Bioreactors are employed increasingly to remove undesirable components from gas streams and to carry out bioprocesses with gaseous nutrients. The problem is to match the type and detailed design of the bioreactor to the characteristics of the component and the microbial metabolism, in order to meet a given set of process requirements at the lowest possible total cost. The higher the specified fractional removal, the greater the advantage of packed-bed and bubble-column reactors in which the gas approaches plug flow. The variation of the gas/liquid interfacial area with gas flow rate makes bubble columns unsuitable for relatively dilute, insoluble components (e.g., hydrocarbons), but the complete mixing of the Liquid is an advantage with more concentrated, soluble components, particularly those (e.g., H2S, SO2) that dissociate in contact with water and may be inhibitory at their equilibrium concentrations. When the component is the carbon/energy source or electron acceptor for metabolism, the amount of viable biomass in the reactor at steady state is limited to the rate of component removal divided by the maintenance requirement of the biomass for the component. However, the volume of water in which this biomass is suspended is a variable to be fixed by the process designer or operator. Thus, a packed-bed bioreactor may be run as a minimum-water biofilm reactor (given a film-forming organism) or as a trickle-bed reactor containing a dilute cell culture (better when the metabolism requires a soluble nutrient or produces a nonvolatile and inhibitory product). In either case, the optimum operating point, close to the onset of mass-transfer Limitation, can be identified by setting the appropriate definition of the Thiele modulus equal to 1. An analysis based on interpolation between these limiting cases suggests a general definition of the modulus. It can be used in conjunction with mass-transfer and liquid-holdup correlations as a guide for selecting the type and size of packing material, operating strategies, and scale-up procedures, which can then be tested by experiment on particular gas streams.