To evaluate the importance of subterranean termites in the detritus cycle it was necessary to determine density and dispersion of termite foraging populations and the types and amounts of wood available to them. Sampling was carried out in a shrub-invaded desert grassland south of Tucson, Ariz. Fifty randomly located, circular quadrats were sampled to study the relationships between foraging termites and the abundance of wood. Standing crop biomass of superficial dead wood was 2127 kg/ha. Five plant species represent 97.6% of the biomass while 6 account for 96.4% of the 450 kg/ha of such wood produced annually. Instantaneous estimates of mean surface foraging numbers ofHeterotermes aureus(Snyder),Gnathamitermes perplexus(Banks),Paraneotermes simplicicornis(Banks),Amitermes wheeleri(Desneux), andA. minimusLight were 6.89, 0.35, 1.31, 1.51, and 0.12 termites/m2. Caste composition varied between the species and was important in determining total biomass. Neither superficial dead wood nor foraging termites were randomly dispersed; rather, they were aggregated on the site. Estimated total number of all 5 termite species was set at 1025/m2with a biomass of 0.414 g/m2, based on a literature ratio between surface and subsurface foragers ofGnathamitermes tubiformans. We have tested the hypothesis that quantity and/or quality of wood available influences termite abundance. Results indicate that the abundance of a particular oligophagous species is correlated with the quantity of a highly perferred wood, where there is no correlation between ployphagous species and the quantity of any species of wood.