Secondary sewage effluent containing about 3 x 104 plaque-forming units of polio virus type 1 (LSc) per ml was passed through columns 250 cm in length packed with calcareous sand from an area in the Salt River bed used for groundwater recharge of secondary sewage effluent. Viruses were not detected in 1-ml samples extracted from the columns below the 160-cm level. However, viruses were detected in 5 of 43 100-ml samples of the column drainage water. Most of the viruses were adsorbed in the top 5 cm of soil. Virus removal was not affected by the infiltration rate, which varied between 15 and 55 cm/day. Flooding a column continuously for 27 days with the sewage water virus mixture did not saturate the top few centimeters of soil with viruses and did not seem to affect virus movement. Flooding with deionized water caused virus desorption from the soil and increased their movement through the columns. Adding CaCl2 to the deionized water prevented most of the virus desorption. Adding a pulse of deionized water followed by sewage water started a virus front moving through the columns, but the viruses were readsorbed and none was detected in outflow samples. Drying the soil for 1 day between applying the virus and flooding with deionized water greatly reduced desorption, and drying for 5 days prevented desorption. Large reductions (99.99% or more) of virus would be expected after passage of secondary sewage effluent through 250 cm of the calcareous sand similar to that used in our laboratory columns unless heavy rains fell within 1 day after the application of sewage stopped. Such virus movement could be minimized by the proper management of flooding and drying cycles.

Virus movement in soil during saturated and unsaturated flow was compared by adding poliovirus to sewage water and applying the water at different rates to a 250-cm-long soil column equipped with ceramic samplers at different depths. Movement of viruses during unsaturated flow of sewage through soil columns was much less than during saturated flow. Viruses did not move below the 40-cm level when sewage water was applied at less than the maximum infiltration rate; virus penetration in columns flooded with sewage was at least 160 cm. Therefore, virus movement in soils irrigated with sewage should be less than in flooded groundwater recharge basins or in saturated soil columns. Management of land treatment systems to provide unsaturated flow through the soil should minimize the depth of virus penetration. Differences in virus movement during saturated and unsaturated flow must be considered in the development of any model used to simulate virus movement in soils.

Adsorption of poliovirus from primary sewage effluent was similar to that from secondary sewage effluent in both batch soil studies and experiments with soil columns 240 cm long. Virus desorption by distilled water was also similar in a soil column that had been flooded with either primary or secondary effluent seeded with virus. These results indicated that adsorption of poliovirus from primary effluent and virus movement through the soil were not affected by the higher organic content of primary sewage effluent.

Short, frequent cycles of flooding soil columns (2 days flooded and 5 days dry) with secondary sewage effluent caused no net removal of N but transformed almost all of the N to nitrate. The net N removal during longer cycles (9–23 days flooded and 5 days dry) was 30%, and half of the N remaining in the water was concentrated into a wave of high‐nitrate water, which represented 10% of the total volume of reclaimed water and was collected immediately after the dry period. Water collected from the columns after the wave of high‐nitrate water passed contained 67% less N than the incoming sewage water.Alternate flooding and drying periods were necessary for consistent N removal. The net N removal was probably due to a combination of several reactions dominated by denitrification. Cation exchange was important in holding NH4+ in the soil until it could be nitrified, thereby concentrating N into smaller volumes of high‐nitrate water. Denitrification is the logical reaction to investigate for higher net N removal because the soil microorganisms nitrified most of the NH4+ and N can be removed from the system as an inert gas by denitrification.

New and innovative methods for measuring soil loss and its impact on productivity are needed to accurately assess the value of soil lost by erosion. Cesium‐137 (137Cs), a radionuclide from nuclear weapons tests which is strongly adsorbed to clay, can be used to trace sediment movement. Measurements of 137Cs concentrations in uneroded soils across the southern United States indicated that the 137Cs input was proportional to the average annual precipitation. Distribution of 137Cs within the profile was related to soil properties rather than to rainfall. Cesium‐137 measurements on a small native grass watershed in Oklahoma showed considerable spatial variability in the 137Cs concentrations, but the variability was random and was not correlated with changes in slope. Cesium‐137 concentrations measured on a 10‐m grid in an adjacent small watershed that had been cultivated for 8 yr showed differences with slope positions even though only 17.8 Mg/ha sediment had been removed from the watershed during an 8 yr measurement period. Also, 137Cs concentrations in the cultivated watershed were significantly lower than in the uncultivated watershed. These data show that 137Cs measurements can be used at low erosion rates if enough samples are analyzed. Further research is needed to determine the number of samples needed for different watersheds and to refine the method.