Background: Small-for-gestational age (SGA) newborns have modified calcium (Ca) transport across the placenta, delayed intestinal maturation and lower serum phosphate (P) levels compared to adequate-for-gestational-age infants (AGA), as well as decreased bone mineral content. In addition, higher concentrations of P were found in colostrum of mothers of SGA infant compared to the mothers of AGA. This led us to ask if there were differences in mineral content between SGA and AGA.Methods: A prospective control case study was done in order to analyse the influence of intrauterine growth retardation (IUGR)(type and intensity) on Ca and P supply through human milk, as well as, its impact on newborn serum levels. We studied 71 pairs of mothers-newborns, divided in two groups: group I(control)-41 pairs of mothers-term, AGA newborn and II(study)-30 mothers-term, SGA infants. This group was subdivided, according to type (symmetric/asymmetric) and intensity (<P3 and >P3-<P10) of the IUGR. Mothers milk, as well as blood samples were collected for Ca and P analysis using standard techniques (milk at the third, fifteenth and thirty days post-partum and blood of umbilical cord, fifteenth and thirty days post-partum). Results: There were no differences in the blood and human milk Ca and P content, between the two groups (analysis of variance). In the study group, the milk(p=0,001) and serum (p=0,002) P levels as well as the milk Ca (p=0,002) levels increased during the first month of life (analysis of variance). There was a trend for correlation between milk and serum Ca and P levels for both groups, but there was a statistical significance (Pearson's correlation/means regression) only for P at the 15th day of life(p=0.04). Regarding type and intensity of IUGR: milk Ca levels were not influenced by type or intensity; milk P levels increased in the groups IUGR symmetric and P<3; serum Ca levels increased in the IUGR >P3-<P10 and serum P levels increased in the type symmetric. Conclusions: These findings suggest a tendency of adaptation of human milk to the growth needs of SGA newborns, mainly those more affected (intensity and timing), with respect to P content.