We describe a modular method for building a large, outdoor experimental stream system that has great flexibility for research projects in fish ecology, behavior, conservation, or management. The system has been in use for more than a decade at the University of Oklahoma Biological Station (Kingston, Oklahoma) and has been used with modification at four other research facilities in the Midwest. Here, we document the system in detail, including specifications for construction of the original system and modifications or improvements at other sites. The system uses commercially available, customized fiberglass round tank and trough units that can be configured in many different ways to create flowing pool and riffle habitats. The system appears to be a good mimic of small natural streams based on system flow rates, establishment of natural substrates and cover, stream chemistry relative to that of a natural creek, and fish behaviors. At least 39 fish species have been used successfully in 1-14-month experiments in these systems and approximately half have reproduced. The system offers great flexibility of design to experimenters, is cost effective, and may be of interest at other facilities that research basic biology, conservation, or management of stream fishes.

No abstract

Collections of 231 catfish from 34 localities were surveyed for mtDNA (399 base pairs of cytochrome b) and morphological evidence of headwater catfish Ictalurus lupus in areas of historical occurrence in Texas and New Mexico. The species is of concern for conservation managers, primarily because of the potential for population losses from competition and hybridization with channel catfish I. punctatus. For cytochrome b, there were two deeply divergent (4.8–6.1%) groups, a channel catfish clade of 14 haplotypes (0.8% to 1.3% divergence) and a headwater catfish clade of two haplotypes (1.0% divergence), associated with morphotypes of channel catfish and headwater catfish, respectively. Morphotypes were based on field identification and a canonical discriminant function utilizing external morphology. All specimens from the Nueces River and the main‐stem Rio Grande and the Pecos River conformed to morphological and mtDNA expectations for channel catfish. Apparently pure populations of headwater catfish were found only in two relatively isolated situations (Rocky Arroyo, New Mexico, and San Solomon Spring, Texas). Additional genetic evidence of headwater catfish was restricted to four populations in streams that are direct tributaries of either the Pecos River or the Rio Grande. Two of these populations (Independence Creek and Dolan Creek—Devils River) were morphologically distinct from, but shifted toward, the morphotype of channel catfish. A third population (Pinto Creek) was morphologically indistinguishable from channel catfish, and individuals from the fourth population (Delaware River) had morphotypes consistent with both species. The Pinto Creek, Independence Creek, and Delaware River populations exhibited mtDNA haplotypes from both species, whereas the Dolan Creek—Devils River population was fixed for a haplotype from the headwater catfish clade. A survey of early collection records tentatively suggests that hybridization between headwater catfish and channel catfish might be a result of relatively recent introduction of the latter to the Rio Grande basin.

No abstract

No abstract