We present a physiological model of isoprene (2-methyl-1,3-butadiene) emission which considers the cost for isoprene synthesis, and the production of reductive equivalents in reactions of photosynthetic electron transport for Liquidambar styraciflua L. and for North American and European deciduous temperate Quercus species. In the model, we differentiate between leaf morphology (leaf dry mass per area, MA, g m−2) altering the content of enzymes of isoprene synthesis pathway per unit leaf area, and biochemical potentials of average leaf cells determining their capacity for isoprene emission. Isoprene emission rate per unit leaf area (μmol m−2 s−1) is calculated as the product of MA, the fraction of total electron flow used for isoprene synthesis (ɛ, mol mol−1), the rate of photosynthetic electron transport (J) per unit leaf dry mass (Jm, μmol g−1 s−1), and the reciprocal of the electron cost of isoprene synthesis [mol isoprene (mol electrons−1)]. The initial estimate of electron cost of isoprene synthesis is calculated according to the 1-deoxy- D-xylulose-5-phosphate pathway recently discovered in the chloroplasts, and is further modified to account for extra electron requirements because of photorespiration. The rate of photosynthetic electron transport is calculated by a process-based leaf photosynthesis model. A satisfactory fit to the light-dependence of isoprene emission is obtained using the light response curve of J, and a single value of ɛ, that is dependent on the isoprene synthase activity in the leaves. Temperature dependence of isoprene emission is obtained by combining the temperature response curves of photosynthetic electron transport, the shape of which is related to long-term temperature during leaf growth and development, and the specific activity of isoprene synthase, which is considered as essentially constant for all plants. The results of simulations demonstrate that the variety of temperature responses of isoprene emission observed within and among the species in previous studies may be explained by different optimum temperatures of J and/or limited maximum fraction of electrons used for isoprene synthesis. The model provides good fits to diurnal courses of field measurements of isoprene emission, and is also able to describe the changes in isoprene emission under stress conditions, for example, the decline in isoprene emission in water-stressed leaves.