Experiments were carried out to investigate the potential for bacterial colonization of different substrates in karst aquifers and the nature of the colonizing bacteria. Laboratory batch experiments were performed using limestone and PVC as substrates, a natural bacterial isolate and a known laboratory strain (Escherichia coli [E. coli]) as inocula, and karst ground water and a synthetic formula as growth media. In parallel, fragments of limestone and granite were submerged in boreholes penetrating two karst aquifers for more than one year; the boreholes are periodically contaminated by enteric bacteria from waste water. Once a month, rock samples were removed and the colonizing bacteria quantified and identified. The batch experiments demonstrated that the natural isolate and E. coli both readily colonized limestone surfaces using karst ground water as the growth medium. In contrast, bacterial colonization of both the limestone and granite substrates, when submerged in the karst, was less intense. More than 300 bacterial strains were isolated over the period sampled, but no temporal pattern in colonization was seen as far as strain, and colonization by E. coli was notably absent, although strains of Salmonella and Citrobacter were each observed once. Samples suspended in boreholes penetrating highly fractured zones were less densely colonized than those in the borehole penetrating a less fractured zone. The results suggest that contamination of karst aquifers by enteric bacteria is unlikely to be persistent. We hypothesize that this may be a result of the high flow velocities found in karst conduits, and of predation of colonizing bacteria by autochthonous zooplankton.