Track plates and cameras are proven methods for detecting and identifying fishers (Martes pennanti) and other mesocarnivores. But these methods are inadequate to achieve demographic and population-monitoring objectives that require identifying sex and individuals. Although noninvasive collection of biological material for genetic analysis (i.e., hair-snaring methods) may help achieve these objectives, they have yet to be evaluated. We incorporated wire- and glue-snares into track-plate enclosures in bait stations deployed at 3 locations in California, USA, in 2002 and 2003 to compare their efficacy. We detected 5 species of carnivores via their tracks, fisher and marten (M. americana) most frequently. We collected 96 hair samples, 71 (80%) of which had sufficient DNA to yield a species identification. The wire-snares were more permeable to all species, but both snares were permeable to fishers. Glue-snares were more effective at collecting hair than the wire-snare configuration used. Small species, such as the marten, did not readily leave hair on the wire-snare. Glue was more cumbersome to handle than wire, but, given that it collected more hair and more reliably, we favor use of a glue-snare for fisher surveys. Glue also was effective at collecting hair that was correctly identified as marten on 72% of visits by martens. The hair-snaring method resulted in a 58% and 75% rate of successful identification of fishers that entered the enclosure for wire and glue, respectively. Most failures were due to either insufficient DNA in the sample or no hair snared during a visit to the bait. The success of verifying the presence of a species also was affected when a second species visited the station between check intervals. We believe that changes in laboratory technique can mitigate this problem. Although only 75% of the visits by fishers to glue-snares resulted in hair that yielded sufficient DNA for confirmation, we believe this success rate guarantees a high probability that the presence of a fisher would be confirmed by at least one of their visits to a multiple-station sample unit in which each station is checked several times. A sample of hair from fishers and martens were selected for individual identification, and 9 of the 12 fisher samples had sufficient DNA to identify a minimum of 6 different individuals. Track and genetic methods need not compete for use by carnivore surveyors; each is inexpensive enough when deployed in the integrated unit we tested here to justify their use as companion methods. However, additional innovations will be necessary to increase the number of species that can be detected by track and genetic means at the same location.