Temporal variation in the composition of ant assemblages (Hymenoptera, Formicidae) on trees in the Pantanal floodplain, Mato Grosso do Sul, Brazil. In this paper we investigate how seasonal flooding influences the composition of assemblages of ants foraging on trees in the Pantanal of Mato Grosso do Sul. During the flood in the Pantanal, a large area is covered by floods that are the main forces that regulate the pattern of diversity in these areas. However, the effects of such natural disturbances in the ant communities are poorly known. In this sense, the objective of this study was to evaluate the effect of temporal variation in assemblages of ants foraging on trees in the Pantanal of Miranda. Samples were collected during a year in two adjacent areas, one who suffered flooding during the wet period and another that did not suffer flooding throughout the year. In 10 sites for each evaluated habitat, five pitfall traps were installed at random in trees 25 m apart from each other. In the habitat with flooding, the highest richness was observed during the flooding period, while there was no significant change in richness in the area that does not suffer flooding. The diversity of species between the two evaluated habitats varied significantly during the two seasons. Most ants sampled belong to species that forage and nest in soil. This suggests that during the flood in flooded habitats, ants that did not migrate to higher areas without flooding adopt the strategy to search for resources in the tree canopy.

Adenocalymma bracteatum is a shrub of dense foliage and yellow flowers, easily found on grasslands areas in Central Brazil. The aim of this study was to determine the reproductive biology and the flower visitors of A. bracteatum in a pasture area nearby Ivinhema city, MS (Brazil). The flowering peak occurs in winter. The flower reflects ultraviolet light. Anthesis begins at 6:30h, and pollen and nectar are the resources to visitors. We captured 1,038 floral visitors. The bees Apis mellifera (L.), Trigona sp., Trigona spinipes (Fabricius), (Hymenoptera: Apoidea) and the ant Cephalotes sp. (Hymenoptera: Formicidae) were the main visitors. The reproductive tests indicate that A. bracteatum is self compatible, justifying its expansion in altered environments; however, the largest reproductive success was dependant on cross-pollination and self-pollination, evidencing the pollinators importance. Adenocalymma bracteatum presents melittophilous syndrome and bumblebees were the main pollinators in the area. The correlations observed between the climatic variables and the main pollinator species were low or medium.

Diversidade de formigas epigéicas (Hymenoptera, Formicidae) em ambientes no Centro-Oeste do Brasil. Foi comparada, através do uso de índices de diversidade e modelos de abundância de espécies, a diversidade das comunidades de formigas epigéicas que ocorrem em duas estruturas vegetacionais diferentes: mata nativa e cultura de eucalipto. Para a captura das formigas foram utilizadas 800 armadilhas de solo do tipo pitfall, em oito amostras distintas. Um total de 85 espécies, distribuídas em 36 gêneros de sete subfamílias foram coletadas nos dois ambientes, sendo que destas, 83 ocorreram na mata nativa e 60 na cultura de eucalipto. A diversidade de espécies de formigas calculada pelo índice de Simpson não foi significativamente diferente entre os ambientes, ao contrário do resultado obtido a partir da aplicação do índice de Shannon, o qual indicou maior diversidade de espécies na mata nativa. O modelo log-series não se ajustou satisfatoriamente aos dados das comunidades de formigas encontradas na cultura de eucalipto e na mata nativa, mas o modelo log-normal mostrou-se adequado para descrever a estrutura das comunidades dos dois ambientes. O modelo broken-stick, que representa uma comunidade bem estruturada, ajustou-se apenas aos dados da mata nativa.

We compared, through the use of diversity indices and species abundance models, the diversity of communities of epigeous ants occurring in two different vegetational structures: native forest and eucalyptus plantation. To capture the ants, 800 soil pitfall traps were used, in eight sample series. A total of 85 species, belonging to 36 genera of seven subfamilies were collected in the two environments. Of these species, 83 occurred in the native forest and 60 in the eucalyptus plantation. The species diversity as evaluated by Simpson's index was not significantly different between the environments, but the Shannon index indicated a higher diversity of species in the native forest. The log-series model could not be adjusted satisfactorily to the data for the ant communities from either the eucalyptus plantation or the native forest; however, the log-normal model appeared appropriate to describe the community structure in both environments. The broken-stick model, which represents a well-structured community, was only adjusted to the data for the native forest sites.