Abstract. Many models that simulate the long-term response of forests to climatic change use the assumption that northern and southern range limits are caused by the deleterious effects of cold and hot air temperatures, respectively, on individual tree growth and that growth declines symmetrically with air temperatures above and below some optimal value in between these extremes. To test the validity of this assumption, we combined physiological data for black spruce, Picea mariana, growing near the treeline in subarctic Quebec with a model of the biophysical and biochemical effects of temperature on photosynthesis. The physiological conditions allow black spruce to grow over a wider range of air temperatures than is reflected in its geographic distribution. In particular, the physiological data suggest that the northern range limit of black spruce is not caused by the direct effects of cold growing-season air temperatures on tree growth and that growth is optimal, with respect to temperature, at the southern range limit. While pollen data indicate large geographic changes in spruce abundance with past climatic changes, the current analyses suggest that the direct effect of air temperature on individual tree growth has not caused these changes. Until we better understand the effects of air temperature on ecological processes, the efficacy of climatic change analyses must be evaluated in terms of model assumptions.