Aim: Little is known of minor neurological dysfunction (MND) in infancy. This study aimed to evaluate the inter-assessor reliability of the assessment of MND with the Touwen Infant Neurological Examination (TINE) and the construct and predictive validity of MND in infancy. Method: Inter-assessor agreement was determined in a sample of 40 infants (24 males, 16 females) aged 3 to 12 months (25 born at term: gestational age 37-41wks, median 39; and 15 born preterm, gestational age 24-35wks, median 32). Thirty typically developing term infants (18 males, 12 females; gestational age 37-42wks, median 40) and 59 preterm infants (34 males, 25 females) born at less than 35 weeks' gestation (gestational age 25-34wks, median 29) participated in the validity study. They were neurologically assessed with the TINE at the corrected ages of 4, 6, 10, and 12 months and with the Hempel assessment at 18 months. Results: The findings indicated that MND can be assessed reliably (inter-assessor agreement: kappa=0.83). MND during infancy was related to prenatal, perinatal, and social factors, and in particular to preterm birth. Neurological condition during infancy was prone to change, but was related to neurological condition at 18 months at all ages tested. Interpretation: We conclude that MND can be determined reliably in infancy. Important considerations in the construct of MND in infancy are its relation to prenatal and perinatal factors, its limited stability, and its moderate predictive value. (Contains 5 tables.)