The 'Soil Landscape of Canada' (SLC) is a consistent and comparable source of information about soil and land attributes nation wide. It is a large scale planning tool used to assess land use potential for large areas of Canada. It maps soil landscape polygons at a scale of 1:1,000,000. The SLC mapping in southern Ontario is a compilation derived from larger scale, county level soil survey maps. The soil landscape polygons generally consist of dominant and subdominant components. The dominant component is mapped and described in terms of a standard set of attributes considered important for plant growth, land management, and terrain sensitivity. The attributes include the mode of deposition, textural group of the parent material, soil development, surface form and slope class. Additional attributes further describe the dominant and subdominant components of each SLC polygon. The full array of attributes are located in a detailed computerized database available on diskette in ASCII Format. The source for much of the soil and land attributes contained in the SLC database is the National Soil Data Base in CanSIS. The SLC is increasingly being used to make decisions regarding the state of the soil resources and land use potential at regional, provincial and national scales. This study assesses the quality of the data contained in the detailed computerized legend and the map polygons, and develops a methodology to sample soils in the landscapes that best represent the SLC polygon. To assess the quality of the SLC database, five soil landscape polygons in the Regions of Niagara and Haldimand-Norfolk and five in Oxford County were selected. The SLC description of the soil and land attributes was compared to the actual soil series mapped in each of the polygons based on their respective soil survey information. To develop a sampling methodology a two phased approach was used. First the study developed a sampling methodology to determine the variability of selected soil attributes within the soil landscape. Based on these results a methodology to sample soil landscape polygons that can be applied over a large area of southern Ontario was developed. Agricultural fields were sampled in a systematic manner, independent of land use history, which resulted in soil samples and soil attributes that are representative of the majority of agricultural soils within each of the SLC polygons. The study determined that there was very limited correlation between the major soil attributes (soil development, mode of deposition, and textural class of the parent material) described by the SLC for the five SLC polygons studied in Niagara and Haldimand-Norfolk. There was, however, a good correlation between the SLC polygon descriptions and the Oxford County Soil Survey. The study also determined that there is significant variability between the pH and soil organic carbon values within the soil landscape and at different locations. Based on this information, a sampling methodology was developed and applied to Oxford County where each of the major soil landscape components within the SLC polygon was sampled. The study concluded that the accuracy of the SLC is highly variable. In some locations the SLC mapping and database correlates well to the detailed soil data. The use of the SLC mapping and data for large scale planning purposes would be appropriate in these locations. In other locations the correlation is poor and may limit the value of the SLC for large scale planning purposes. Recompilation of the SLC database and mapping may be required before interpretative decisions can be made depending on the level of accuracy required for the users objectives. Further, depending on the level of decision making it may be preferable to use the more detailed soil map information rather than relying on the SLC.