Abstract Till is a favoured sample medium for locating mineral deposits in glaciated shield terrains of Canada and Fennoscandia because it best reflects the primary composition of the bedrock source area. In the sampling phase, an important and costly component of till surveys, sample density, sample depth and sample method must be chosen according to the needs of the exploration program. Surface till sampling methods in forested areas differ from those used in permafrost terrain. However, in both areas, concentrations of labile ore minerals and their products of decomposition can be detected in the fine fraction (< 2 mm) of weakly oxidized till. In thin drift-covered areas, till samples are collected by hand excavation or trenching at < 5 m depth. In areas of thicker drift, more expensive methods such as reverse circulation rotary drills, rotasonic drills and portable drills are used to collect till samples at depth and to determine lateral and vertical variations in till geochemistry. Laboratory methods are an essential part of till geochemical surveys. The choices of the size fraction and analytical methods are determined by the nature and composition of the expected bedrock target, and by costs.