It is suggested that the much publicized discrepancy between observed surface global mean temperature and global mean atmospheric temperature from 1979 to the present may be due to the fact that the atmosphere underwent a jump in temperature in 1976 (before the satellite temperature series began), and that the surface response was delayed for about a decade due to the ocean heat capacity. The ocean delay depends on both climate sensitivity and vertical heat transport within the ocean. It is shown that the observed delay is best simulated when sensitivity to doubling of CO2 is less than about 1C.