Total gas storage capacity in many shale gas reservoirs arises from three sources: compressed gas in the micro-and nano-scale pores, adsorbed gas on the inner surfaces of pores in kerogen-the main constituent of organic materials in shale-and dissolved gas molecules in kerogen. The storage capacity of and transport processes for each of these three sources of gas are different, and accordingly, appropriate models and production strategies should be developed on the basis of actual physics. Of great relevance to field development and management is knowing the contribution of each source to daily production and ultimate gas recovery. A valuable body of literature has addressed the first two sources, but the last source has not been studied in detail. We developed a technique to measure the gas capacity from each of the above-mentioned three sources simultaneously, in a small piece of a shale sample, through batch pressure decay experiments. Temporal pressure decay is recorded using a quartz high-resolution pressure transducer for several days. The pressure-decline curve shows distinctive slope changes representing different storage processes. Pressure decline at the earliest time represents gas migration into the micro- and nano-pores. This is followed by a change in the slope of the pressure decline in an intermediate time, representing gas adsorption onto the inner surface of kerogen pores. At the later time, the slope of the pressure decline changes again, representing gas diffusion into kerogen. In the samples we tested, we found that dissolved gas in kerogen can contribute about 22% of the total gas-in-place. Assuming that gas molecules diffuse into the walls of the pores in kerogen, we used a Fickian diffusion model and a parameterestimation technique to estimate the gas molecular diffusion coefficient in kerogen. We measured the diffusion coefficient of methane in amorphous kerogen as being on the order of 10(-20)m(2)/s. (C) 2013 Elsevier B.V. All rights reserved.