The Proceedings of the Eighth International Conference on Creationism (2018)

characters to calculate baraminic distances. Most characters of the lower jaw and lower jaw dentition were retained, but all but one character of the palatal dentition, and all the skull roof, braincase and postcranial characters were eliminated, including the new characters introduced by Callier et al. (2009). Clack et al.’s (2016) matrix consisted of 45 taxa scored for 213 cranial, mandibular and postcranial characters. Characters were sourced from Ahlberg and Clack (1998), Clack (1998), Clack et al. (2012), Clack and Finney (2005), Klembara et al. (2014), Ruta and Clack (2006) and Ruta et al. (2002), several of them with modifications and with the addition of five new characters. The taxa included two elpistostegids, six Devonian tetrapods and a sampling of other Devonian to Carboniferous tetrapods and fishes. We performed our calculations on two subsets of taxa. In both cases, several Carboniferous and Permian taxa representing multiple families were removed in order to eliminate excessive outgroups. Our first analysis excluded 32 taxa and used a subset of 13 taxa, including two of the new Tournaisian tetrapods ( Perittodus , Diploradus ) described by Clack et al. (2016). A third ( Ossirarus ) had to be eliminated because it had too few characters in common with the other taxa. After filtering at 0.75 character relevance cutoff, we used 32 characters to calculate baraminic distances. All postcranial and upper dentition characters were eliminated after filtering, as were most of the palatal and general skull characters. Our second analysis excluded 36 taxa and used a subset of 9 taxa, and included only the Devonian tetrapods, the elpistostegids and Eusthenopteron . After filtering at 0.75 character relevance cutoff, we used 50 characters to calculate baraminic distances. All postcranial characters were eliminated after filtering, along with most of the palatal and general skull characters. RESULTS The baraminic distance correlation results for Ahlberg and Clack’s (1998) whole matrix are summarized in Figure 1. Two blocks of positive correlation are evident, one comprising the ‘more derived’ Carboniferous and Permian taxa and a second smaller one comprising the Devonian taxa plus one Lower Carboniferous taxon ( Whatcheeria ). Almost all members of the first group are negatively correlated with all members of the second group, with only three exceptions: Greererpeton , Megalocephalus and Crassigyrinus . Megalocephalus is negatively correlated with all but one member of the Devonian group. However, Greererpeton is negatively correlated with only four of the nine members of the Devonian group, and Crassigyriunus is negatively correlated with only one of the Devonian taxa. Moreover, Crassigyrinus is the only taxon that does not show positive correlations with all other members of its own group. Bootstrap values are generally good, ranging from 27% to 100% with a median value of 94%. The 3D MDS results indicate the same groupings (Figure 2), with an obvious cluster of Carboniferous and Permian taxa separated from a cluster of Devonian taxa + Whatcheeria . The Devonian taxa include seven tetrapods plus one elpistostegid ( Panderichthys ). The Carboniferous and Permian taxa represent multiple families of tetrapods. The 3D stress was 0.133 with minimal stress of 0.121 at four dimensions. We suspected that most of the Carboniferous and Permian taxa were so different from the Devonian taxa that they were obscuring any discontinuities among the Devonian taxa. We therefore removed most of the Carboniferous and Permian forms from the analysis and ran it again, with a subset of 11 taxa comprising the Devonian forms plus Whatcheeria , Greererpeton and Crassigyrinus . The baraminic distance correlation results for this subset of Ahlberg and Clack’s (1998) matrix are summarized in Figure 3. There are two blocks of taxa, one comprising the Devonian tetrapods + Whatcheeria and another comprising Greererpeton + Crassigyrinus . Only one Devonian tetrapod ( Obruchevichthys ) is correlated negatively Garner and Asher ◀ Devonian and Carboniferous tetrapodomorphs ▶ 2018 ICC 463 Figure 1. BDC results for Ahlberg and Clack’s (1998) whole matrix, as calculated by BDISTMDS (relevance cutoff 0.75). Closed squares indicate significant, positive BDC; open circles indicate significant, negative BDC. Black symbols indicate bootstrap values >90% in a sample of 100 pseudoreplicates. Grey symbols represent bootstrap values <90%. Figure 2. Three dimensional MDS applied to Ahlberg and Clack’s (1998) whole matrix. Devonian tetrapods are shown in blue, elpistostegids in red and Carboniferous and Permian tetrapods in yellow.