During 2009, the Met Office implemented a new surface observing system, the Meteorological Monitoring System (MMS). We report the results of a comparison of screen temperature as measured by MMS, and the Semi-Automatic Meteorological Observing System (SAMOS), the most widely used of the systems which MMS replaces. In each system, temperature was measured at 1-min resolution using 100 Ω platinum resistance thermometers. Comparison of MMS and SAMOS temperatures in separate Stevenson screens revealed acceptably small mean temperature differences (0.05 °C); however, MMS temperatures tended to be 0.3–0.5 °C higher than SAMOS temperatures during conditions of strong solar radiation, with extreme differences of up to 0.8 °C. Such differences were largely absent when MMS and SAMOS temperatures were measured from within the same screen, suggesting that differences in the immediate environment of each screen may have been responsible. However, even within the same screen, a diurnal cycle in the (SAMOS–MMS) temperature differences of amplitude (peak-to-peak) ∼0.04–0.06 °C was observed, which was likely due to the effects of low-angle solar radiation. Additional testing using a third, independent temperature unit revealed a slight temperature dependency in the SAMOS measurements; extrapolation of the results suggests differences of 0.2–0.3 °C between SAMOS and MMS temperatures at ambient temperatures >30 °C (MMS warmer). Overall, however, the differences between SAMOS and MMS temperatures were small, being well within World Meteorological Organization (WMO) accuracy limits for mean and extreme temperatures. The results therefore demonstrate that any step change in climatological temperature records associated with the introduction of MMS may be considered negligible.