Increases in clinical complications associated with maternal obesity have generated interest in increasing physical activity (PA) and exercise levels as an intervention to improve pregnancy outcomes.The objective of this study was to examine the relationship between BMI categorisation and PA and exercise levels as pregnancy advances.This was an observational study in a large university maternity hospital. Women were recruited at their convenience before they left hospital after delivering a baby weighing 500 g or more. They completed a detailed customised physical activity and exercise questionnaire. BMI categorisation was based on the measurement of weight and height in early pregnancy.Of the 155 women recruited, 42.5 % (n = 66) were primigravidas and 10.3 % (n = 16) were smokers. Mean Body Mass Index (BMI) was 24.6 kg/m(2) and 14.2 % (n = 22) were obese, based on a BMI >29.9 kg/m(2). Overall, women decreased their exercise from an average 194 min (range 0-650 min) per week pre-pregnancy to 98 min antenatally (range 0-420 min) (p < 0.0001). Obese women exercised least pre-pregnancy and antenatally at 187.5 and 75 min per week, respectively, compared with 193.2 and 95.5 min per week in the normal BMI group and 239.3 and 106.7 min per week in the overweight group. The mean gestation at which all women reduced their activity levels was 29 weeks.We found that women decreased their PA and exercise levels significantly in the third trimester and, thus, in the absence of a medical contra-indication there is considerable scope for an exercise intervention to improve activity and exercise levels as pregnancy advances. However, an increase in PA levels in obese women needs further studies to determine whether it will improve the clinical outcomes for the woman and her offspring.