Laserfiche WebLink
(0.00036 or 0.036%} and the proportion of variation attributable to population differences <br />were reduced (ca. 8%). <br />Phylogenetic resolution was limited due to the large number of haplotypes differing by a <br />small number of characters, resulting in more than 1000 most parsimonious trees. Clustering <br />of haplotypes by estimates of sequence divergence provided more resolution; however, the <br />most divergent haplotypes differed by only 0.7%. The consensus of the first 1000 most <br />parsimonious trees and the phenetic network were different, although haplotypes were not <br />geographically partitioned in either analysis (for example, see Figure 2). Individuals from <br />GRIT canals, Little Colorodo River, Lake Mead, Lake Powell, Green River and upper <br />Colorado River either shared or exhibited haplotypes most similar to those found in the most <br />diverse population, Lake Mohave. <br />Discussion <br />Razorback suckers were once abundant and widespread throughout the Colorado River basin. <br />Human activities have severely impacted this species, with dams modifying habitat and <br />disrupting potential corridors for movement throughout the basin. Even prior to human <br />modification, however, razorbacks were most likely distributed discontinuously, tending to be <br />localized in wide, non-canyon reaches of the mainstem Colorado and lazger tributaries, like <br />the Green and Gila rivers (Minckley, 1991). Lack of continuity could have resulted in <br />isolation and divergence of populations, although longevity, fecundity, and size of the species <br />made it probable that enough individuals dispersed among regions to prevent divergence of <br />subpopulations. <br />Restriction endonuclease analysis of mtDNA (this report) and allozymes (report to be <br />provided by Buth and Haglund) were used to quantify the amount of genetic variation within <br />10 <br />