9470 - Laserfiche WebLink

Home Browse Search

The distribution of haplotypes strongly reinforces the pattern detected by the F statistics. The San Juan system is unique and the other subbasins are more similar to their adjacent basins (the Colorado River subbasin to the Dolores River subbasin and the White River subbasin to the Yampa River subbasin), yet they are still unique from one another in the ratios of the dominant haplotypes. As has been noted earlier, the unique minor haplotypes tend to be most abundant in the individual populations. The isolating mechanism generating this pattern appears to be both distance and time. The presence of unique minor haplotypes within populations supports time as an important factor because these unique forms have not spread to other parts of the basin. Unfortunately we did not survey the speckled dace haplotypes in the subbasins to the Colorado and Green Rivers within Utah, but it is likely that haplotype frequencies in other portions of the drainage basin will further reinforce the observed patterns. The Gunnison system would be expected to be much closer to the Colorado River subbasin than is the Dolores subbasin. Unfortunately low sample size and lack of additional sampling sites does not allow us to address this relationship. Sequence similarity was also examined using the Tamura-Nei index (Tamura and Nei 1993). MEGA (Kumar et a12001) was used to generate clusters of the similarity matrix (Figures 3a, 3b). Because this is a similarity based index, populations and subbasins can be joined in a synthetic index, regardless of individual evolutionary trajectories. Thus multiple unique haplotype sequences within a population are combined with haplotypes shared with other populations to generate the synthetic stand for a given location. This allows a visual representation of the overall similarity of populations and subbasins. The relationships between basins are apparent in Figure 3a. The San Juan is the most dissimilar of the subbasins examined in the upper Colorado Basin. The remaining subbasins are more similar to one another, with the Colorado River subbasin being the most distinct, followed by the Gunnison and Dolores. The White and Yampa are the most similar subbasins. The associations among drainages are more clearly shown in the cluster of relationships among sites (Figure 3b). The San Juan populations are most distant from other speckled dace populations, and interestingly, the San Juan populations tend to show more similarity to the overall cluster than to each other. They do not form a separate cluster of their own. The Colorado, Gunnison and Dolores populations are quite similar to one another. Slight changes in frequencies of various haplotypes would easily shift relationships among these populations. Both the White and the Yampa populations form clusters where populations within the subbasins are more similar to one another than to the adjacent subbasin. One exception to this is the Piceance Creek sample from the White River (WhPi; Figure 3b; site #9), which is clustering basally with the Yampa subbasin populations. This population (site # 9 on Figure 2), was pulled to the Yampa subbasin (Figure 3) because of the high proportion of haplotype K (haplotype # 54; Table 9). The population distance cluster does show the White and Yampa river populations as being most similar, but they clearly have separate evolutionary trajectories. Phylogenetic relationships among the speckled dace populations were examined using PAUP (Phylogenetic Analysis Using Parsimony; Swofford 1998). We used longnose dace as our out group, and included speckled dace from the Virgin River and Gandy marsh (Utah) to help anchor the relationships. We identified 96 speckled dace haplotypes within the Colorado River 36