Paleomagnetic measurements on sedimentary rocks have been used in the past to reconstruct plate motions, but recent work has shown there to be a systematic shallowing of paleomagnetic inclinations in these rocks when compared with volcanic rocks. Thus a possibly serious error exists in paleolatitude calculation conducted to date. This study will systematically test the expected latitudinal dependence of inclination shallowing using lithified Atlantic DSDP sediments. Detailed study of the process will be conducted on a small group of preselected samples. The results will help resolve discrepancies in the existing database and will thereby permit accurate reconstruction of past tectonic motions.