The aims of this work were to assess timing of complementary feeding in infants and to precise the underlying factors that may cause inappropriate complementary feeding. Patients and methods. – A cohort of 167 newborns, born in the same maternity ward during a 6 week-period, was prospectively analyzed. Only healthy neonates were included in the study. A phone questionnaire was filled at 4 and 6 months of age to evaluate modality of complementary feeding. Multivariate analysis (segmentation tree, analyse by multiple correspondence) was used to study factors associated with inappropriate diversification. Results. – Out of the 167 neonates included in the study, 132 mothers could be contacted at 4 months and 116 at 6 months of age. Sixty-seven per cent of mothers started breast feeding at birth. Among these, 33% still breastfed -at least partially- at 4 months and 17% at 6 months. Fifty-two percent of mothers started complementary feeding before 4 months, and 24% of infants received gluten at 4 months of age. Multi-gravida mothers, mothers aged more than 35 years old and mothers who gave infant or follow-up formulae before 4 months, started complementary feeding significantly earlier ( P < 0.05). Infants who were formula fed received more frequently complementary feeding before the age of 4 months than breast fed infants (57% vs 33%, P < 0.05). Conclusion. – Our study showed that half of infants were introduced solid food too early and allowed to identify a population at risk that could benefit from nutritional intervention programs.