Supplementary material A.DNA sequencing and phylogenetic methods DNA was extracted from fresh frozen tissues (Karsenia koreana, MVZ 246033 and MVZ 246034; Hydromantes brunus MVZ 247158) using a method modified from Brufford et al. (1992)1. A 1,503 base pair portion of the recombination activating gene-1 (Rag-1) was amplified in several overlapping fragments using standard protocols. PCR products were run out on 1.5 % agarose gels. Successful amplifications were either directly cut from the gel and extracted using Qiagen Gel Extraction kit, or cleaned with a Millipore PCR96 cleanup kit (MontágeTM). Big Dye (ABI) was used for cycle sequencing reactions sequenced on an ABI 3730 capillary sequencer. Individual sequences were edited and multiple sequence alignments were made in Sequencher TM vers 3.1 (Gene Codes Corp). The two K. koreana had identical Rag-1 sequences, so we only included one in the phylogenetic analyses. We compared our new sequences of K. koreana and H. brunus with recently published Rag-1 sequences for salamanders2,3(Appendix 1). The sequences that we chose represent all extant families of salamanders and the majority of genera of plethodontids. Modeltest vers 3.64 chose GTR + I + as the most appropriate model of sequence evolution4. Bayesian analyses implemented via the program MRBAYES vers 35, applying a six parameter model plus invgamma, was run with four chains for 5.2 million generations (with the first 200,000 discarded as burn-in). The resulting 50% majority-rule consensus of the 50,000 post burn-in trees was computed in PAUP* to determine the topology and posterior probabilities for each node6. To determine the amount of evolutionary distance among our taxa, the sumt command in MR BAYES was used to calculate the average branch length across all post burn-in trees. Pseudobranchus (Sirenidae) was selected as the outgroup2. PAUP* was used to generate a matrix of Maximum Likelihood distances based on parameters (Base frequencies: A = 0.2797, C = 0.2477, G = 0.2446, T = 0.2280; Substitution model: [A-C] = 1.4290, [A-G] = 3.4255, [A-T] = 1.0241, [C-G] = 0.6478, [C-T] = 5.4303, [G-T] = 1.0000; I = 0.4544; = 1.4226) chosen by Modeltest4. Maximum Likelihood bootstrapping was performed in PAUP* and run for 600 pseudoreplicates. The most likely tree was identical to the Bayesian tree.
References 1. Brufford, M. W., Hanotte, O., Brookfield, J. F. Y. & Burke. T. Single-locus and multilocus DNA fingerprint. In The South American Herpetofauna: Its Origin, Evolution, and Dispersal. Molecular Genetic Analysis in Conservation (ed Hoelzel AR), pp. 225-270. IRL Press, Oxford (1992):
2. Chippindale, P. T., Bonett, R. M., Baldwin, A. S. & Wiens, J. J. Phylogenetic evidence for a major reversal of life history evolution in plethodontid salamanders. Evol. 58, 2809-2822 (2004).
3. Wiens, J. J., Bonett, R. M. & Chippindale, P. T. Ontogeny discombobulates phylogeny: paedomorphosis and higher-level salamander relationships. Syst. Biol. In Press (2005).
4. Posada, D. & Crandall, K.A. MODELTEST: testing the model of DNA substitution. Bioinformatics14, 817-818 (1998).
5. Huelsenbeck, J.P. & Ronquist, F.R. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics17, 754-755 (2001).
6. Swofford, D.L. PAUP*: Phylogenetic analysis using parsimony (and other methods), v. 4.0. b10. (Sinauer Associates, Sunderland, MA, 2001).
Supplementary material B Appendix 1. Specimens used for molecular analyses.