Pathogens are the leading cause of surface water impairments in Virginia. Currently, Nonpoint Source (NPS) pollution models are most frequently used to determine the maximum allowable loading rates of bacteria from identified sources and they typically simulate bacterial transport to surface waters as a planktonic or free pollutant. Very few models attempt to partition between the planktonic and attached phases primarily because data on bacteria partitioning during overland flow events are currently not available. A field study was conducted to evaluate the partitioning of E. coli and enterococci between the planktonic and attached phases in runoff from pasturelands and to identify the particle sizes to which the fecal indicators preferentially attached. Transport plots were constructed on pastureland with high vegetative cover to simulate well managed pastureland and bare box plots containing three different soil types were used to simulate bare or overgrazed pasturelands. Partitioning ratios were calculated for both studies and used to compare fecal indicator attachment in runoff from different pastureland conditions, from three soils types and between E. coli and enterococci. The average partitioning ratio in runoff from the plots with high vegetative cover was 0.06 for E. coli and 0.30 for enterococci. Partitioning ratios were much higher from bare soil box plots, ranging from 0.38 to 0.55 for E. coli and 0.42 to 1.79 for Enterococcus. In runoff from both the transport plots and the box plots, at least 50% of all attached cells were associated with particles retained by an 8µm screen. These partitioning ratios can be incorporated into NPS models and used to optimize selection of best management practices.