Hormonal herbicides, used in pastures, can suffer drift and reach forests. The sensitivity and potential phytoremediation of native species to herbicide residues should be evaluated. The objective of this study is to evaluate the initial development of native Atlantic Forest tree species in soil contaminated with hormonal herbicides. The experiment was conducted in a greenhouse in a 4 x 8 factorial scheme. The first factor had the control and the herbicide Tordon® in three doses (0.166, 0.333 and 0.666 L ha-1) and the second consisted of the forest species Anadenanthera colubrina (Vell.), Cassia ferruginea (Schrad.) Schrad. ex DC., Dalbergia villosa (Benth.) Benth., Machaerium nyctitans (Vell.) Benth., Machaerium opacum Vogel, Piptadenia gonoacantha (Mart.) JF Macbr., Senegalia polyphylla (DC.) Britton and Rose, Senna macranthera (DC Collad.) HS Irwin and Barnaby. The emergence, height, survival, emergence speed index, intoxication, root volume, stem diameter, root and shoot dry mass, leaf area and leaf numbers of the forest species were evaluated. The A. colubrina, D. villosa and M. opacum initial development was reduced by the herbicides 2.4-D plus picloram residues. S. macranthera and P. gonoacantha are tolerant to this mixture and, therefore, show potential for phytoremediation of degraded areas containing residues of these compounds.
Keywords: Hormonal herbicides; native species; phytoremediation; residual effect.