In order to obtain a process capable of speeding the germination of D. tortuosum (Sw.) DC. seeds, two experiments were conducted with one-seeded loment (control) segments submitted to manual threshing; manual threshing followed by chemical scarification, also manual, with sandpaper no 220; mechanical threshing (home processor); chemical scarification with H,S0, (95%) for 1, 5, and 8 min; pre-heating at 530C for 4, 10, and 16 h in forced-air oven; imbibition with H2O at 800C for 1, 3, 5, and 10 min; H20 at 27 C for 2 h; and imbibition with thermal swich (H2O at 80 0C/5 min an d H20 at 130 C/ 1 min) . For treat ment evaluation one used germination (G), soil seed ling emergence (E ), first count of germination (FG), and emergence (FE) tests, germination (I. V.G.) and emergence (I. V.E.) seed index, and seed ling length (S L). A completely randomized block design with 4 replications of 100 (G, FG, I.V. G.) or 20 seeds (SL) per treatment in the first experiment, and 4 replications of 50 (G, FG, I.V.G.) or 100 seeds (E, FE, I. V.E. ) per treatment on the second experiment was used. In the first experiment, the treatments which showed a significant dormancy decrease (D) and consequent germination increase (G) in comparison with the control (D = 82%; G =15%) were, in a decreasing efficacy order: manual threshing and scarification (D=3%; G=92%), mechanical thre shing (D = 13%; G = 81%), imbibition in H20 at 800C for 1 min (D = 68%; G = 29%) and 3 min (D=65%; G=32%). The second experiment (control with D = 93%; G = 3%) pointed out: threshing and manual scarification (D=2%; G=93%), mechanical threshing (D = 2%; G = 87%), imbibition into H20 at 800C for 5min and thermal switch (both with D = 85% and G=11%). The FC, I.V.G ., I.V. E., SL, and E vigour tests confirmed these results. Thus, manual scarification with 220 sandpaper and mechanical threshing (with a processor) are recommended to overcome D. tortuosum seed dormancy.
weed; germination; emergence