The decomposition of autumn shed leaves of hackberry (Celtis occidentalis L.) and elm (Ulmus americana L.) in a gallery forest reach and a grassland reach of Kings Creek was examined by placing 24 artificial leaf packs of each species at the two stream sites. Sampling was done after 24 hours and weekly thereafter for a seven week period. No differences between elm and hackberry leaf decomposition rates were measured. The mean percentage weight loss of elm leaves during the sampling period was 90.3% of their ash-free dry weight in the gallery forest reach and 88.9% in the grassland reach. The decay coefficients were -0.0211 and -0.0199 respectively. Hackberry leaves lost 93.8% of their ash-free dry weight at both sites (decay coefficient of -0.02 5). Larvae of the leaf shredding insect, Tipula, were found to influence the decomposition process in the grassland reach but not in the gallery forest reach. The Tipula larvae were estimated to exert their greatest influence on the elm leaves. Headwater streams in forested ecosystems are characterized by the import of allochthonous organic matter, usually in the form of leaf litter (Fisher and Likens, 1973; Petersen and Cummins, 1974; and others). The decomposition of terrestrial leaf material is a major pathway of energy flow through such stream ecosystems (Anderson and Sedell, 1979). The rate of decomposition is controlled by a variety of factors including stream microbial populations (Suberkropp and Klug, 1976), stream water temperatures (Bott, 1975; Suberkropp and Klug, 1976; Paul et al., 1978), chemical composition of the leaf material (Benfield et al., 1979), invertebrate populations (Anderson and Sedell, 1979; Wallace et al., 1982; Kirby et al., 1983), and current velocity (Boling et al., 1982; Gurtz et al., 1982). However, the initial fragmentation of the leaf material is dependent upon mechanical breakage in conjunction with microbial decomposition (Benfield et al., 1979). Rounick and Winterbourne (1983) observed that differences in leaf decomposition rates were associated primarily with differences in the benthic invertebrates related to the physical characteristics of the streambed. There are several mechanisms This content downloaded from 157.55.39.151 on Fri, 29 Jul 2016 04:00:56 UTC All use subject to http://about.jstor.org/terms 18 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE involved in the degradation of allochthonous organic matter in streams. The interactions among them, while various, are associated with watershed characteristics. Kings Creek is a stream with its headwaters in grassland rather than in forest. The riparian vegetation changes to forest downstream. The upper reaches are bordered by riparian vegetation dominated by grasses. The stream side vegetation of the middle reaches is comprised of small trees, shrubs and grasses. The vegetation change to gallery forest in the lower reaches is abrupt and dramatic. The purpose of this study was to determine the decomposition rates of artificial packs of leaves from elm (Ulmus americana L.) and hackberry (Celtis occidentalis L.) in two of these distinctly different riparian reaches of Kings Creek. The hypothesis tested was that autumn shed leaves of elm and hackberry trees would decompose more rapidly in a gallery forest reach of Kings Creek than in a grassland reach because these leaves form a part of the natural organic input and stream microbes would be more adapted to decomposing the leaves. Additionally, the microbes in the grassland reaches were hypothesized to be less active on substrates not experienced naturally.