The lowermost Cretaceous (Berriasian) Sierra del Pozo Formation is divisible into metre-scale cycles that are bundled in a four-tiered hierarchy of cycles and cyclic sequences. At the smallest scale, sixth-order cycles, thought to be the product of precessionally forced sea-level fluctuations (c. 20 ka), average less than a metre in thickness, shallow upward and are bounded by surfaces where deeper facies abruptly overlie shallower facies. Bundles of sixth-order cycles, called fifth-order sequences, are recognized by two types of asymmetric patterns in facies distribution. First, more pronounced facies changes occur at sixth-order cycle boundaries lower in a sequence, whereas smaller facies changes occur at cycle boundaries higher in a sequence. Secondly, subtidal marine limestones (i.e. biomicrites and biosparites) are the dominant facies lower in a cyclic sequence, whereas more restricted or shallower water facies (i.e. mud-cracked microbial laminites, dolomites, shale and clay) are more predominant higher in a cyclic sequence. The bundling of sixth-order cycles is explained as a product of periodic change (100 ka) in the eccentricity of the earth's orbit. The degree of orbital eccentricity modulates the magnitude of precessional sea-level fluctuations, which in turn determine the relative facies contrasts at sixth-order cycle boundaries. Larger scale fourth- and third-order sequences are defined by similar patterns in facies contrast at successive sequence boundaries and by a change in predominant facies type from bottom to top. These patterns are explained as the product of 400 ka and 2 Ma periodic variations in the eccentricity of the earth's orbit. In summary, the strength of the precessional signal varies in consort with periodic changes in orbital eccentricity at three time scales producing a stacked hierarchy of cycles and sequences of cycles. Change in obliquity of the earth's axis may modulate the effects of the precessional signal and thereby modify patterns of preserved cyclic structure. This interpretation of cyclicity in the Sierra del Pozo section is markedly different from that of Jiménez de Cisneros & Vera (1993), who attributed all rock cycles to a single process (obliquity) even though the cycles they described varied in thickness from less than 1 m to more than 4 m. The cycles they described are shown here to be either precessional cycles or composite sets of precessional cycles (100 ka sequences). In this new hierarchical interpretation, ninety-six 100 ka and twenty-four 400 ka sequences are recognized, extending the time of deposition of the Sierra del Pozo section to more than 9·6 Ma.