Detection of Larval Remains after Consumption by Fishes JASON D. SCHOOLEY,* ABRAHAM P. KARAM,BRIAN R. KESNER,PAUL C. MARSH, CAROL A. PACEY, AND DARREN J. THORNBRUGH 1 School of Life Sciences, Arizona State University, Post Office Box 874601, Tempe, Arizona 85287-4601, USA Abstract.—In southwestern North America, consumption of native fish larvae by nonnative predators has imperiled native populations. Field-acquired dietary analyses have provided little evidence of this cause–effect relationship. In this study, small, nonnative green sunfish Lepomis cyanellus, bluegills L. macrochirus, red shiners Cyprinella lutrensis, fathead minnow Pimephales promelas, and yellow bullheads Ameiurus natalis were each fed a single larva of the native razorback sucker Xyrauchen texanus. Gut content analysis revealed that prey detection generally became increasingly difficult over a short postconsumption time period under laboratory conditions. For green sunfish, bluegills, and yellow bullheads, significant relationships between prey detection and time were revealed; the probability of prey identification was initially 50% or greater for about 30 min postconsumption, whereas few prey (3%) were identifiable at 60 min postconsumption. For red shiners and fathead minnow (pooled for analysis), no relationship was evident; these two species completely masticated their prey, thus hindering identification. Green sunfish and bluegills swallowed prey whole, and yellow bullheads damaged larvae during consumption. Many larvae were discovered in the foregut, and 25% were regurgitated during predator fixation. Use of gut content analysis as evidence of predation on native fish larvae by small, nonnative fish is problematic and unreliable due to rapid mechanical and chemical digestion of fragile larval tissues. Declines of native fish populations have been exacerbated in part by nonnative fish predation on early life stages (Courtenay and Stauffer 1984; Lever 1996; Fuller et al. 1999). In southwestern North America, there is increasing evidence that native fishes can be quickly extirpated when their larvae occupy the same habitat as nonnative fishes (Johnson and Hines 1999; Dudley and Matter 2000; and similarly, Lemly [1985]); there is also direct evidence of native young being consumed en masse by nonnative fishes (Marsh and Langhorst 1988). Still, few direct observations support the broad-scale recruitment losses that can occur when native fish larvae encounter nonnative fishes. Most studies have cited difficulties with identification of small larvae or eggs in predator gut contents due to rapid digestion (Hunter 1981; Folkvord 1993; Kim and DeVries 2001). Explanations for this paucity of evidence may include specimen collecting and preservation techniques (Crowder 1980), timing of collections and gut analysis (Ruppert et al. 1993), predator digestion and evacuation rates (Lohr and Fausch 1996; Brandenburg and Gido 1999), and fragility of larval tissues (Brandt et al. 1987). The intent of this study was to explore these issues with a laboratory experiment on gut analysis of nonnative predators performed within short time intervals after consumption of a native fish larva. The razorback sucker Xyrauchen texanus is an example of a native fish with successful annual reproduction but near-total recruitment failure (Marsh et al. 2003; Minckley et al. 2003). Most field-collected evidence points to consumption of all larvae by nonnative fishes, particularly green sunfish Lepomis cyanellus, as a likely cause for population collapse (Marsh and Langhorst 1988; Johnson and Hines 1999). Results under controlled laboratory conditions demon- strated the likely difficulties in field detection and assessment of the effect of predation by small-bodied, nonnative predators. Methods During 23–24 February 2006, individuals of five regionally common nonnative species (Table 1)—red shiner Cyprinella lutrensis, fathead minnow Pime- phales promelas, green sunfish, bluegill L. macro- chirus, and yellow bullhead Ameiurus natalis—were collected from the Salt River and Sycamore Creek, Maricopa County, Arizona, by use of a backpack electrofisher (Smith-Root, Inc.; Model 12-A POW [programmable output waveform]). Red shiners, green sunfish, and yellow bullheads were targeted as representative common nonnative predators of native larvae, but sampling provided less-than-adequate abundances of these species. Bluegills and fathead minnow were readily available and were therefore collected as a contingency. Fish were transported under oxygenated conditions to a laboratory at Arizona State University (ASU), where each species was assigned to one of five 37.9-L aquaria. Feed was withheld from nonnative predators for 72 h before the experiment began. * Corresponding author: jdschooley@gmail.com 1 Present address: Division of Biology, Kansas State University, 232 Ackert Hall, Manhattan, Kansas 66506, USA. Received August 3, 2007; accepted December 22, 2007 Published online June 19, 2008 1044 Transactions of the American Fisheries Society 137:1044–1049, 2008 Ó Copyright by the American Fisheries Society 2008 DOI: 10.1577/T07-169.1 [Note]