In September-October 1994, the FLUPAC cruise was carried out in the western equatorial Pacific as a French contribution to the JGOFS programme. One leg of that cruise included a zonal transect along the equator, from 170 degrees E to 150 degrees W. Physical and nutrient data from that section presented unusual features: zonal currents in the upper layer were mostly found in opposition to local forcing, i.e. westerly winds and westward flow west of 180 degrees, easterly trade winds and weakly eastward flow elsewhere; east of 170 degrees W strong oscillations of the meridional flow (0.8 m s(-1) peak to peak) extended from the surface to the thermocline; a warm (higher than 30 degrees C), fresh (S<34.8) and salinity-stratified surface water mass spread, out from 170 degrees E to 172 degrees W, where it ended through a salinity front; east of that limit a deep homogeneous layer was found, but with relatively low vertical mixing; the thermocline deepened steadily from 170 degrees E eastward to a depth of 160 m at 160 degrees W; the boundary of nutrient enriched waters was displaced by about 2800 km eastward as compared with climatology, with 1 mu M surface NO3 concentration at 165 degrees W; a relative nutrient minimum (less than 3 mu M NO3) was found embedded in the enriched area east of that longitude. These observations are explained with the help of large-scale data from the TAO mooring array: 61 days of westerly winds in the western Pacific before the cruise had triggered a 0.4 m s(-1) current jet that advected surface waters to the east; a downwelling Kelvin wave propagated east of that wind patch, depressed the thermocline, and changed geostrophic surface current to eastward; tropical instability wave activity caused localized nutrient depletion by meridional advection; west of 180 degrees, propagation of an upwelling Rossby wave from the east may have contributed to westward flow. These results stress the importance of intraseasonal variability on physical and nutrient content data obtained during a cruise, and the usefulness of concurrent large-scale observations. (C) 1998 Elsevier Science Ltd. All rights reserved.