9327 - Laserfiche WebLink

Home Browse Search

and Santa Cruz systems. These groupings generally remained intact in subsequent analyses which deleted some other populations (Figs. 16,17) so long as the component samples were themselves retained in the analysis. Thus, overall, while these lower basin samples collectively are genetically relatively similar (all grouping at the 0.95 level or above), they group mosaically with only few subpatterns reflecting geography or morphological groupings. It is notable that some samples from the same streams separated by a few kilometers failed to cluster. The two samples from the upper Verde River (at Perkinsville [VDP] and Tapco [VDS+VBT]), though separated by perhaps fewer than 20 stream kilometers (Fig. 1), clustered in widely divergent groups (Fig. 15), while the samples from upper and lower Eagle Creek in the upper Gila Basin were divergently placed but at a somewhat higher level of similarity. Possible reasons for this and other groupings will be further discussed below under Allelic Variation and Taxonomic Implications. DeMarais' (1992) phenetic analyses incorporated samples from fewer and, in some cases, different Gila populations of the Gila and Bill Williams basins compared to those reported herein. Referring to populations examined by both studies., dendrograms presented in his report are at variance with Figure 15 and others herein at several junctures, though some (e.g., Cherry - Tonto crs.; Bonito Cr. - Redfield Canyon) do cluster relatively similarly. Comparison of allelic variation at loci examined in both studies reveals that findings were relatively congruent with minor exceptions as noted below. DeMarais examined five loci not examined herein and at least one (AH-A) may have provided sufficient variation to effect allignment of samples; conversely, many more loci were screened herein than by DeMarais and, of the 24 that ultimately entered the BIOSYS analysis, at least 15 differed from his, including several variable ones. The lack of agreement between the respective phenetic analyses therefore may be a function of both the partial differences in the suites of variation analyzed and the overall mix of populations represented. While offering different allignments, DeMarais' analyses likewise show an overall lack of pattern with respect to geography or supposed morphological groupings with some exceptions (e.g., clustering of morphologically similar samples from Spring, Redfield, and Bonita crs.; geographically proximate Blue & San Carlos rivers, etc.). Allelic Variation while some loci, which were polymorphic in samples from other regions studied, were monomorphic or nearly so among samples from the Gila Basin (e.g., AK-1, bGAL, IDH-2, Figs. 2,3,8), others showed considerable variation in allele frequencies. EST-2, GPI-1, MPI-2, and PGM are particularly 11