RESUMO
Hyperarid Atacama soils are reported to contain significantly reduced numbers of microbes per gram of soil relative to soils from other environments. Molecular methods have been used to evaluate microbial populations in hyperarid Atacama soils; however, conflicting results across the various studies, possibly caused by this low number of microorganisms and consequent biomass, suggest that knowledge of expected DNA concentrations in these soils becomes important to interpreting data from any method regarding microbial concentrations and diversity. In this paper we compare the number of bacteria per gram of Atacama Desert soils determined by real-time quantitative polymerase chain reaction with the number of bacteria estimated by the standard methods of phospholipids fatty acid analysis, adenine composition (determined by liquid chromatography - time-of-flight mass spectrometry), and SYBR-green microscopy. The number determined by real-time quantitative polymerase chain reaction as implemented in this study was several orders of magnitude lower than that determined by the other three methods and probably underestimates the concentrations of soil bacteria, most likely because of soil binding during the DNA extraction methods. However, the other methods very possibly overestimate the bacteria concentrations owing to desiccated, intact organisms, which would stain positive in microscopy and preserve both adenine and phospholipid fatty acid for the other methods.