This article reviews the state of research on sorption of gases (CO(2), CH(4)) and water on coal for primary recovery of coalbed methane (CBM), secondary recovery by an enhancement with carbon dioxide injection (CO(2)-ECBM), and for permanent storage of CO(2) in coal seams. Especially in the last decade a large amount of data has been published characterizing coals from various coal basins world-wide for their gas sorption capacity. This research was either related to commercial CBM production or to the usage of coal seams as a permanent sink for anthropogenic CO(2) emissions. Presently, producing methane from coal beds is an attractive option and operations are under way or planned in many coal basins around the globe. Gas-in-place determinations using canister desorption tests and CH(4) isotherms are performed routinely and have provided large datasets for correlating gas transport and sorption properties with coal characteristic parameters. Publicly funded research projects have produced large datasets on the interaction of CO(2) with coals. The determination of sorption isotherms, sorption capacities and rates has meanwhile become a standard approach. In this study we discuss and compare the manometric, volumetric and gravimetric methods for recording sorption isotherms and provide an uncertainty analysis. Using published datasets and theoretical considerations, water sorption is discussed in detail as an important mechanisms controlling gas sorption on coal. Most sorption isotherms are still recorded for dry coals, which usually do not represent in-seam conditions, and water present in the coal has a significant control on CBM gas contents and CO(2) storage potential. This section is followed by considerations of the interdependence of sorption capacity and coal properties like coal rank, maceral composition or ash content. For assessment of the most suitable coal rank for CO(2) storage data on the CO(2)/CH(4) sorption ratio data have been collected and compared with coal rank. Finally, we discuss sorption rates and gas diffusion in the coal matrix as well as the different unipore or bidisperse models used for describing these processes. This review does not include information on low-pressure sorption measurements (BET approach) to characterize pore sizes or pore volume since this would be a review of its own. We also do not consider sorption of gas mixtures since the data base is still limited and measurement techniques are associated with large uncertainties. (C) 2011 Elsevier B.V. All rights reserved.