The upper waters of the Northeast Pacific Ocean contain very low frequency temperature fluctuations which have amplitudes of more than 1°C. Hydrographic measurements at 60°N, 149°W, and sea surface temperatures (SST) on a 5° grid over the Northeast Pacific are used to examine these variations. The very low frequency (VLF) 20- to 30-year fluctuation in SST found at and north of 55°N is not evident at lower latitudes. This VLF fluctuation exists throughout the water column on the shelf of the northern Gulf of Alaska. Contained within the hydrographie data on the shelf are responses to El Niño-Southern Oscillation (ENSO) forcing. However, ENSO responses are not evident in the SST data. The propagation characteristics of SST anomalies through the region are not consistent from one event to another. One SST anomaly moved eastward through the region over a 3-year period (1956–1959), whereas the 1983 anomaly appeared simultaneously throughout the Gulf of Alaska. Correlations with local wind stress and wind stress curl are very poor, implying that the temperature variability is not wind forced. The causes for these temperature anomalies are uncertain. Though climate changes due to increases in greenhouse gases might be amplified at high latitudes, heating due to global warming is discounted. Coupling of the temperature fluctuations with solar activity and lunar tides is possible especially at high latitudes and the periods of the temperatures, tides, and solar activity are well matched. In any case, the recent upper ocean warming is probably not a result of large-scale global change but is, rather, part of the VLF zonal signal. Below normal water and air temperatures should occur over the next 5–15 years. This VLF signal must be considered and understood before we will be able to measure the effects of high-latitude climate changes.