Resumo. In view of the widespread increase in herbicide-resistant weeds, biotechnology companies have developed dicamba-tolerant soybean and cotton cultivars. This technology can, however, increase the risk of the product drifting to adjacent areas. This study was developed with the objective of the to evaluate the phytotoxicity and biometric variables of young eucalyptus plants exposed to subdoses of the herbicide dicamba. The experiment was carried out under field conditions in Rio Verde, state of Goiás, Brazil. The treatments were represented by the application of 0 (control), 7.5, 15, 30, 60, 120 or 240 g ae ha-1 of dicamba 45 days after the seedlings were planted in the field. In terms of phytotoxicity, the dicamba doses of 120 and 240 g ae ha-1 caused greater damage to the eucalyptus plants in all periods of evaluation. The predominant symptoms were epinasty, increased number of shoots and necrosis and senescence of young branches and leaves. The herbicide doses of 120 and 240 g ae ha-1 significantly compromised plant height and diameter, number of branches and dry mass of leaves and roots, interfering with the growth and development of the eucalyptus crop. The results indicate that the effect of subdoses of the herbicide dicamba can interfere with the proper development of young eucalyptus plants, which may cause losses in the initial planting phase and future losses for producers.