The 36Cl dating method is increasingly being used to determine the surface-exposure history of Quaternary landforms. Production rates for the 36Cl isotopic system, a critical component of the dating method, have now been refined using the well-constrained radiocarbon-based deglaciation history of Whidbey and Fidalgo Islands, Washington. The calculated total production rates due to calcium and potassium are 91±5 atoms 36Cl (g Ca)−1 yr−1 and are 228±18 atoms 36Cl (g K)−1 yr−1, respectively. The calculated ground-level secondary neutron production rate in air, Pf(0), inferred from thermal neutron absorption by 35Cl is 762±28 neutrons (g air)−1 yr−1 for samples with low water content (1–2 wt.%). Neutron absorption by serpentinized harzburgite samples of the same exposure age, having higher water content (8–12 wt.%), is ∼40% greater relative to that for dry samples. These data suggest that existing models do not adequately describe thermalization and capture of neutrons for hydrous rock samples. Calculated 36Cl ages of samples collected from the surfaces of a well-dated dacite flow (10,600–12,800 cal yr B.P.) and three disparate deglaciated localities are consistent with close limiting calibrated 14C ages, thereby supporting the validity of our 36Cl production rates integrated over the last ∼15,500 cal yr between latitudes of 46.5° and 51°N. Although our production rates are internally consistent and yield reasonable exposure ages for other localities, there nevertheless are significant differences between these production rates and those of other investigators.