The aims of the Atlantic Meridional Transect (AMT) programme [www.amt-uk.org] are to quantify the nature and causes of ecological and biogeochemical variability in the planktonic ecosystems of the Atlantic Ocean, and to assess the effects of this variability on biological carbon cycling and air–sea exchange of radiatively active gases and aerosols. Marine and atmospheric data have been collected twice a year along a 13,500 km transect in the Atlantic Ocean between 50°N and 52°S since 1995. The cruise track enables biogeochemical measurements to be made within the poorly studied North and South Atlantic oligotrophic gyres as well as within equatorial and coastal upwelling regions. The range of ecosystems sampled during AMT has facilitated the calibration and validation of newly developed optical, microbiological, molecular and analytical techniques and provided a testbed for comparative ecology and the development of atmospheric and oceanographic models. This paper describes the rationale and methodology of the programme. Upper-ocean measurements of density, nitrate and chlorophyll a (Chl a) are presented to illustrate seasonal, inter-annual and decadal variability in hydrography. Seasonal distributions of dissolved nitrous oxide (N2O) and methane (CH4) are used to derive estimates of the sea–air flux of these gases in the South Atlantic Gyre. Observations made during AMT and published since 2000 are reviewed, and the key findings are highlighted. The extent to which the programme aims have been achieved is discussed and improvements for the future suggested.