Objective: To assess the effects of birth size, growth and feeding in infancy on serum lipids in 12-month-old infants. Design: A longitudinal observation study on infants' consumption and growth. Food and growth records were made every month. At 6, 9 and 12 months, food records were weighed to calculate intake. Serum total cholesterol (TC), high-density lipoprotein-cholesterol (HDL), low-density lipoprotein (LDL) -cholesterol and triglyceride concentrations were analysed at 12 months. Settings: Birth and growth information was gathered from maternity wards and healthcare centres in Iceland and food consumption data at home. Subjects: Randomly selected newborns (n ¼ 180) according to the mother's domicile and 77% (n ¼ 138) participated, of them 75% (n ¼ 103), came in for blood sampling. Results: Among boys, a 1 kg higher birth weight resulted in a 0.79 mmol/l higher TC (P ¼ 0.005), but nonsignificant after adjustment for growth. Duration of breastfeeding was related to LDL-cholesterol (B ¼ 0.06 70.02, P ¼ 0.020, adj. R 2 ¼ 0.039), adjusting for gender. Independent of size at birth and breastfeeding, increase in length from 6 to 12 months and in weight from birth to 12 months were negatively related to TC (B ¼ À0.45570.156, P ¼ 0.008 and B ¼ À1.08670.474, P ¼ 0.032, respecitvely) in boys. PUFA was the strongest nutrient variable predicting TC (B ¼ 0.332, adj. R 2 ¼ 0.24, P40.001). Cod liver oil consumption increased both TC and LDL-cholesterol in girls (B ¼ 0.14170.051, P ¼ 0.008 and B ¼ 0.11270.047, P ¼ 0.021, respectively). Conclusion: Slower growth of high birth weight infants and breastfeeding contributes to higher TC concentration at the age of 12 months. Nutrient intake in infancy also affects lipid profile. The effect of birth weight, growth and nutrient intake in infancy on lipid profile is different for boys and girls.