Field manipulations of forest litter were conducted to measure effects of litter structure on abundances of forest floor arthropods in natural and artificial leaf litter. Artificial litter was made of nondecomposable vinyl. Both artificial and natural litter occurred in three structural treatments: flat, natural and complex. Using this design we could compare the interacting effects of litter as a trophic base and as a spatially heterogeneous environment on litter arthropods. In 1 of 4 months of the study, microAraneae were significantly (p< 0.05) more abundant in litter with complex structure: both natural and artificial. In contrast, Collembola and Acari abundances were unaffected by structural complexity of litter but were significantly (p<0.05) higher in natural leaves compared to artificial leaves. Abundances of dominant macroarthropod predators (Araneae) increased significantly (p<0.05) with increased litter structural heterogeneity in 1 of 4 study-months. Non-Araneae predators showed weak or no responses to litter complexity. Diplopods and isopods were more abundant in natural litter, while other nonpredatory macroarthropods were equally abundant in natural and artificial litter. INTRODUCTION Comparison of litter-dwelling arthropods from different forests shows arthropod abundance increased with increased litter depth and complexity (Lowrie, 1948; Luczak, 1963; Duffey, 1966; Berry, 1967; Huhta, 1971, 1976; Jocque, 1973; Uetz, 1975, 1976; McBrayer et al., 1971; Seastedt and Crossley, 1981; Bultman et al., 1982). Unfortunately, interpretation of these correlative studies is hampered because forests with structurally complex litter (composed of curly and twisted leaves rather than flat ones) tend also to have deep litter. [This correlation, however, does not always hold; litter of beech-maple forests is structurally simple (being primarily flat beech leaves) but often quite deep.] As a consequence of this general trend between litter depth and complexity, effects of these litter parameters on arthropod abundance were confounded. Further, effects of changes in litter depth on arthropod abundance may be caused by factors associated with litter depth, such as the habitat space or nutritional base available to arthropods. In sum, from previous work we cannot determine whether abundances of litter-dwelling arthropods are affected by the depth, structural complexity or nutritional quality of litter. To elucidate effects of litter complexity and nutritional quality on arthropod abundance, we conducted field experiments using natural and artificial leaves of varied structure and constant depth. Artificial, nonnutritive leaves provided a control for litter acting as a trophic base for the arthropod community; thus, influences of structure and nutritional content could be separated. We compared arthropod faunas inhabiting artificial leaves of varied structure (flat, natural and complex) to those of natural leaves of identical structural regimes.