Phosphorus (P) and calcium (Ca) are stored in bone, a tissue that can be mobilized when mineral requirements are high, as during lactation. While this flow has been quantified in primiparous sows, it has not been assessed in multiparous sows. The objectives of this study are to validate this flux in primiparous sows and to quantify it in multiparous sows based on phosphocalcic intake and to study bone mineralization in suckling piglets. Bone mineral content (BMC; g) of the whole body of 24 primiparous and 24 multiparous sows (parity 3 or 4) and their piglets was measured by X-ray absorptiometry at farrowing (d2) and weaning (d26). Four lactation feeds were formulated to achieve different nutritional intakes: Lact100 (9.0 g Ca; 3.0 g digestible P/kg), Lact50 (4.5 g Ca; 1.5 g digestible P/kg), and these same treatments supplemented with 750 FTU/kg exogenous phytase (Lact100+FTU and Lact50+FTU). Sow BMC were influenced by dietary treatments (P < 0.05). At d26, the BMC of Lact50 sows was lower (P < 0.05) than that of Lact100 and Lact100+FTU sows, which had similar BMC and Lact50+FTU was intermediated. Therefore, only Lact50 sows demineralized during lactation due to a lack of Ca and P. Primiparous sows had a lower BMC than multiparous sows, but the kinetics of BMC between farrowing and weaning were independent of parity. Dietary treatments had no impact on piglet mineral concentration (g BMC/kg PV). The results of this study confirm the ability of sows to mobilize their bone reserves during lactation. The intensity of this demineralization is mainly determined by the dietary P content, independently of the parity of the sows. This ability to draw on bone mineral reserves allows to maintain P and Ca concentrations in milk, even when the intake of digestible P and Ca is reduced to 50% of the requirements. These data will be used to develop a model for predicting the use and retention of P and Ca in lactating sows, with the aim of defining the requirement more precisely.