Most animals experience fluctuations in temperature during development, but studies of energetics have ignored the potential influence of these thermal fluctuations. We measured the energetics of Sceloporus undulatus lizard embryos under two conditions that differ realistically in the mean and variance of temperature (diel cycles of 20 degrees-30 degrees and 20 degrees-34 degrees C). Our goal was to determine whether embryos in warm nests would expend more energy to develop than embryos in cool nests. We quantified metabolic rates during development, durations of incubation, and sizes at hatching. To describe changes in metabolic rate during incubation, we used the Akaike Information Criterion to determine the best statistical model among a set of six candidates. Once the best model was determined, the energetic cost of development was estimated by integrating metabolic rates over the period of incubation. We found that some form of sigmoidal model provided the best fit to the data for the majority of embryos (75%). Although embryos in the warmer treatment hatched earlier, the cost of development (approximate to 1.6 kJ) did not differ significantly between embryos in the two treatments. This estimate of energy expenditure at fluctuating temperatures accords with previous estimates of energy expenditure at constant temperatures, suggesting that embryonic metabolism under realistic thermal conditions does not differ substantially from that under constant conditions.