On the basis of experiments presented in Part I of this series, most of the published thermobarometers relevant to four-phase peridotites are tested here for their ability to reproduce experimental conditions. They were rejected if any systematic discrepancy in either pressure or temperature was discernible. This test cautions against the use of all published versions of thermometers based on the compositions of coexisting ortho- and clinopyroxenes and the use of existing barometers based on the Al content of orthopyroxene coexisting with garnet. Therefore, we formulated new versions of the two-pyroxene thermometer and the Al-in-opx barmometer: [GRAPHICS] with K(D)* = (1-Ca*)cpx/(1-Ca*)opx, Ca* = CaM2/(1-Na(M2)) and X(Fe)px = Fe/(Fe + Mg); T(BKN) is in degrees Kelvin and P is in kilobars. Our new barometer is of the form [GRAPHICS] (C1-C3) and site occupancies are given in the text. Temperatures may also be calculated from the Ca content of opx alone: [GRAPHICS] This thermometer can be applied both to the CMS and the natural system experiments, which may indicate that Fe and Na have counter-balancing effects on the Ca content of opx. The partitioning of Na between opx and cpx can also serve as a useful thermometer, and was calibrated from natural rock data: [GRAPHICS] where T is in degrees Kelvin, P is in kilobars, and D(Na) = Na(opx)/Na(cpx). The following three published thermobarometers based on further exchange reactions are capable of reproducing experimental conditions: (a) exchange of Ca between olivine and clinopyroxene as a barometer (P(KB)), (b) exchange of Fe and Mg between garnet and clinopyroxene as a thermometer (T(Krogh)), (c) exchange of Fe and Mg between garnet and olivine as a thermometer (T(O'Neill)).