The oxidation of 2-butyl radicals (and to a lesser extent 1-butyl radicals) has been studied over the temperature range of 298-735 K. The reaction of Cl atoms (formed by 360 nm irradiation of Cl-2) with n-butane generated the 2-butyl radicals in mixtures of n-C4H10, O-2, and Cl-2 at temperatures below 600 K. Above 600 K, 2-butyl radicals were produced by thermal combustion reactions in the absence of chlorine. The yields of the products were measured by gas chromatography using a flame ionization detector. Major products quantified include acetone, acetic acid, acetaldehyde, butanone, 2-butanol, butanal, 1- and 2- chlorobutane, 1-butene, trans-2-butene, and cis-2-butene. At 298 K, the major oxygenated products are those expected from bimolecular reactions of 2-butylperoxy radicals (butanone, 2-butanol, and acetaldehyde). As the temperature rises to 390 K, the butanone decreases while acetaldehyde increases because of the increased rate of 2-butoxy radical decomposition. Acetone and acetic acid first appear in significant yield near 400 K, and these species rise slowly at first and then sharply, peaking near 525 K at yields of similar to 25 and similar to 20 mol %, respectively. In the same temperature range (400-525 K), butanone, acetaldehyde, and 2-butanol decrease rapidly. This suggests that acetone and acetic acid may be formed by previously unknown reaction channels of the 2-butylperoxy radical, which are in competition with those that lead to butanone, acetaldehyde, and 2-butanol. Above 570 K, the yields of acetone and acetic acid fall rapidly as the yields of the butenes rise. Experiments varying the Cl atom density, which in turn controls the entire radical pool density, were performed in the temperature range of 410-440 K. Decreasing the Cl atom density increased the yields of acetone and acetic acid while the yields of butanone, acetaldehyde, and 2-butanol decreased. This is consistent with the formation of acetone and acetic acid by unimolecular decomposition channels of the 2-butylperoxy radical, which are in competition with the bimolecular channels that form butanone, acetaldehyde, and 2-butanol. Such unimolecular decomposition channels would be unlikely to proceed through conventional transition states because those states would be very constrained. Therefore, the possibility that these decomposition channels proceed via roaming should be considered. In addition, we investigated and were unable to fit our data trends by a simplified ketohydroperoxide mechanism.