The suitability of scales for ageing Micropterus salmoides was determined by comparing the precision of growth zone counts on scales with those obtained from sectioned sagittal otoliths from a sample of 496 fish collected from Wriggleswade and Mankazana Impoundments in the Eastern Cape, South Africa. Otoliths (1.4% rejected) were more readable than scales (41.7% and 7.5% rejected) for Wriggleswade and Mankazana Impoundments respectively. Otolith readings were more precise (average percentage error (APE) = 13.6%; coefficient of variation (CV) = 15.8%) than scales (APE = 18.0%; CV = 21.9%) for the total sample. Growth zone counts between structures were not symmetrically distributed around the agreed ages (otolith age = scale age) of the fish from Wriggleswade Impoundment (Bowker's test H²: χ²= 136, d.f. = 27, P < 0.05) and graphical comparison showed that scales tended to underestimate age, particularly for fish older than five years. Scales are therefore not suitable structures for ageing South African M. salmoides which may attain ages of up to 13 years.