We have previously demonstrated that fetal hypoxia-ischemia decreases the total antioxidant activity in premature rabbit brain. Improvements in an assay for the measurement of total antioxidant activity in biological samples are reported. The assay utilizes the monitoring of formation of the ABTS+ radical by measuring the absorbance at 734 nm. The previous method of measuring end-point absorbance after a fixed time is unsuitable for biological samples since the pattern of absorbance change after addition of biological samples is markedly different from that of pure chemical solutions of antioxidants. The area under the curve for absorbance of the ABTS+ radical formation over time was a more accurate measurement of total antioxidant activity in brain and plasma than end-point absorbance at a fixed time or lag phase determination of the onset of radical formation. Furthermore, another modification of the assay enables the delineation of various components of antioxidant activity in biological samples. A post-addition modification distinguishes between ABTS+ scavenging activity and inhibition of radical formation. Water-soluble antioxidants like ascorbic acid and uric acid constitute scavenging activity observed in biological samples. Units of scavenging activity can be expressed in Trolox equivalents following comparison to a Trolox standard curve. It is noteworthy that proteins constitute the bulk of radical-formation inhibitors in brain and plasma. The amount of inhibitors of radical formation can be expressed as a percentage of a solution containing no antioxidants. Interestingly, the post-addition modification enabled the detection of a slow reacting scavenger in plasma, which was not detected in brain homogenates. This slow-reacting scavenger is most probably a protein of size >30,000 Mr, that is not albumin, superoxide dismutase or catalase. Addition of the strong oxidant, peroxynitrite, to plasma decreases the scavenging activity but does not affect the slow-reacting component. The post-addition modification gets optimal results with addition of 20 μl sample. Addition of greater quantities of sample causes a dilution error that needs to be corrected for. The delineation of various components of total antioxidant activity may be useful in the development of different antioxidant treatments in clinical situations of oxidative stress.