Imbibition of tobacco seed for 24 hours and then air-drying increased subsequent germination in the dark. Exposure to light during this pre-sowing treatment further increased dark germination. Germination performance in parallel tests in the light indicated that the pre-sowing treatments broke dormancy in all seeds that were dormant but had not affected their viability. Dormancy varied between cultivars. In some, all viable seeds were quiescent. In others dormancy could be completely broken either by the imbibition and drying treatment on its own or by this treatment in the presence of light, or by both treatments. Although the germination characteristics of the different cultivars may have been inherited, they could equally have been due to the provenance of their respective seeds. For example, in one cultivar (Kutsaga E1), less than 20 per cent of 3-month old seeds were quiescent, compared with more than 50 per cent in older seeds. In about 20 per cent of the seeds dormancy could be broken by imbibition and drying alone, whereas light was also required to break it in the remainder. Dormancy is usually broken within a few days of sowing under normal (light) conditions. Nevertheless, even under these conditions, the pre-sowing treatments improved germination rate. They could, therefore, assist seedling uniformity by increasing the rate of synchronisation of germination.