54, 56, 59-62, 64, 68, 70, 74, 76, 80-81, 88, 92, 95, 99-100, 103-104, 107, 110-111, 123, 125, 132, 140-141, 147-151, 160-161, 165-166, 168, 170, 175, 181-182, 189-190, 192, 195, 206, 209-210, 212, 219, 224, 230-231, 235-238, 240, 243, 245, 247, 250-251, 254-256, 259, 264, 268-269, 281-282, 287, 295, 297-298, 301, 303-304, 306, 314315, 318-319, 321, 326-328, 332, 335-338, 341-345, 350, 352, 358359, and 362-391. As a result, we evaluated four different character matrices, the original craniodental character matrix reduced to the fourteen taxa here evaluated, a subset of high-relevance craniodental characters with character relevance a > 0.5, the postcranial characters, and a combined set of the high-relevance craniodental characters and all postcranial characters. Baraminology analysis. Following Sinclair and Wood (2021), we used both simple matching (baraminic) distances and Jaccard distances for distance correlation and cluster analysis. We performed distance correlation analysis with both distance metrics as well as Pearson and Spearman correlations. We performed cluster analysis using medoid partitioning and fuzzy analysis with both types of distances. For each set of analyses, we used all characters present in the four character sets described above. All clustering and distance correlations calculations were done using BARCLAY (https://coresci. org/barclay). Any additional analyses were performed in R. RESULTS All craniodental characters. Before presenting and interpreting the postcranial characters and their clustering patterns, we sought to develop a comparison group of just craniodental characters for the fourteen taxa used in the postcranial character matrix. We derived this craniodental matrix from Wood’s (2020) matrix by combining the Asian and African Homo erectus taxa into a single Homo erectus s.l., as described in the methods. Because our previous analysis (Sinclair and Wood 2021) used a larger set of 17 taxa that included H. rudolfensis and Au. anamensis, in addition to separating African and Asian H. erectus into separate taxa, we performed all distance correlation and cluster analyses on just the present fourteen taxa to determine how the smaller taxon sample might alter previous results. Figure 1. Distance correlation results for all craniodental characters. Correlations and distance metrics are shown in the diagram. Filled squares indicate significant, positive distance correlation. Open circles indicate significant, negative correlation. WOOD AND BRUMMEL Hominin Baraminology Reconsidered 2023 ICC 253
RkJQdWJsaXNoZXIy MTM4ODY=