Two relative paleointensity records of the Jaramillo subchron were obtained during Ocean Drilling Program Leg 167 at Sites 1010 and 1021 on the California margin. The two sites are ~1300 km apart and differ in sediment composition and deposition rate. The demagnetization of the natural remanent magnetizations (NRM) of discrete samples revealed a vertical drilling overprint at both sites which could be removed by alternating field demagnetization at 20 mT for the sediments from Site 1010. This was confirmed by a positive reversal test on the characteristic remanent magnetization of the discrete samples across the Jaramillo subchron. Rock magnetic criteria for determining relative paleointensities are fulfilled for the samples from Site 1010. Because of variations in the rockmagnetic properties, two short intervals had to be excluded from relative paleointensity determinations at Site 1021. Normalization of the NRM at several demagnetization steps with various magnetic parameters yielded similar results only for Site 1010, and contributions from viscous remanent magnetization could not be detected in the sediments from this site using the pseudo-Thellier method of Tauxe et al. <1995>. The relative paleointensity data of Site 1021 show variations of up to 50% between normalizations with different magnetic parameters and therefore had to be excluded from further analysis. Several fine-scale intensity features are detectable in the record of Hole 1010C, displaying a maximum in relative paleointensity shortly above the lower Jaramillo transition, a prominent local minimum in the middle of the subchron, and a gradual decrease toward the upper Jaramillo reversal. A comparison of this record from the California margin with several published records of relative paleointensity across the Jaramillo subchron with different sedimentation rates and sediment composition shows reasonable agreement between the paleointensity patterns but differences in amplitudes. This suggests an influence of postdepositional magnetization processes which were not removed during normalization. ¿ 1999 American Geophysical Union