To test for the effects of far-red light on preventing budset in Picea abies, seedlings of six populations originating from latitudes between 67°N and 47°N were grown for 4–8 weeks in continuous incandescent (metal halogen) light at 300 µmol m−2 s−1 and 20°C and then transferred, at the same temperature, to a daily regime of 8 h incandescent light (300 µmol m−2 s−1) followed by 16 h cool white fluorescent light (40 µmol m−2 s−1). (Cool white lamps are deficient in far-red light, with a R/FR ratio of 7.5 compared with 2.0 for the incandescent lamps.) All the seedlings from 67° and 80% of those from 64° stopped extension growth and set terminal buds within 28 days of the change of regime. The seedlings from 61° and further south continued growing, as did control seedlings from 67° grown as above but with incandescent light at 20 µmol m−2 s−1 replacing cool white illumination. To distinguish between a clinal and ecotypic pattern of variation, the interval between 64° and 59° was investigated by growing populations originating from that area in the same regimes as before. After 28 days in the cool white day-extension regime, the percentage budset was 86 for the population from 64°, 0 for the population from 59° and 25–50 for the intermediate populations; i.e. the populations showed a clinal variation in requirement for far-red light according to latitude. Thus northern populations of Picea abies appear to behave as ‘light-dominant’ plants for the photoperiodic control of extension growth and budset, whereas the more southern populations behave as ‘dark-dominant’ plants.