a b s t r a c tThe GreenFeed (GF) system (C-Lock Inc., Rapid City, USA) is used to estimate total daily methane emissions of individual cattle using short-term measurements obtained over several days. Our objective was to compare measurements of methane emission by growing cattle obtained using the GF system with measurements using respiration chambers (RC) or sulphur hexafluoride tracer (SF 6 ). It was hypothesised that estimates of methane emission for individual animals and treatments would be similar for GF compared to RC or SF 6 techniques. In experiment 1, maize or grass silage-based diets were fed to four growing Holstein heifers, whilst for experiment 2, four different heifers were fed four haylage treatments. Both experiments were a 4 × 4 Latin square design with 33 day periods. GreenFeed measurements of methane emission were obtained over 7 days (days 22-28) and compared to subsequent RC measurements over 4 days (days 29-33). For experiment 3, 12 growing heifers rotationally grazed three swards for 26 days, with simultaneous GF and SF 6 measurements over two 4 day measurement periods (days 15-19 and days 22-26). Overall methane emissions (g/day and g/kg dry matter intake [DMI]) measured using GF in experiments 1 (198 and 26.6, respectively) and 2 (208 and 27.8, respectively) were similar to averages obtained using RC (218 and 28.3, respectively for experiment 1; and 209 and 27.7, respectively, for experiment 2); but there was poor concordance between the two methods (0.1043 for experiments 1 and 2 combined). Overall, methane emissions measured using SF 6 were higher (P<0.001) than GF during grazing (186 vs. 164 g/day), but there was significant (P<0.01) concordance between the two methods (0.6017). There were fewer methane measurements by GF under grazing conditions in experiment 3 (1.60/day) compared to indoor measurements in experiments 1 (2.11/day) and 2 (2.34/day). Significant treatment effects on methane emission measured using RC and SF 6 were not evident for GF measurements, and the ranking for treatments and individual animals differed using the GF system. We conclude that under our conditions of use the GF system was unable to detect significant treatment and individual animal differences in methane emissions that were Abbreviations: CI, confidence interval; DM(I), dry matter (intake); GF, GreenFeed; LW, live weight; LWG, LW gain; NDIR, non-dispersive infrared; RC, respiration chambers; RFID, radio frequency identification; SF6, sulphur hexafluoride tracer. identified using both RC and SF 6 techniques, in part due to limited numbers and timing of measurements obtained. Our data suggest that successful use of the GF system is reliant on the number and timing of measurements obtained relative to diurnal patterns of methane emission.