Tropical mountain rain forests are among the most species rich regions in the world. The RBSF (Reserva Biológica San Francisco) area in southern Ecuador is a hotspot of biodiversity. To characterise this exceptional high diversity it is essential to investigate the complex interactions between the organisms and their abiotic environment. A number of groups of organisms, including birds, moths, orchids and mosses, have been investigated, whereas knowledge on others, including soil invertebrates, is minute. This study focuses on (1) the characterisation of the hidden diversity of soil animal species, in particular that of oribatid mites (Oribatida), (2) the composition and function of soil microarthropods including their trophic relationship, (3) decomposition processes including altitude (temperature, moisture), litter type (low and high quality) and microarthropods (small and large mesh of litterbags). (1) At the study site 193 species of 48 families of oribatid mites were determined which is high but compared to temperate forest ecosystems not extraordinary divers. The proportion of not described species is assumed to be around 40 %. Nine new species of ptyctimous oribatid mites were described. Many species are restricted to tropical regions indicating specific community structures in the tropics. Furthermore, the oribatid mite community differed along the elevation gradient from 1850 to 2270 m indicating distinct oribatid mite communities at different altitudes. (2) The high diversity of oribatid mites and decomposer soil animals in general, is one of the great riddles in soil ecology. Whereas in deciduous forests the soil microarthropods are well characterised, taxonomical knowledge in tropical rainforests and especially in mountain regions is low. One aim of this study was to investigate the microarthropod soil community. The density of microarthropods and oribatid mite communities at three horizons (L, F/H, Ah) and on the bark of adjacent trees along an elevation gradient (1850, 2020, 2200 and 2270 m) in the RBSF forest were studied. Oribatid mites were the most abundant group followed by Collembola and Gamasina. The compared to temperate forests generally low number of microarthropods declined with elevation in the order 1850 > 2020 > 2270 ~ 2200 m and with soil depth (L, F/H, Ah). Microarthropods on bark were less abundant than in soil. Declining numbers of soil microarthropods with altitude is concluded to be due to harsh abiotic conditions (e.g. lower temperatures, waterlogging, solar radiation). Low microbial biomass and resource quality presumably also contribute to the low abundances of soil microarthropods. The trophic structure of 32 species and potential basal food resources (litter of Graffenrieda emarginata) was investigated by analysing natural variations in stable isotope ratios (15N/14N; 13C/12C). The results indicate that the soil food web is similar to that of temperate forests and spans about four trophic levels. Primary decomposers, i.e. litter feeding species, were rare, potentially reflecting low litter quality. A large number of ‘decomposer’ animals were in fact predatory or necrophagous, suggesting that various putative decomposer soil animal species (especially in oribatid mites) presumably feed on other soil invertebrates, in particular nematodes or animal carcasses. (3) We studied leaf litter decomposition of two abundant tree species with higher and lower litter quality (Graffenrieda emarginata, Purdiaea nutans) and a mixture of both in a litterbag field experiment at two altitudes (1850 and 2280 m). Litter quality was measured as microbial biomass and N content. Decomposition rates at the studied tropical mountain rain forest were generally low (average of 32 % y-1). The slow decomposition rates may have been due to low litter quality (e.g. high C-to-N ratios) but also due to low temperatures. Litter generally decomposed slower at higher elevation supporting our assumption that temperature is a major driving force for litter decomposition at our study sites and indicating that organic matter accumulates at high altitudes. P. nutans litter after 2 and 6 months of exposure decomposed slower than that of G. emarginata, but not at the end of the experiment, after twelve months. We suggest that litter chemistry affect decomposition mainly at early stages of decomposition. After 12 months the mixture of G. emarginata and P. nutans litter decomposed significantly faster than both single litter types indicating that combining the two litter types accelerates decomposition processes ('non-additive effect'). Soil microarthropods contributed little to decomposition processes. Microbial biomass and density of microarthropods in the litterbags were higher at 1850 than at 2280 m indicating higher biological activity at lower altitudes. Species composition was similar in both litter types supporting previous findings that the structure of soil decomposer microarthropod communities is little affected by litter type.