The restriction of water flow in 1959 in Queen Creek in Whitlow Ranch Dam, Pinal County, Arizona, has caused the development of a 15-ha riparian island upstream behind the dam. The herpetofaunas of the riparian interior, riparian edge, desert wash, and upland habitats were sampled to assess the value of this type of development for mitigating continued losses of riparian habitat. Total species richness was 4 in the riparian interior, 7 in the riparian edge, 14 in the desert wash, and 15 in the desert upland. Many of the locally expected riparian species were absent. The lack of invasion by typical riparian species probably results more from biogeographic considerations and flooding patterns than from structural and physical conditions of this newly formed riparian habitat. Regression models for species abundance emphasize the importance of using floristic information rather than summary variables in developing animal-habitat relationships. J. WILDL. MANAGE. 50(4):752-761 Riparian vegetation supports some of the richest terrestrial vertebrate faunas in the arid Southwest (Carothers et al. 1974, Brown et al. 1977). The dependence of many bird species on riparian plant assemblages has been well documented (Szaro 1980, Ohmart and Anderson 1982). Other riparian fauna-flora relationships are less well known, but there doubtless are mammalian, reptilian, and amphibian species that also depend on riparian zones (Hubbard 1977). Reptiles and amphibians have received relatively little attention with respect to the importance of riparian habitats to their distribution, density, and diversity. Jones (1981a) and Vitt et al. (1981) reported on faunal richness and abundance for riparian communities of reptiles and amphibians, but compiling species lists remains primarily guesswork in most areas. Data resulting from government-sponsored inventories frequently go unpublished or are available only in unreviewed report form (Scott and Campbell 1982). Although some literature is available on habitat requirements of single species or small species groups in riparian habitats (Vitt and Ohmart 1974, van Loben Sels 1976, Vitt et al. 1981, Tinkle 1982) more sitespecific distribution and species-specific habitat use data in riparian situations are needed. Unfortunately, human activities of the southThis content downloaded from 157.55.39.181 on Thu, 29 Sep 2016 05:36:27 UTC All use subject to http://about.jstor.org/terms J. Wildl. Manage. 50(4):1986 DESERT RIPARIAN HERPETOFAUNA * Szaro and Belfit 753 western United States have resulted in degradation of critical wildlife habitat particularly in riparian zones (Patton 1977). Recent studies emphasize the importance of watershed vegetation management and flood control dams for subsequent development of riparian vegetation and their impact on riparian bird populations (Szaro 1981; Szaro and Jakle 1982, 1985). The development of riparian vegetation above and below dams and the creation of artificial oases along canals or other water developments are options available to habitat managers for mitigation of the continued loss of riparian habitat. Primary purposes of this study were to: (1) develop habitat models for reptile and amphibian species abundance, (2) determine the utility of summary variables such as diversity indices and single measures of plant community components such as low or high shrub cover in predicting species abundance, (3) determine the response of a Sonoran Desert herpetofauna to vegetative changes resulting from a flood control dam, (4) determine if there was a subsequent invasion of riparian-dependent species, and (5) determine the mitigation potential of a flood control structure for the habitat degradation of those areas with rich riparian herpetofaunas. Data also were gathered on community composition and species habitat preference in comparison with other riparian areas. We thank R. B. Bury, J. Collins, K. B. Jones, and R. King for their thought-provoking reviews of this manuscript. Special thanks to J. D. Gomez for her help in preparing this manu-