AbstractThe paper reports the first research on karst solution processes and rates in New Zealand. The study area is an IHD representative basin in the northwest corner of the South Island in a mountain range consisting principally of Ordovician marble. The climate is sunny and warm (17°C) in summer and wet and cool (7°C) in winter. Average precipitation is 2,158 mm of which 525 mm evapotranspires, yielding a discharge of 51–75 l/s/km2in the river basin studied. Almost half of the catchment of 45.1 km2consists of karst which occurs mainly as a doline covered plateau at 600–900 m within which most drainage is subterranean. Water tracing is with fluorescein defined drainage patterns. Marble solution was established by estimating inputs, throughputs and outputs of water and dissolved calcium and magnesium in both autogenic and allogenic karst drainage systems. Particular attention was paid to estimating errors. Water samples for chemical analysis were taken irregularly for approximately one year, and a rating curve relating chemical load to discharge was established. The best estimate of solution loss from the basin yields a mean rate of 100 ±M24 m3/km2/a. Of this 80 per cent is derived from solution of marble by autogenic waters, mostly in the top 10–30 m of the marble outcrop. The remaining 20 per cent is accomplished by allogenic stream solution. Approximately 9.9 per cent of the dissolved calcium and magnesium load leaving the basin originates from non‐karst rocks and 4.6 per cent is initially introduced by rainfall.River flows that are exceeded only 5 per cent of the time transport approximately 44 per cent of the annual dissolved load, while mean to low flows that occur for 75 per cent of the time transport 35 per cent of the annual solute load. This confirms the importance of low frequency‐high magnitude events, but indicates also that in corrosion systems high frequency events of moderate to low magnitude can also accomplish sign