The response of large stores of carbon in boreal forest soils to global warming is a major uncertainty in predicting the future carbon budget. We measured the temperature dependence of decomposition for upland boreal peat under black spruce forest with sphagnum and feather moss understory using incubation experiments. CO2 efflux rates clearly responded to temperature, which ranged from -10¿ to +8¿C by ~2¿C increments. At temperatures below 0¿C, significant decomposition was observed in feather moss peat but not in wetter sphagnum peat. Above 0¿C, decomposition was exponentially related to temperature, corresponding to a Q(10) (the ratio of the rate of CO2 evolution at one temperature divided by that at a temperature 10¿C cooler) of 4.4 for feather moss and 3.1 for sphagnum peat. The greatest change in CO2 evolution rate with temperature occurred between -2¿ and 0¿C, which coincided with the phase transition of soil water. We saw no large change in the rate of CO2 evolution between incubation experiments separated by a 6 month storage period for feather moss peat. Stable C isotope measurements of evolved CO2 and the rate of change of CO2 evolution with time suggest different substrates are used to sustain heterotrophic respiration above and below freezing. Radiocarbon signatures of CO2 respired from both types of peat reflected significant contributions from C fixed in the last 35 years (bomb 14C) as well as C fixed prior to 1950. We observed no change in the Δ14C of respired CO2 with temperature. Isotopic signatures of peat components showed that a combination of substrates must contribute to the CO2 evolved in our incubations. Decomposition of fine roots (which made up less than 7% of the total peat C) accounted for ~50% of respired CO2 in feather moss peat and for ~30% of respired CO2 in sphagnum peat. Fine-grained (<1 mm), more humified material that makes up 60--70% of the bulk peat organic carbon contributed significantly to heterotrophic respiration (~30% in feather moss and ~50% in sphagnum moss peat), despite slow decomposition rates. Increased temperatures caused enhanced decomposition from all pools without changing their relative contributions. Because the contribution of peat decomposition is a small portion of total soil respiration at the study site, increased respiration rates would be difficult to measure as increased fluxes in the field. Nonetheless, sustained warming could lead to significant loss of C from these peat layers.