Water quality has implications for the health of our ecosystem and the welfare of our population. Agriculture is one of the major contributors of non-point source pollution that contaminates our nation's water supplies. Understanding how farmers substitute manure for commercial fertilizers allows us to better understand the level of nitrogen that enters the soil and can seep into our waterways. In this paper, we explore the factors that help determine farmers' substitution rates between the two types of fertilizers. Location, crop type, and time all could play important roles. We analyze USDA farm level survey data for both crop and livestock farms covering the years 1996 to 2002 to create substitution rate estimates used on corn, soybean, and wheat fields. While the substitution rates we found did not appear to change over the time frame examined, we did find that crop type and location significantly affected the rates that farmers use. Additionally, and perhaps most importantly, the substitution rates we did find did not conform to the "perfect substitution" between N coming from commercial sources and manure used in much of the literature. This means that previous studies could have underestimated the potential level of pollution of our water by our nations' farms.