 14.7 mm, FLl 18.4 mm, TFl 5.8 mm, HLl 18.7 mm and FTl 6.8 mm. Total length 105 mm. Relative finger lengths 1 < 4 < 2 < 3. Relative toe lengths 5 < 1 < 2 < 4 < 3. Nlinks left 3, right 6, total 9; colour pattern reticulated. Medium grey ventral colouration with no black spots and a medium number of small white spots.
Second paratype – mncn 51801. Female. Excellent storage condition. SVl1 58 mm, SVl2 61 mm, ILd 29 mm, Hw 10.8 mm, Hl 15.8 mm, FLl 20.0 mm, TFl 7.1 mm, HLl 21.3 mm and FTl 7.6 mm. Total length 120 mm. finger lengths 4 < 1 < 2 < 3. Relative toe lengths 5 < 1 < 2 < 3 = 4. Nlinks left 6, right 6, total 12; colour pattern reticulated. Grey ventral colouration with few small black spots and few small white spots.
Other material from the type locality – Additional morphological data on 11 individuals from the type locality are in the Appendix table A1.
Locality and date of collecting – Granja, Portugal, at 7.262 western longitude and 38.318 northern latitude, elevation 160 m above sea level. Collected by J. W. Arntzen on 11 March 2023.
Diagnostic features – the newly recognized subspecies is morphologically distinguishable from the southern, nominotypical subspecies by a high number of green coloured transversal bands (also referred to as ‘links’), by long fingers and toes in males and possibly by a relatively big head. Microsatellite data for a section of the range show a genetic profile that is drastically different from that of the adjacent southern form, as studied in the southern section of Doñana National Park.
Taxonomic status – The northern form is described as a subspecies. This decision is debatable because the currently available microsatellite genetic data suggest that genetic exchange with the nominotypical form in the short contact zone is absent or rare. The low level of genetic exchange may be due to a local environmental barrier, or to strong selection against hybrids.
Derivatio nominus – the subspecies name ‘lusitanicus’ acknowledges that the range of the newly recognized taxon coincides with the former Roman province of Lusitania.
Suggested vernacular name – Lusitanian pygmy newt.
Distribution – The south-western corner of the Iberian Peninsula with the exception of the Betic Cordillera. Approximate latitudinal range borders are the range of T. marmoratus at ca. 40 degrees north and the Guadalquivir to the south. In the east ranging till ca. 3.0 degrees west, perhaps 1.5 degrees west. Widely allopatric to T. p. pygmaeus, yet presumably with a narrow and sharp contact zone located inside the Doñana National Park. See figs 1 and 2.
