Fatores responsáveis pela presença de populações da mirmecófita tococa guianensis em áreas de cerrado sem formigas mutualistas obrigatórias

Abstract

This study determined the factors that enable the occurrence of populations Tococa guianensis in some areas the Cerrado without obligate ants. More specifically, it was investigated (1) influence of trichomes, opportunistic and obligate ants on leaf herbivory T. guianensis, (2) the pressure of herbivores on T. guianensis differs between these locations and (3) there is a differential investment (tannin concentration, trichomes of density and leaves toughness) between plants from sites with and without obligate ants. It was also assessed whether traits related to mutualism plant (domatia size) are changed due to the lack of mutual partner. Furthermore, we evaluated the factor limiting the occurrence of obligate ants Allomerus octoarticulatus in some sites in do Cerrado. This study was realized in three locations in central Brazil, two of them Aragarças (ARA) in Goiás, Cachoeira da Fumaça (CACHO) in Mato Grosso and Uberlândia (UDI) in Minas Gerais. The effect of trichomes and ants was assessed by a factorial experiment in two populations with obligate ants (A. octoarticulatus) in ARA and CACHO and in an absence of obligate ants in UDI. To investigate the effect if the pressure of herbivory differs between sites with and without associated ant and if there is an investment non-biotic defense was made a reciprocal transplant experiment with seedling originated from sites where the species occurs with and without the mutualistic partner. Moreover, it was also made morphological measurements of domatia different populations. To know which biotic and/or abiotic factor limits the occurrence of ant in UDI was compared to the survival and queen s size of A. octoarticulatusin ARA and in UDI after six months. The trichomes affected herbivory in all three populations, but the differences between herbivory in plants with or without trichomes was relatively small. Opportunistic ants did not affect herbivory rate in UDI. Obligate ants had a strong effect on the herbivory rate in ARA, but surprisingly, not in CACHO. All seedling transplanted, regardless of source, were attacked in more in ARA than CACHO and UDI. In UDI plants were less damaged than in other places. The UDI population showed higher tannin content both in adult and in seedling. The trichomes density also differed between sites, ARA had more trichomes than other places and individuals CACHO showed higher leaf toughness than ARA and UDI. The survival of queen s A. octoarticulatus was lower in UDI than in ARA, but the difference depended on the number of queens that colonized the plant. However, the size of the colonies of queens surviving until the end experiment did not differ between sites.Overall, the results of this study suggest that T. guianensis has more of a defense strategy against herbivory insects and this plasticity defenses maybe favoring the occurrence of local population without the presence ofobligate ants. The absence of A. octoarticulatus in UDI seems to be more related to dispersal ability of queens and distance from local that queens occurs in greater quantities than with abiotic conditions sites.