The trade winds over the central Pacific are observed to weaken several months after the appearance of anomalously warm surface waters in the eastern equatorial Pacific Ocean. The following results obtained with a numerical model indicate how this relaxation of the winds affect the later stages of El Niño. A weakening of the westward trade winds causes a zonal redistribution of heat in the equatorial oceans and a warming of the eastern part of the basin. The warming depends on the zonal extent of the region over which the winds relax, and on the length of time T for. which the winds relax. As T increases the warming in the east increases until it asymtotes to a maximum value when T exceeds the adjustment time of the basin (which is ∼400 days in the case of the Pacific Ocean). Maximum heating is associated with a permanent weakening of the winds, unless the winds reverse direction and become eastward. Even weak eastward winds for a short period can cause disproportionately large temperature increa... Abstract The trade winds over the central Pacific are observed to weaken several months after the appearance of anomalously warm surface waters in the eastern equatorial Pacific Ocean. The following results obtained with a numerical model indicate how this relaxation of the winds affect the later stages of El Niño. A weakening of the westward trade winds causes a zonal redistribution of heat in the equatorial oceans and a warming of the eastern part of the basin. The warming depends on the zonal extent of the region over which the winds relax, and on the length of time T for. which the winds relax. As T increases the warming in the east increases until it asymtotes to a maximum value when T exceeds the adjustment time of the basin (which is ∼400 days in the case of the Pacific Ocean). Maximum heating is associated with a permanent weakening of the winds, unless the winds reverse direction and become eastward. Even weak eastward winds for a short period can cause disproportionately large temperature increa...