Resumo -A dificuldade de controle das plantas daninhas tem sido considerada como um dos principais entraves ao sucesso dos projetos de recuperação da vegetação em áreas degradadas. O presente trabalho teve o objetivo de avaliar o efeito de diferentes herbicidas pré-emergentes sobre a emergência, sobrevivência e desenvolvimento inicial de plantas das espécies nativas amendoim bravo (Platypodium elegans), mutamba (Guazuma ulmifolia), carobinha (Jacaranda micrantha) e angico cuiabano (Anadenanthera colubrina). O experimento foi desenvolvido em casa de vegetação no delineamento experimental inteiramente casualizado, com cinco repetições. Os tratamentos foram constituídos pela aplicação em pré-emergência dos seguintes herbicidas e doses: isoxaflutole (75, 150 e 300 g i.a. ha -1 ), pendimenthalin (500, 1000 e 2000 g i.a. ha -1 ), trifluralin (600, 1200 e 2400 g i.a. ha -1 ) e uma testemunha padrão sem a aplicação de herbicidas. O efeito dos herbicidas foi avaliado quanto ao número de plântulas emergidas, número de folhas por plântula, altura de plântulas, diâmetro de plântulas na altura do coleto e matéria seca de raiz e folhas aos 35 dias após a germinação. Com os resultados obtidos foi possível afirmar que todos os herbicidas e doses utilizadas não prejudicaram a emergência e o desenvolvimento inicial das espécies P. elegans e J. micrantha. As doses de 150 e 300 g i.a. ha -1 do herbicida isoxaflutole foram não seletivas para a espécie arbórea G. ulmifolia. Apenas a dose de 600 g i.a. ha -1 do herbicida trifluralin apresentou seletividade para a espécie A. colubrina.Abstract -The weed control difficulties has been considered one of the most obstacles to success of forest recovery projects at degraded areas. This experiment aimed to evaluate the different pre-emergent herbicides effect on seed emergence, seedling survival and initial development of Brazilian native tree species Platypodium elegans, Guazuma ulmifolia, Anadenanthera colubrine and Jacaranda micrantha. The experiment was conducted under greenhouse conditions in a completely randomized design with five replications. The treatments were constituted by the pre-emergence application of the following herbicides and doses: isoxaflutole (75, 150 and 300 g a.i. ha -1 ), pendimenthalin (500, 1000 and 2000 g a.i. ha -1 ), trifluralin (600, 1200 and 2400 g a.i. ha -1 ) and untreated herbicide as a check control. The effect of the herbicides was evaluated for the number of seedlings emerged, number of leaves per seedlings, height of seedlings, diameter of stem and dry matter of root and leaves at 35 days after germination. With the results obtained it was possible to affirm that all herbicides and their respective rates were safe to P. elegans and G. ulmifolia trees. Isoxaflutol at 150 and 300 g a.i. ha -1 was not selective to G. ulmifolia specie. Only trifluralin at 600 and 1200 g a.i. ha -1 sowed selectivity to A. colubrina specie.

The occurrence of weeds has been the main obstacle to vegetation recovery in Cerrado areas unduly occupied by agricultural and livestock activities. This study evaluates the effect of different pre-emergent herbicides on the emergence, survival, and early development of seedlings of the native species jatobá (Hymenaea stigonocarpa), urucum (Bixa orellana), dry flour (Albizia hasslerii), and white jurema (Mimosa interrupta). The experiment was carried out under greenhouse conditions in a completely randomized experimental design with five replicates, where the treatments were constituted by herbicides: isoxaflutole, applied at the doses of 100, 200, and 300 g ai ha-1 ; pendimethalin, at 500, 1000, and 4000 g ai ha-1 ; and trifluralin, at 600, 1200, and 2400 g ai ha-1 ; in addition to a control treatment, without the application of herbicides. Herbicide phytotoxicity was evaluated at 35 days after emergence for the number of emerged seedlings, number of leaves per seedling, seedling height and diameter, and root and shoot dry weight. With the results obtained, it is possible to affirm that under the conditions of the experiment, all the herbicides and doses used did not affect the emergence and early development of seedlings of species H. stigonocarpa and B. orellana. For species A. hasslerii, the dose of 300 g ai ha-1 of herbicide isoxaflutole was not selective. All doses of the herbicide trifluralin showed selectivity for species M. interrupta.