Soil mapping assigns soil at individual locations to predefined categories (classes). By nature soils exist as a continuum both in the spatial and attribute domains and often do not fit into predefined discrete categories without over-simplification. One approach to mitigate this problem in predictive digital soil mapping is the combination of fuzzy logic-based class assignment with raster GIS representation model. This allows the continuous spatial variation of soils to be expressed at much greater details than what has been achieved in soil survey based on the area-class model. However, such an approach faces at least two challenges: defining the central concept of a soil category, and determing membership for soils at individual locations in a given soil category. Prototype theory offers a potential solution to these two challenges. Emerging from ideas of family resemblance, centrality and membership gradation, and fuzzy boundaries relected in fuzzy set theory, prototype theory stresses the fact that category membership is not homogenous and that some members are better representatives of a category than others. A prototype can be viewed as a representation of the category, that 1) reflects the central tendency of the instances' properties or patterns; 2) consequently is more similar to some category members than others; and 3) is itself realizable but not necessarily an instance. Based on this notion, we developed a prototype-based approach to acquire and represent knowledge on soil-landscape relationships and apply the knowledge in digital soil mapping under fuzzy logic. The prototype-based approach was applied in a case study to map soils in central Wisconsin, U.S.A. The study shows that the soil spatial information derived using our approach is more accurate in terms of both soil series prediction and soil texture estimation than that derived either from the traditional soil survey or from a case-based reasoning approach.