Knowledge of soil phosphorus (P) levels is an essential component of nutrient management planning for crop production and for tools that assess the risk of P loss from agricultural fields. Historically, soil P testing has been used to estimate P availability for crops and is now being used directly or as a component of P indices to assess the risk of P loss from fields to water resources. A basic assumption of soil testing for crop production is that the soil sample collected represents a certain area with an acceptable margin of error. Therefore, appropriate soil sampling methods are needed for effective agronomic P management and environmental P assessment. A recent SERA-17 position paper (Vadas et al., 2006) addressed the issue of soil sampling depth. This position paper summarizes important practical aspects of topsoil P variability (within 20 cm depth) across the landscape with an emphasis on within-field variability, and sampling methods that have been recommended or proposed to assess soil P and suggests practical soil sampling options for agronomic and environmental purposes.