Theories concerning the formation of bedded chert traditionally have emphasized either depositional or diagenetic processes. Major and rare earth element data from Franciscan assemblage (Mesozoic) and Claremont Formation (Miocene) bedded chert sequences, along with physical observations such as the presence of rare and highly corroded radiolarians in shale interbeds, are most consistent with a dominantly diagenetic origin of chert-shale couplets and are incompatible with many depositional theories. Chemical distributions between Franciscan and Claremont bedded chert=shale closely match chemical fractionations recorded by Monterey Formation and Deep Sea Drilling Project-sampled cherts formed by diagenetic SiO (sub 2) dissolution, transport, and reprecipitation, suggesting that diagenetic migration of SiO (sub 2) from proto-shale to proto-chert is also largely responsible for chert-shale couplets. Identical Ce anomalies (Ce/Ce*) found in immediately adjacent chert-shale layers indicate that turbidites or other transport mechanisms are not responsible for the alternating beds. Neither the chemistry of the chert-shale couplet nor the overall stratigraphy of the sequences is consistent with couplet formation being caused by productivity fluctuations. Chemical mass balance calculations reconstructing the total bulk sediment composition suggest that modern siliceous sequences do not contain enough labile biogenic SiO (sub 2) to form entire stratigraphies of bedded chert.