The GRAVI-1 experiment was brought on board the International Space Station by Discovery (December 2006) and carried out in January 2007 in the European Modular Cultivation System facility. For the first run of this experiment, lentil seedlings were hydrated and grown in microgravity for 15 h and then subjected for 13 h 40 min to centrifugal accelerations ranging from 0.29 x 10(-2) g to 0.99 x 10(-2) g. During the second run, seedlings were grown either for 30 h 30 min in microgravity (this sample was the control) or for 21 h 30 min and then subjected to centrifugal accelerations ranging from 1.2 x 10(-2) g to 2.0 x 10(-2) g for 9 h. In both cases, root orientation and root curvature were followed by time-lapse photography. Still images were downlinked in near real time to ground Norwegian User Support and Operations Center during the experiment. The position of the root tip and the root curvature were analyzed as a function of time. It has been shown that in microgravity, the embryonic root curved strongly away from the cotyledons (automorphogenesis) and then straightened out slowly from 17 to 30 h following hydration (autotropism). Because of the autotropic straightening of roots in microgravity, their tip was oriented at an angle close to the optimal angle of curvature (120 degrees -135 degrees ) for a period of 2 h during centrifugation. Moreover, it has been demonstrated that lentil roots grown in microgravity before stimulation were more sensitive than roots grown in 1 g. In these conditions, the threshold acceleration perceived by these organs was found to be between 0 and 2.0 x 10(-3) g and estimated punctually at 1.4 x 10(-5) g by using the hyperbolic model for fitting the experimental data and by assuming that autotropism had no or little impact on the gravitropic response. Gravisensing by statoliths should be possible at such a low level of acceleration because the actomyosin system could provide the necessary work to overcome the activation energy for gravisensing.