Newts in time and space: the evolutionary history of Triturus newts at different temporal and spatial scales

Espregueria Themudo, G. (2010)

Doctoral Thesis

The interspecific relationships in the genus Triturus are incompletely known. InCHAPTER 2, we attempt to resolve them by using allozyme and mtDNA data. Despitethe large number of markers used, relationships continue to elude us. The resultssuggest that speciation in the group occurred during a short time period (the end of theMiddle Miocene). Paleogeographic reconstructions of the presumed centre of origin (theBalkans) support this hypothesis. We proposed here that T. macedonicus should beraised to full species given its allopatric distribution and high genetic divergence with T.carnifex.The best way to test the scenario presented in CHAPTER 2 is to look at multipleindependent markers that, unfortunately, were not readily available. CHAPTER 3describes the process through which dozens of markers were designed and tested for thegenus Triturus. Out of more than fifty markers tested, five provided promising resultswith enough variability to study the phylogeny and phylogeography of the genus. Thisopened the door not only for CHAPTER 4, but also for 5 and 6.Taking CHAPTER 2 as the starting point and with the tools developed inCHAPTER 3, CHAPTER 4 attempts to decipher the history of the genus Triturus. Thestudy includes samples from 15 individuals of the seven species of the group. Locationswere selected to cover most of the variability in the group, with the exception of areasclose to other species. Hybridization is known to occur in these areas, and could bias theinferences made. Results show that all the genes, except the two mtDNA ones, haveincongruent phylogenetic signals. We used phylogenetic networks to visualize thealternative phylogenetic signals and have built a phylogenetic tree based on a Bayesianhierarchical method that obtains the species tree based on individual gene trees. Thisapproach successfully resolved the branching order of the newts, although time intervals are very narrow, confirming the near simultaneous speciation scenario of CHAPTER 2.We also found a high genetic differentiation between the two forms of the southerncrested newt (T. karelinii) and proposed that they should be raised to full species (T.karelinii and T. arntzeni).As described in CHAPTER 4, hybridization can have a confounding effect onphylogenetic inferences. Incomplete lineage sorting can also produce similar patterns asgene flow, further complicating matters. CHAPTER 5 takes a new approach indistinguishing between gene flow and incomplete lineage sorting, only possible inspecies with very well defined species borders and limited dispersal capability as thenewts, which limits gene flow to a narrow geographical region.As can be seen in the previous chapters, the crested newts show complex(although interesting) patterns resulting from compound interactions, especially in theBalkans. The marbled newts, on the other hand, are only two species, and therefore wecan cover their evolutionary history in more detail, as it is simpler. CHAPTER 6 studiesthe phylogeography of the two species of marbled newts. Being just a pair and not agroup of species, relationships are not problematic. Their sibling relationship is wellestablished (see CHAPTER 2). This chapter also explores how differences in ecologicalconditions (see CHAPTER 8) along a hybrid zone affect its structure and the amount ofgene flow between species.The morphological distinction of adults of the two marbled species is relativelystraightforward, even though variation is present. Eggs and larvae, on the contrary, areeasy to spot and collect, but impossible to distinguish. CHAPTER 7 describes a cheapand fast molecular technique that allowed the identification of the large number ofsamples used in CHAPTERS 8 and 9.Based on a morphological and genetic identification of individuals of the twospecies of marbled newts and published distributional data, CHAPTER 8 identifiedecological factors associated with the range border of the two species of marbled newt.Ecological models defining the distribution of the two species differed in four mainareas: the northern most region of contact close to Aveiro, the rest of the coastal area(see also CHAPTER 9), the region coinciding with the Tejo river, and the remaindercontact zone going until Madrid. In the coastal area of Portugal, close to Caldas da Rainha, the northern marblednewt was found where only pygmy marbled newts were thought to occur. Thisprompted a detailed study on the distribution of the two species in this area described inCHAPTER 9. This study revealed a small pocket of populations of the northern speciessurrounded by populations of its sister southern species. Given the distance of thispocket to the main distribution, we believe that the enclave was created by T. pygmaeusmoving north, superseding T. marmoratus, rather than the latter species expandingsouthwards.

• Triturus
• Phylogenetics
• Phylogeography
• Geographical Information system
• Amphibia
• Nuclear genes
• Newts
• Allozymes
• Mitochondrial DNA

• All authors: Espregueria Themudo, G.
• Supervisor: Gittenberger, E.
• Co-supervisor: Arntzen, J.W.
• Qualification: Doctor (dr.)
• Awarding Institution: Faculty of Science, Leiden University
• Date: 2010-03-10
• Court: LEI Universiteit Leiden
• Sponsorship: This thesis was financially supported by a Fundação para a Ciência e para a Tecnologia (FCT) PhD grant (ref. SFRH/BD/16894/2004).