Species composition and autecology of basidiomycete assemblages and decomposition of wood from woodrat middens in the Chihuahuan and Sonoran deserts

Although wood decomposition has been examined in arid habitats, the contributions of basidiomycetes to this process and the ecology of these organisms have not been studied extensively. Fungal activity in arid environments is restricted by high temperatures and low water potentials of the substrates for much of the year. This study focused on the autecology and spatial and temporal patterns in species composition of wood decomposing basidiomycete assemblages associated with woodrat (Neotoma spp.) middens in the Chihuahuan and Sonoran Deserts. Selected fungal isolates were examined to determine growth responses to selected temperatures and water potentials, and to determine their contributions to wood decomposition in woodrat middens.

Composition of basidiomycete assemblages associated with woodrat middens differed significantly in the two deserts with respect to temporal and spatial patterns. Species richness was greater in the Sonoran than in the Chihuahuan Desert Fungi from both deserts; however, had temperature and water potential optima that were similar to those of fungi from mesic environments. Decomposition rates were greatest in association with Phanerochaete omnivorum. a dominant basidiomycete of the Sonoran Desert. Overall, decomposition was greatest for wood in the Sonoran Desert versus in the Chihuahuan Desert. These differences may be the result of historical development of the two deserts: the Sonoran traditionally has been considered a woody desert, whereas the Chihuahuan has only recently been converted from grassland to desert shrub habitats via desertification. Additionally, the Sonoran desert is warmer than the Chihuahuan Desert, and this is reflected in the temperature optima for isolates obtained from the Sonoran compared to those for the Chihuahuan. Of the Sonoran isolates examined, two had temperature optima of 20°C, six had optima of 25°C, one exhibited no difference at 25-30°C, and five had optima of 30°C. This is in contrast to the Chihuahuan Desert isolates: one had an optimum of 20°C, whereas the other nine had optima of 25°C. None had optima above 25°C.

Peak numbers of isolates from both deserts were collected from May through October, with peak numbers from the Chihuahuan occurring in late summer through autumn. In the Sonoran Desert, greater numbers of isolates were obtained in late winter and early spring. These differences in seasonal patterns in isolate densities may be attributed to the difference in rainfall patterns between the two deserts. Dominant basidiomycete taxa coUected in the Chihuahuan Desert were Peniophora tamaricicola (42% of all Chihuahuan isolates) and Hyphoderma pallidum (9%). In the Sonoran Desert, Peniophora tamaricicola (28% of Sonoran Desert isolates), Mvcoacia austrooccidentale (19%), Phanerochaete omnivomm (6%). and Hvphoderma sp. (5%). Peniophora tamaricicola. believed to be an introduced species, and Hyphoderma spp. were among the few species in common between the deserts. Peniophora tamaricicola accounted for 25% of the total number of isolates from both deserts.

While species composition of the basidiomycete assemblages from the middens is determined by abiotic characteristics of each desert as well as differences in woody substrates, the main factor which determine the structure of these assemblages is the midden microenvironment Midden microenvironment may also account for the lack of significant differences in the numbers of species and isolates per midden for both deserts. Middens may represent isolated "islands," with each midden developing a unique assemblage of fungal species characteristic of the particular ecosystem. Analogous to the fungal unit communities that develop on other discrete pieces of organic matter, such as dead leaves, is the occurrence of specific basidiomycete species in each midden with few taxa occurring in all middens. The basidiomycete species composition of a particular midden may reflect primarily the type of material collected, the random colonization of woody litter prior to inclusion in the midden, and the movement of material between middens.