In this paper we describe the effects of photoperiod (9, 12, 15, 17, 19 and 21 h) and temperature (21/17 and 16/12°C) on rate of leaf appearance and phyllochron in two spring wheats, a semi-winter wheat, and a winter wheat. Under long photoperiods only, all leaves on the main culm of a cultivar emerged at a common rate within a temperature regime, so it was acceptable to assign a specific phyllochron to leaves irrespective of their level of insertion. Increased temperature significantly decreased phyllochron, but the degree of this effect differed between cultivars. As photoperiod was shortened below the optimum, phyllochron lengthened marginally and similarly in all varieties (by approximately 0.1 days per hour change in photoperiod). For very short photoperiods this was true only for the first six leaves, whilst for leaves at higher insertions there was a major effect of reducing photoperiod on lengthening phyllochron. The actual daylength required for producing this major effect on phyllochron was cultivar-dependent. These results suggest that, when making predictions of heading date using phyllochron, it may not be acceptable to assume that leaf number and time are always linearly related at shorter photoperiods, particularly when considering leaves at higher insertions.