Search

A reliable phylogeny is crucial for understanding the evolution and radiation of animal taxa. Phylogenies based on morphological data may be misleading due to frequent convergent evolution of traits—a problem from which molecular phylogenies suffer less. This may be particularly relevant in oribatid mites, an ancient soil animal taxon with more than 11,000 species, where the classification of species into high-ranking taxa such as superfamilies is equivocal. Here, we present a molecular phylogeny of 317 oribatid, 4 astigmatid and 17 endeostigmatid mite species/taxa based on 18S rDNA sequences. We aimed at testing the validity of the 41 superfamilies of oribatid mites recognized by Norton and Behan-Pelletier (in Krantz and Walter, A manual of Acarology, 3rd ed., Texas Technical University Press, Lubbock, 2009). The results indicate that 17 of the 41 oribatid mite superfamilies are monophyletic but that 18 superfamilies are paraphyletic or polyphyletic (5 superfamilies were only included with one species and Microzetoidea were not included). Our findings point to the importance of convergent evolution in polyphyletic oribatid mite taxa. Convergent evolution and the old age of mites likely resulted in a mosaic-like distribution of morphological characters impeding phylogenetic reconstructions based on morphology, calling for molecular approaches to improve oribatid mite systematics. [ABSTRACT FROM AUTHOR]

Elevational gradients in species diversity and species area relationships are two well established patterns that are not mutually exclusive in space and time. Elevation and area are both considered as good proxies to detect and characterize the patterns of species diversity distribution. However, such studies are hampered by the incomplete biodiversity data available for ecologists, which may affect the pattern perceptions. Using the large dataset of oribatid mite communities sampled in Georgia, we tested the effects of altitude and area on species distribution using various approaches, while explicitly considering the biases from sampling effort. Our results showed that elevation and area are strongly correlated (with increasing absolute elevation, land area decreases) and both have strong linear effects on species diversity distribution when studied separately. Approaches based on multiple regression and direct removal of co-varied factors, indicated that the effect of area can actually override the effect of elevation in describing the oribatid species diversity distribution along with elevation. On the other hand, the bias of sampling proved significant in perception of elevational species richness pattern with less effect on elevational species area relationship. We suggest that the sampling alone may be responsible for patterns observed and thus should be considered in ecological studies when eligible. [ABSTRACT FROM AUTHOR]

Many aboveground animals and plant communities have been studied along elevational gradients whereas studies on soil animals are scarce. Here, we studied oribatid mite community distribution along an elevational gradient from 600 to 2200 m in forest ecosystems of the Western Lesser Caucasus Mountains in Georgia. Overall, 86 oribatid mite species were found at the study sites. Oribatid mite densities were generally low (~9800 ind./m), and 74 % of all species reproduced sexually indicating that resource conditions at the study sites are generally poor. Oribatids mainly comprised Brachypylina (76 %), Mixonomata (13 %), Desmonomata (6 %) and Enarthronota (5 %). Oribatid mite community structure changed along the elevational gradient and the changes correlated with temperature, pH, litter thickness and density of the herb layer. The dominance of sexually reproducing taxa and low overall abundance indicate that the studied elevational gradient is characterized by poor resource conditions for soil microarthropods. Oribatid mite species richness and density declined with elevation suggesting that decreasing temperature in concert with resource limitation is a main driver of oribatid mite communities whereas stochastic factors (such as mid-domain effects) are of minor importance. [ABSTRACT FROM AUTHOR]