The concentration of ketone bodies in plasma and of carnitine in various maternal, fetal, and neonatal tissues was examined during the developmental period in rats. Plasma ketone levels were low in the fetus, increased 10-fold during the first 24 h postpartom, and thereafter gradually declined such that normal values were found at the end of the suckling period. An almost identical profile was observed for liver carnitine concentrations in the baby rats. The converse was true for heart tissue, the carnitine content of which was low at birth and steadily increased to adult levels with the time of suckling. The primary source of carnitine in neonatal tissues, at least during the first 2 to 3 days postpartum, was shown to be the mother rat whose liver and milk carnitine content was very high at this time and fell as nursing continued. Experiments in which the fate of [14C]butyrobetaine, the immediate precursor of carnitine, was followed after injection into nursing mother rats indicated movement of carnitine from maternal liver leads to maternal plasma leads to milk leads to neonatal tissues. The above findings support the view expressed earlier that one prerequisite for the development of a high ketogenic profile in liver may be an elevation in the tissue carnitine concentration. Additional factors, however, are clearly involved as evidenced by the observation that in the fed state perfused livers from nursing mother rats synthesized ketone bodies from oleic acid at low rates compared with those seen after a 24 h fast, despite the fact that tissue carnitine levels were equally elevated in both groups. This paradox is likely related to the fact that in the fed state such livers also contained large quantities of glycogen, depletion of which through fasting was accompanied by marked acceleration of ketogenesis from oleate. The data indicate, therefore, that maximal ketogenic capacity of the liver requires for its induction an increase in carnitine coupled with a decrease in glycogen content of the tissue.