It has often been proposed that the well-known standard activation enthalpyentropy linear plots usually found in kinetic studies for many series of homologous reactions are mere artefacts generated by the experimental errors committed in the determination of the rate constants. Here it is shown that the experimental errors can explain the existence of a standard activation enthalpy-entropy linear correlation only when the temperature obtained from the slope of that plot (compensation temperature, Tc) is lower than or equal to the mean experimental temperature used in that study (Tm), that is c m T T  . However, when Tc > Tm it is necessary to accept the existence of a real, physicochemically meaningful correlation between the values of o H and o S  for each member of the series of reactions. Four different explanations (thermodynamic interpretation, solvent-cage effect, formation of a binding intermediate and a Hinshelwood-like model) for real enthalpy-entropy compensation effects are analysed in some detail.