The influence of daily light integral (DLI) before vernalization and vernalization temperature and duration on growth and flower development was determined for seed-propagated perennials Aquilegia xhybrida Sims 'Remembrance', Coreopsis grandiflora Hogg ex Sweet 'Sunray', and Lavandula angustifolia Mill. 'Hidcote Blue'. Seedlings were grown under two DLIs (4 or 14 mol.m(-2).d(-1)) for 5 weeks before being vernalized at -2.5, 0, 2.5, or 5 T for 2,4,5, or 8 weeks. 'Remembrance' and 'Sunray' plants were vernalized in the dark, while 'Hidcote Blue' plants were vernalized in light at 5 to 10 mumol.m(-2).s(-1) for 9 hours/day. After vernalization, plants were forced under a 16-h photoperiod in the greenhouse at 20+/-2 degreesC. 'Remembrance' plants flowered uniformly when vernalized at 0 to 2.5 T for 2 weeks or longer, and flower number, plant height, time to visible bud or to flower were generally not influenced by vernalization temperature or duration. No 'Sunray' plants flowered without vernalization, and only a low percentage flowered with 4-week vernalization. Compared with low DLI, a 14 mol.m(-2).d(-1) before vernalization delayed flowering by 7 to 20 days in 'Remembrance', but there were no substantial differences in flowering characteristics of 'Sunray'. 'Hidcote Blue' plants were best vernalized in the light at 5 degreesC for 8 weeks to obtain rapid and uniform flowering and the highest number of inflorescences. Flowering and survival percentages of 'Hidcote Blue' were much lower for plants at 14 mol.m(-2).d(-1) DLI compared to 4 mol.m(-2).d(-1).