Resumo A espécie Glycoxylon inophyllum (Mart. ex Miq.) Ducke tem como centro de dispersão a Amazônia. Enfatizado o comportamento ecológico desta planta, comum em campinas amazônicas, sob dois aspectos: 1 — quanto a sua variação estrutural dentro de uma área de campina da Reserva Biológica do Convênio INPA-SUPRAMA, onde duas formações típicas de campina amazônica estão presentes, campina e campinarana; 2 — quanto a aspectos químico-ecológicos por ela produzidos sobre plântulas da mesma espécie ou de outras, principalmente Lagenocarpus verticillatus (Sprengel) T. Koyama & Maguire. Para este segundo aspecto, é feito um estudo de competição por nutrientes e umidade do solo, para melhor conhecimento do comportamento da espécie no ecossistema.

Summary A study of the ecology of Glycoxylon inophyllum (Mart, ex Miq.) Ducke was made considering two main aspects: 1) the structural variation of the species in a transect through campinarana-campina-campinarana measuring 60x5 m. This was correlated with variations in water table. Comparison was made between density of individuals, height of tree, basal area and size of crown. The results showed that the number of individuals of G. inophyllum is greater in the central campina than in the campinarana. Also in the campina the trees are more tortuous and rachitic. 2) The ecological chemistry of the species. Here, firstly studies of competition for nutrients and soil moisture were made. The results showed that there is little dependence of G. inophyllum seedlings on the nutrients when they are transplanted to the humic soils of G. inophyllum and Aldina hetrrophylla Spr. ex Benth. The soil moisture does not affect the seedling growth in campinarana. The lowest value obtained during the period November 1974 — 1975 April was 64.46% in December. Thus, this is not a limiting factor, because the frequent rains supplied sufficient moisture for normal plant growth. Finally, tests were made for allelopathy in both the laboratory and the field. In the laboratory a bioassay of the hypocotyil was made growing lettuce (Lactuca sativa l.) in Petri dishes on filter paper watered with aqueous extracts of intact leaves, cut leaves, intact roots, bark flake from the trunk, and finally in stems leached with water. Controls with distilled water were made. The results showed that hypocotyls of lettuce were growth stimulated by the intact leaf extracts, while the other extracts produced growth inhibition. However, only the extract of bark flakes showed a really significant growth difference from the control. In addition, tests were made with seeds of native species of the campina (Glycoxylon inophyllum and Lagenocarpus verticillatus (Sprengel) T. Koyama & Maguire). In the laboratory results were not obtained because of interference by Boletus fungi. In the field two areas of 2m were delimited for each species on the humic soils of G. inophyllum and A. heterophylla. These areas were sown with seeds of L. sativa and Glicine max (l.) Merr. In the case of L. sativa a germination of 20% was obtained, followed by a decline in seedling nuber due to mortality. In the case of G. max germination of 50% was obtained in the area of G. inophyllum and 70% in the area of A. heterophylla, also followed by a decline of seedlings. The reason for the decline in both areas was a massive attack by herbivores. Presumably this indicates thas these plants are not toxic to the herbivores of the campina. G. inophyllum plants that are near to these sample areas did not show signs of predation. We conclude that Janzen (1974) is correct in affirming that tropical species of campina accumulate toxic materials in their organs to avoid excessive attack by herbivores.