In the course of the ICOS (Integrated Carbon Observation System) Demonstration Experiment a feasibility study on the usefulness of a travelling comparison instrument (TCI) was conducted in order to evaluate continuous atmospheric CO<sub>2</sub> and CH<sub>4</sub> measurements at two European stations. The aim of the TCI is to independently measure ambient air in parallel to the standard station instrumentation, thus providing a comprehensive comparison that includes the sample intake system, the instrument itself as well as its calibration and data evaluation. Observed differences between the TCI and a gas chromatographic system, which acted as a reference for the TCI, were −0.02 ± 0.08 μmol mol<sup>−1</sup> for CO<sub>2</sub> and −0.3 ± 2.3 nmol mol<sup>−1</sup> for CH<sub>4</sub>. Over a period of two weeks each, the continuous CO<sub>2</sub> and CH<sub>4</sub> measurements at two ICOS field stations, Cabauw (CBW), the Netherlands and Houdelaincourt (Observatoire Pérenne de l'Environnement, OPE), France, were compared to co-located TCI measurements. At Cabauw mean differences of 0.21 ± 0.06 μmol mol<sup>−1</sup> for CO<sub>2</sub> and 0.41 ± 0.50 nmol mol<sup>−1</sup> for CH<sub>4</sub> were found. For OPE the mean differences were 0.13 ± 0.07 μmol mol<sup>−1</sup> for CO<sub>2</sub> and 0.44 ± 0.36 nmol mol<sup>−1</sup> for CH<sub>4</sub>. Offsets arising from differences in the working standard calibrations or leakages/contaminations in the drying systems are too small to explain the observed differences. Hence the most likely causes of these observed differences are leakages or contaminations in the intake lines and/or their flushing pumps. For the Cabauw instrument an additional error contribution originates from insufficient flushing of standard gases. Although the TCI is an extensive quality control approach it cannot replace other quality control systems. Thus, a comprehensive quality management strategy for atmospheric monitoring networks is proposed as well.