Laboratory experiments were performed on four different soils, using 100 cm long columns, to determine the extent of virus movement when wastewater percolated through the soils at various hydraulic flow rates. Unchlorinated secondary sewage effluent seeded with either poliovirus type 1 (strain LSc) or echovirus type 1 (isolate V239) was continuously applied to soil columns for 3 to 4 days at constant flow rates. Water samples were extracted daily from ceramic samplers at various depths of the column for the virus assay. The effectiveness of virus removal from wastewater varied greatly among the different soil types but appeared to be largely related to hydraulic flow rates. At a flow rate of 33 cm/day, Anthony sandy loam removed 99% of seeded poliovirus within the first 7 cm of the column. At flow rates of 300 cm/day and above, Rubicon sand gave the poorest removal of viruses; less than 90% of the seeded viruses were removed by passage of effluent through the entire length of the soil column. By linear regression analyses, the rate of virus removal in soil columns was found to be negatively correlated with the flow of the percolating sewage effluent. There was no significant difference in rate of removal between poliovirus and echovirus in soil columns 87 cm long. The rate of virus removal in the upper 17 cm of the soil column was found to be significantly greater than in the lower depths of the soil column. This study suggests that the flow rate of water through the soil may be the most important factor in predicting the potential of virus movement into the groundwater. Furthermore, the length of the soil column is critical in obtaining useful data to predict virus movement into groundwater.