Different calculation methods, based on needle geometry, for estimating both projected area (PLA) and total surface area (TLA) of foliage in Norway spruce [Picea abies (L.) Karst.] were compared. Seventy-eight shoots of four age classes were sampled from both the basal and top thirds of crowns. Three dimensions (the length, minor and major diameters) of needles were taken, and the needle shape was approximated to a parallelepiped or ellipsoid. There was a perfect coincidence of the measured and estimated values of PLA calculated as the width of the needle projection multiplied by needle length, and corrected for needle taper (method III), or when the needle projection was treated as a rectangle joined with half-ellipses at both ends (method IV). The most reliable estimations of TLA resulted from treating the needle sides as faces of the parallelepiped tapering at their ends in the form of half-ellipses. The ratio of TLA to PLA ranged from 2.2 to 4.0 depending on the needle morphology. Needle minor diameter (anatomical width; D 1 ) was found to be a better morphological index of the spruce foliage than needle flatness, i.e. the ratio of major to minor diameter. Expressing the factor for converting PLA to TLA as a function of D 1 considerably improved the precision of the estimates. Close relationships were established between specific leaf area, expressed on both a projected area (SLA P ) and total surface area basis (SLA T ), needle dry weight (R 2 was 0.799 and 0.852, respectively) and minor diameter of needles (R 2 was 0.701 and 0.554, respectively).