Statistically reliable data on the in situ rates of growth, substrate consumption, and product formation are required to test the validity of the mathematical models developed for microbially enhanced oil recovery and in situ bioremediation processes. A simple, replicable porous-core system that could be aseptically divided into sections at various times was developed to follow the kinetics of microbial growth and metabolism in situ. This core system was used to study the kinetics of growth and the mode of penetration of strains of Escherichia coli through anaerobic, nutrient-saturated, fine Ottawa sand (permeability of 7.0 p.m2 and porosity of 37%) under static conditions. The in situ rate of growth of a wild-type, motile, chemotactic strain, RW262, was two times slower inside cores than it was in liquid cultures. The mode of metabolism of galactose by strain RW262 was not altered inside cores, as acetate was the only product detected either inside the cores or in liquid cultures. Without applied advective force, strain RW262 grew exponentially and moved through cores at a rate of about 0.1 m/day. The cell population moved through cores in a band-like fashion, as the front of the moving cells consisted of high cell concentrations (greater than 105 cells per ml). Until the breakthrough of the cells occurred, galactose consumption and acetate production were observed only in the proximal sections of the core, showing that the cell propagation preceded the complete depletion of the substrate or the accumulation of large amounts of products. A motile, nonchemotactic strain of E. coli (RP5232) penetrated cores faster than did its chemotactic parental strain (RP437), which can be explained by differences in their mode of growth inside the cores. Unlike the wild-type, chemotactic strain RP437, which grew and moved through cores in a band-like fashion, cells of the nonchemotactic strain moved through cores in a diffuse manner, as the front of the moving cells consisted of low cell concentrations (103 cells per ml). The appearance of nonchemotactic cells in a section of the core was not necessarily followed by an increase in cell concentration in that section with time. For the nonmotile strain RP2912, a high cell density (107 cells per ml) in a section of the core was observed before cells were detected in the next section. This suggested that the transport of nonmotile cells through porous material requires a high cell density and may occur by a physical displacement process in which some of the progeny cells are forced into the less populated regions of the core.