Time-series analysis is used widely in ecology to study complex phenomena and may have considerable potential to clarify relationships of genetic and demographic processes in natural and exploited populations. We explored the utility of this approach to evaluate population responses to management in razorback sucker, a long-lived and fecund, but declining freshwater fish species. A core population in Lake Mohave (Arizona-Nevada, USA) has experienced no natural recruitment for decades and is maintained by harvesting naturally produced larvae from the lake, rearing them in protective custody, and repatriating them at sizes less vulnerable to predation. Analyses of mtDNA and 15 microsatellites characterized for sequential larval cohorts collected over a 15-year time series revealed no changes in geographic structuring but indicated significant increase in mtDNA diversity for the entire population over time. Likewise, ratios of annual effective breeders to annual census size (Nb/Na) increased significantly despite sevenfold reduction of Na. These results indicated that conservation actions diminished near-term extinction risk due to genetic factors and should now focus on increasing numbers of fish in Lake Mohave to ameliorate longer-term risks. More generally, time-series analysis permitted robust testing of trends in genetic diversity, despite low precision of some metrics.

Mastication, digestion, and rapid evacuation rates make visual identification of larval fish remains in the digestive tracts of predatory fishes problematic. Recent advances in molecular technology, however, have increased the likelihood of identifying remnants of partially digested larval prey, thereby enabling assessment of predator impacts on local populations. Conducting controlled laboratory experiments, we evaluated the utility of quantitative polymerase chain reaction (qPCR) for identification of Razorback Sucker Xyrauchen texanus larvae in the digestive tracts of Green Sunfish Lepomis cyanellus and Western Mosquitofish Gambusia affinis. Primers and a probe were developed to amplify a fragment from Razorback Sucker mtDNA. Tests using a suite of potential predators and prey indicated that these primers and probes amplified only mtDNA from Razorback Sucker and Flannelmouth Sucker Catostomus latipinnis, an allopatric and allochronic species in the lower Colorado River. Amplification with primers for Razorback Sucker-specific DNA fragments identified Razorback Sucker DNA in the digestive tracts of 87.5% of Green Sunfish and Western Mosquitofish at 2 h postfeeding. After 12 h, DNA fragments were identified in 75.0% of Green Sunfish but in only 28.6% of Western Mosquitofish. No Razorback Sucker DNA was found in 24-h postfeeding samples, though it was detected in 12.5% of both Western Mosquitofish and Green Sunfish at 48 h postingestion. The sensitivity of qPCR provides a useful tool for extending the time period in which Razorback Sucker larvae can be identified beyond that of traditional visual analyses of stomach contents.

Razorback sucker (Xyrauchen texanus) was once common and widely distributed throughout the Colorado River drainage of western North America. Water development and predation by non-native species led to significant decrease in the species' range, and dramatic reduction in size of remaining populations. Previous analyses of mtDNA variation determined that most variation was found within locations and that haplotypes were randomly distributed relative to geography, indicating these samples represent remnants of a single, basin-wide population. In addition, both diversity and number of haplotypes declined progressively down-to upstream, consistent with geologically-recent expansion into the northern portions of the basin. Analyses of variation at 13 microsatellite loci also identified a decrease in genetic variation from downto upstream, also consistent with the hypothesis of recent expansion. Analyses of population structure identified three distinct groups, but the majority of microsatellite variation was found within populations. Most individuals from the upper Colorado River were identified as a discrete unit. These individuals exhibited high levels of relatedness, indicating this represented an isolated group of closely related individuals. There also were significant differences between populations above and below the Grand Canyon; however, estimates of H were relatively low. Given nothing is known of local adaptation in this species, populations above and below the canyon should be managed as independent units; however, if numbers become too low it will be possible to translocate individuals from southern populations northward to increase levels of genetic variability and decrease relatedness within units. These results also illustrate the need for careful consideration of all available information when using molecular data in identifying units for management.

Gathering sex data is frequently a complicated matter for fish species that lack sexually dimorphic traits. Secondary sexual characteristics are cryptic or nonexistent in many minnows (Cyprinidae) during early life stages or outside of the breeding season, which may require lethal methods for determining sex. This complicates many conservation actions and population genetic studies, including (1) the development of hatchery broodstocks for threatened and endangered species, (2) sex ratio determination in wild populations, (3) the influence of sex ratio on the effective population size estimates, and (4) limitations on the implementation of genetic tagging and parentage assignment in wild populations. Here, we remedy this issue for four western North American chub species (Gila spp.) through restriction-site-associated DNA sequencing of Bonytail G. elegans, which identified a single locus indicating an XY sex determination system and no candidate loci consistent with a WZ system. This locus cross-amplified in three additional Gila species (Chihuahua Chub G. nigrescens, Virgin Chub G. seminuda, and White River Chub G. jordani), although species-specific differences necessitated the development of three novel primer/probe sets. Real-time PCR with hydrolysis probe assays was used to genotype sex-linked single-nucleotide polymorphisms for each species. The measurement of endpoint fluorescence in a real-time PCR thermal cycler allowed for rapid and accurate sex genotyping of these species, with accuracy >97% relative to phenotypic sex identification when three PCR replicates were employed. These assays provided high precision, with unanimous results across >87% of the replicates and a majority consensus genotype for 98.6% of the samples. The locus identified herein was conserved across all of the Gila species that we tested (n = 6) and other western North American cyprinids; however, sex-linked polymorphisms were found only in the four species that are named above. Consequently, this locus has promise for the identification and development of sex-linked assays for other Gila and closely related cyprinids.