Piovesan & Adams (2000) criticized the EUROFLUX network of carbon flux measurements based on the assumption that the results were ‘surprising’, as they appeared to contradict the generally accepted relationship that rates of soil and plant respiration increase with increasing temperature. The EUROFLUX studies showed an overall opposite trend, with higher rates of ecosystem respiration at the cooler, northern latitudes. In this paper the concept of ‘surprising’ results in ecology is reviewed and the original claims of Piovesan & Adams reassessed. It is concluded that results from the EUROFLUX network do not contradict the established respiration/temperature relationship for at least two reasons. First, ecosystem respiration is the net outcome of a wide range of underlying processes and a simple respiration/temperature relationship may not be observed if other nontemperature dependent processes dominate. Second, relatively higher temperatures at the northern hemisphere sites during the period over which the EUROFLUX observations were made are consistent with the hypothesis that the measured fluxes reflect a non‐equilibrium, transient loss of carbon. It is further concluded that the potential importance of inter‐year variability for determining short‐term patterns of carbon flux may provide the key to reconciling the temperature/respiration relationship highlighted by Piovesan & Adams, and the latitudinal respiration trends evident in the EUROFLUX results.