Aim To assess the hypothesis that free-living prokaryotes show a pattern of 'no biogeography' by examining the scaling of soil prokaryotic diversity and by comparing it with other groups' biogeographical patterns.Location Two sites in the tropical deciduous forest of Chamela, Jalisco, on the western coast of Mexico. MethodsWe examined the diversity and distribution of soil prokaryotes in two 8 × 8 m quadrats divided in such manner that we could sample at four spatial scales. Restriction fragment length polymorphisms of 16S rRNA genes were used to define operational taxonomic units (OTUs) that we used in lieu of species to assess diversity. ResultsWe found highly structured species assemblages that allowed us to reject multiple predictions of the hypothesis that soil bacteria show 'no biogeography' . The frequency distribution of range size (measured as the occupancy of quadrats) of OTUs followed a hollow curve similar to that of vertebrates on continents. Assemblages showed high levels of beta diversity and a non-random nested pattern of diversity. OTU diversity scaled with area followed a power function with slopes z = 0.42 and 0.47. Main conclusionsWe demonstrate a non-ubiquitous dispersal for soil prokaryotes, which suggests a complex biogeography similar to that found for terrestrial vertebrates.

The Cuatro Cienegas basin (Coahuila, México) is a composite of different water systems in the middle of the desert with unusually high levels of endemism and diversity in different taxa. Although the diversity of macrobiota has been well described, little is known about the diversity and distribution of microorganisms in the oligotrophic ponds. Here we describe the extent and distribution of diversity found in aquatic prokaryotic communities by analysis of terminal restriction fragment length polymorphisms (T-RFLP) of 16S rRNA genes and phylogenetic analysis of cloned genes. Twelve locations within the basin were sampled. Among all the samples, we found a total of 117 operational taxonomic units (OTUs) using T-RFLPs, which ranged in any single sample from four to 49. OTU richness and Shannon diversity indices for different sites varied, but none were particularly high. 16S rRNA gene sequence data showed 68 different phylotypes among 198 clones. The most abundant phylotypes were Gamma- and Betaproteobacteria, and extreme halophiles. The differences among sites were significant; 45 TRFs were found only once, and 37% of the total diversity was represented by differences between sites, suggesting high beta-diversity. Further studies are needed to test whether this is a direct consequence of environmental heterogeneity in the basin.

Arid ecosystems are characterized by high spatial heterogeneity, and the variation among vegetation patches is a clear example. Soil biotic and abiotic factors associated with these patches have also been well documented as highly heterogeneous in space. Given the low vegetation cover and little precipitation in arid ecosystems, soil microorganisms are the main drivers of nutrient cycling. Nonetheless, little is known about the spatial distribution of microorganisms and the relationship that their diversity holds with nutrients and other physicochemical gradients in arid soils. In this study, we evaluated the spatial variability of soil microbial diversity and chemical parameters (nutrients and ion content) at local scale (meters) occurring in a gypsum-based desert soil, to gain knowledge on what soil abiotic factors control the distribution of microbes in arid ecosystems. We analyzed 32 soil samples within a 64 m2 plot and: (a) characterized microbial diversity using T-RFLPs of the bacterial 16S rRNA gene, (b) determined soil chemical parameters, and (c) identified relationships between microbial diversity and chemical properties. Overall, we found a strong correlation between microbial composition heterogeneity and spatial variation of cations (Ca2, K+) and anions (HCO, Cl−, SO) content in this small plot. Our results could be attributable to spatial differences of soil saline content, favoring the patchy emergence of salt and soil microbial communities.

The aim of this study was to analyze C and N dynamics, as well as, soil bacterial community structure within soil micro-and macro-aggregates in a tropical deciduous forest in México. We measured, for three landscape positions and three seasons of the year: total, microbial and available forms of C and N; potential C and N mineralization; and soil bacterial communities by using t-RFLPs. The highest total C concentrations were found in the north-slopes and in the dry season (DS) samples. In general, microaggregates had higher concentrations than macroaggregates of available C and N forms, and microbial C. Similarly, micro-aggregates had the highest potential C mineralization and net N mineralization. We detected 149 different OTUs (operational taxonomic units) from which 50% was shared by the two aggregate size fractions, 25% was exclusive to micro-aggregates and the 25% left was found only in macro-aggregates. Top-hills were richer in OTUs than north and south-slopes. The Unweighted Pair Group Method with Arithmetic mean (UPGMA) analysis indicated clear differences in community composition between the two aggregate size-fractions in relation to the presence of OTUs. These results suggest that the main difference between micro-and macro-aggregates is due to the community structure within each soil fraction and this difference could affect soil nutrients dynamics.