Molecular phylogeny of Austrofundulus Myers (Cyprinodontiformes: Rivulidae), with revision of the genus and the description of four new species. Author Tomas Hrbek Author Donald C. Taphorn Author Jamie E. Thomerson text Zootaxa 2005 825 1 39 http://www.zoobank.org/urn:lsid:zoobank.org:pub:A65C9E57-187D-4503-AD3A-E7E4135A77BF journal article z00825p001 A65C9E57-187D-4503-AD3A-E7E4135A77BF Austrofundulus Phylogeny We conducted maximum likelihood and maximum parsimony analyses of the data. The results of the two analyses are nearly identical. In maximum likelihood analysis, the Colombian Austrofundulus are sister to all other Austrofundulus (Fig. 2), while in maximum parsimony analysis the Colombian Austrofundulus are sister to all other Austrofundulus and three Rachovia species from the Maracaibo basin ( R. hummelincki , R. pyropunctata and R. brevis ) (Fig. 3). In both analyses the Guajira Austrofundulus are sister to all other Lago Maracaibo basin Austrofundulus . Austrofundulus transilis is the sister taxon to A. limnaeus from the Rupununi, Guyuana, and in turn this clade formed the sister clade to the Tucacas population of A. limnaeus . These two A. limnaeus and A. transilis populations form the sister clade to the Lago Maracaibo populations of A. limnaeus (Fig. 2). Likelihood of the maximum likelihood topology is -ln L = 13502.97562 with α = 0.465919. Tree length of the maximum parsimony topology is 2422 steps, the consistency index is 0.689 and retention index is 0.792. Monophyly of all populations is strongly supported statistically, as are majority of among population relationships (Figs. 2, 3). Further evidence supporting monophyly and distinctiveness of these populations are unique length variations at positions 430 to 475 of the 12S rRNA (Table 2). TABLE 2. Diagnostic pattern of length variation and base-pair composition at positions 430 to 475 of the 12S rRNA. Intraspecific as well interspecific variation is shown.
R. maculipinnis TTAAAC--------ATTAATAAAA -CCCAACATCACTTA--AAGG
R. pyropunctata TTAAACC-------ATTAATAAGCCACATGGATCATCAAC-AGGG
R. hummelincki TTAAATC-------ATTAATAAGTCACATGGATCATCAAC-AGGG
R. brevis TTAAAAT-------ATTAATAAAT--CCTTAAATTAACTA-AAGG
A. transilis TCAAAT-----TACATATATATAAACTTCTATAACTCTCA-AAGG
A. rupununi TCAAA--------TATATATATAAACTTCTGTAACTCTCA-GAGG
A. rupununi TCAAA----TATATATATATATAAACTTCTGTAACTCTCA-GAGG
A. leohoignei TCAAAA----TTATTCATATATAAACTTCTATAACTCTTA-AAGG
A. limnaeus TTAAAT---------TATAAATAAACCCCTATAACTATTA-AAGG
A. limnaeus TTAAAC---------TATAAATAAATCCCTATAACTATTA-AAGG
A. guajira TCAAAC---------TATATATAAATCTCTATAACTCTTA-AAGG
A. leoni TCAAAC---------TATAAATAAACCCCTATAACTATTA-AAGG
A. leoni TCAAAC---------TATAAATAAACCCCTATAACAATTA-AAGG
A. leoni TCAAAC---------TATAAATAAACCCCGATAACAATTA-AAGG
A. leoni TAAAAC- --------TATAAATAAACCCCTATAACTATTA-AAGG
FIGURE 2. Maximum likelihood phylogenetic tree on 2653 aligned mtDNA base pairs. Likelihood of the topology is -ln L = 13502.97562 with α = 0.465919. Numbers above nodes are bootstrap values (Felsenstein, 1985). In general, divergences are well supported. Key to genera: A. = Austrofundulus ; R. = Rachovia . FIGURE 3. A strict consensus of four single most parsimonious mtDNA trees based on 2653 aligned base pairs, 796 of which were parsimony informative. The length of the trees is 2422 steps. Numbers above nodes are bootstrap values (Felsenstein, 1985) followed by Bremer support values (Bremer, 1994). In general, divergences are well supported. Key to genera: A. = Austrofundulus ; R. = Rachovia .