Individual-level diet variation can be easily quantified by gut-content analysis. However, because gut contents are a 'snapshot' of individuals' feeding habits, such cross-sectional data can be subject to sampling error and lead one to overestimate levels of diet variation. In contrast, stable isotopes reflect an individual's long-term diet, so isotope variation among individuals can be interpreted as diet variation. Nevertheless, population isotope variances alone cannot be directly compared among populations, because they depend on both the level of diet variation and the variance of prey isotope ratios. We developed a method to convert population isotope variances into a standardized index of individual specialization (WIC/TNW) that can be compared among populations, or to gut-content variation. We applied this method to diet and carbon isotope data of four species of frogs of the Brazilian savannah. Isotopes showed that gut contents provided a reliable measure of diet variation in three populations, but greatly overestimated diet variation in another population. Our method is sensitive to incomplete sampling of the prey and to among-individual variance in fractionation. Therefore, thorough sampling of prey and estimates of fractionation variance are desirable. Otherwise, the method is straightforward and provides a new tool for quantifying individual-level diet variation in natural populations that combines both gut-content and isotope data.

Aim Relationships between elevation and litter-dweller harvestman (Arachnida: Opiliones) species richness along three elevational gradients in the Brazilian Atlantic Forest were evaluated. Specifically, three candidate explanatory factors for the observed patterns were tested: (1) the mid-domain effect, (2) the Rapoport effect, and (3) the influence of environmental variables on species density and specimen abundance.Location Cuscuzeiro, Corcovado and Capricó rnio mountains, in Ubatuba (23°26¢ S, 45°04¢ W), a coastal municipality in São Paulo state, south-eastern Brazil.Methods We recorded harvestman species and abundance through active sampling using 8 · 8-m plots in both summer and winter. At each plot we measured the temperature, humidity and mean litter depth. Harvestman species richness per elevational band was the sum of all species recorded in each band, plus the species supposed to occur due to the interpolation of the upper and lower elevational records. Differences between observed and expected species richness per elevational band, based on the mid-domain effect, were examined through a Monte Carlo simulation. The Rapoport effect was evaluated using both the midpoint method and a new procedure proposed here, the 'specimen method'. We applied multiple regression analysis to evaluate the contribution of each environmental variable (elevation, temperature, humidity and litter depth) on species density and specimen abundance per plot.Results Harvestman abundance and species richness decreased at higher elevations in the three mountains. The decrease in species richness was not monotonic and showed a plateau of high species richness at lower elevations. The number of harvestman species per elevational band does not fit that predicted by the mid-domain effect based solely on geometric constraints assuming hard boundaries. Species with their midpoints at higher elevations tended to cover broader elevational range sizes. Both the midpoint method and the specimen method detected evidence of the Rapoport effect in the data. At fine spatial scales, temperature and humidity had positive effects on species density and specimen abundance, while mean litter depth had no clear effect. These relationships, however, were not constant between seasons.Main conclusions Our results suggest that harvestman species density declines at higher elevations due to restrictions imposed by temperature and humidity. We found a pattern in species range distribution as predicted by the elevational Rapoport effect. However, the usual rescue effect proposed to explain the Rapoport effect does not apply in our study. Since the majority of harvestman

No abstract

No abstract

Comparisons of species richness among assemblages using different sample sizes may produce erroneous conclusions due to the strong positive relationship between richness and sample size. A current way of handling the problem is to standardize sample sizes to the size of the smallest sample in the study. A major criticism about this approach is the loss of information contained in the larger samples. A potential way of solving the problem is to apply extrapolation techniques to smaller samples, and produce an estimated species richness expected to occur if sample size were increased to the same size of the largest sample. We evaluated the reliability of 11 potential extrapolation methods over a range of different data sets and magnitudes of extrapolation. The basic approach adopted in the evaluation process was a comparison between the observed richness in a sample and the estimated richness produced by estimators using a sub‐sample of the same sample. The Log‐Series estimator was the most robust for the range of data sets and sub‐sample sizes used, followed closely by Negative Binomial, SO‐J1, Logarithmic, Stout and Vandermeer, and Weibull estimators. When applied to a set of independently replicated samples from a species‐rich assemblage, 95% confidence intervals of estimates produced by the six best evaluated methods were comparable to those of observed richness in the samples. Performance of estimators tended to be better for species‐rich data sets rather than for those which contained few species. Good estimates were found when extrapolating up to 1.8‐2.0 times the size of the sample. We suggest that the use of the best evaluated methods within the range of indicated conditions provides a safe solution to the problem of losing information when standardizing different sample sizes to the size of the smallest sample.