Bacteriophage transport was investigated in laboratory column experiments using sandy soil, a controlled field study in a sandy wash, and laboratory experiments using fractured rock. In the soil columns, the phage MS-2 exhibited significant dispersion and was excluded from 35 to 40% of the void volume but did not adsorb. Dispersion in the field was similar to that observed in the laboratory. The phage f2 was largely excluded from the porous matrix of the two fractured-rock cores studied, coming through 1.2 and 2.0 times later than predicted on the basis of fracture flow alone. Because of matrix diffusion, nonsorbing solutes were retarded by over a factor of three relative to fracture flow. Results of both granular-medium and fractured-rock experiments illustrate the inability of a solute tracer to provide estimates for dispersion and effective porosity that are applicable to a colloid. Bacteriophage can be used to better estimate the maximum subsurface transport rate of colloidal contaminants through a porous formation. (Copyright (c) 1989 American Society for Microbiology.)