Study 1: Aims: Compare the uniaxial MTI/CSA accelerometer and the biaxial Actiwatch accelerometer against direct observation of total physical activity and minute-by-minute physical activity in 3-4 year olds. Methods: MTI/CSA-7164 and Actiwatch accelerometers simultaneously measured activity during 35-45 minute sessions of structured play in 78, 3-4 year olds. Rank order correlations between accelerometry and direct observation were used to assess the ability of the accelerometers to assess total activity. Within-child minute-by-minute correlations were calculated between accelerometry output and direct observation. Results: For assessment of total activity MTI/CSA output was significantly positively correlated with direct observation (r 0.72, p<0.001), Actiwatch was not (r 0.16, p>0.05). Conclusion: The present study suggests that for epidemiological assessment of total physical activity in young children the MTI/CSA-7164 provides greater accuracy than the Actiwatch. Study 2: Aim: To cross-validate the 1100 counts/ minute cut-off for the assessment of sedentary behaviour in an independent sample of young children using the MTI/CSA accelerometer. Methods: A previously developed cut-off for MTI-CSA accelerometry output (validation study) in 30 healthy Scottish 3-4 year olds, was cross-validated against direct observation in an independent sample of (n = 52) healthy Scottish 3-4 year olds. Results: In the cross-validation study sensitivity was 83%: 438/528 inactive minutes were correctly classified. Specificity was 82%: 1251/1526 non-inactive minutes were correctly classified using this cut-off. Conclusion: Sedentary behaviour can be quantified objectively in young children using accelerometry.