OBJECTIVE To investigate the effect of a breakfast meal on bioelectrical impedance (BI). METHODS Three separate interventions. METHODS A university based study. METHODS Young, healthy volunteers recruited from staff. Twenty-nine subjects (11 men, 18 women), ten subjects (two men and eight women) and 13 subjects (2 men and 11 women) completed the first, second and third protocol, respectively. METHODS Total body BI (protocol 1) or both total body BI and segmental BI (namely arm, leg or torso BI); (protocol 3) was measured in the fasting state and for up to 5 h after the consumption of a breakfast meal containing 28% energy from fat. In the second protocol, total body BI was measured in the same way on two occasions after subjects consumed isocaloric meals containing either 28% energy or 4% energy from fat. RESULTS Consumption of a 2300 kJ meal was followed by a significant (P = 0.0002) decrease in BI (95% confidence intervals 12.5 and 35.3), a change which occurred 2 h after the meal and continued until 5 h, irrespective of meal fat content. The fall in total body BI was accounted for primarily by a fall in the BI of the limbs, with virtually no contribution from the torso. CONCLUSIONS To ensure consistency in the interpretation of BI for body composition analysis, it is important that measurements are made in the fasting state.