A carbon budget for the Nordic Seas and Arctic Ocean is compiled by combining volume flows from the literature and carbon chemistry data, mainly from the Transient Tracers in the Ocean/National Academy of Sciences (TTONAS) 1981 expedition. The net advective export of dissolved inorganic carbon from the Nordic Seas and Arctic Ocean is found to be approximately 0.23*106 moles C s-1 (0.09 Pg C yr-1). Including a tentative estimate of the increase of the carbon inventory in the region, which is due to the supply of anthropogenic CO2, gives an estimate of the net uptake of atmospheric CO2 amounting to 0.11 Pg C yr-1. If the ice-free area is assumed to be 3.1*1012 m2, this corresponds to a net uptake of 3 moles C m-2 yr-1 over the ice-free area. Input of carbon to the North Atlantic via the Arctic Ocean and Nordic Seas comprises transport of carbon from the Pacific through the Bering Strait, riverine carbon input, and net uptake of atmospheric CO2. On the basis of data from the literature the input of carbon from the Pacific is found to be approximately 0.63 Pg C yr-1, giving a total flux of carbon to the North Atlantic from the Nordic Seas and Arctic Ocean amounting to approximately 0.8 Pg C yr-1. A prominent feature of the transports is the large gross transport across the Greenland-Scotland ridge and the concentration difference between the inflowing North Atlantic water and the outflowing deep water, that is, the northern limb of the Atlantic thermohaline circulation. The gross carbon transport associated with deep and intermediate water overflow across the Greenland-Scotland ridge is estimated to 4.5 Pg C yr-1, which is at least 5% of the global gross exchange between surface and deep ocean. The role of the region in the global carbon cycle may therefore be summarized as a transit area for carbon from the Pacific and from river runoff, uptake of atmospheric CO2, and a substantial transformation of surface water to deep water. A large fraction of the regional storage of anthropogenic CO2 appears to be due to advection rather than enhanced regional air-sea uptake. ¿ American Geophysical Union 1996