Gradational development of domainal slaty cleavage, its origin and relation to chlorite porphyroblasts in the Martinsburg Formation, eastern Pennsylvania

Abstract

Slaty cleavage frequently consists of narrow folia or domains of phyllosilicates alternating with microlithons richer in quartz. Within cleavage domains the {001} planes of phyllosilicates are usually parallel or subparallel to the domain boundaries, but they may be at a small but persistent angle. In microlithons the phyllosilicates are commonly not parallel to the cleavage domains. Domainal cleavage may grade into pervasive slaty cleavage with few or no microlithons and a parallel orientation of all the phyllosilicates, or may grade within distances as short as a few millimeters into pelite with no cleavage and with the phyllosilicates either with no preferred orientation or a weak to strong orientation parallel to bedding. Study of cleavage specimens, mainly from the Martinsburg Formation of Pennsylvania and New Jersey, indicates that cleavage development is related to strain variation and to volume loss, at least as measured by deformed quartz-calcite veins in the XZ plane of a fold. Mass transfer of the more soluble constituents under conditions of low-grade metamorphism (chlorite zone) and under the influence of deviatoric stress leads to the development of “solution ways” initiated at heterogeneities in the fabric which result in solution cleavage that is extended by terminal and lateral propagation. Stress gradients result in gradational development of solution cleavage, with minimal development in “strain shadows.” The “solution ways” develop into cleavage domains by accumulation of inert materials, particularly carbonaceous grains, iron ores such as framboidal pyrite, and phyllosilicates. However, the preferred orientation of phyllosilicates is mainly the result of crystallization of new grains in the domains because of the breakdown of detrital or diagenetic phyllosilicates and of other minerals such as feldspar or epidote under low-grade metamorphic conditions. This interpretation is supported by lack of bent or deformed phyllosilicates in and along the boundaries of cleavage domains. (Larger detrital phyllosilicates, with {001} at a high angle to cleavage, commonly show deformation.) Further support is provided by the occurrence of chlorite grains with their {001} planes approximately parallel to the cleavage in “strain shadows” on competent grains such as framboidal pyrite and by the shape change of diagenetic chlorite porphyroblasts as traced from zones with no domainal cleavage into zones with well-developed domainal cleavage. Mechanical rotation appears to be of minimal importance in the development of cleavage domains that are unrelated to a crenulation structure; there will be grain-boundary sliding and other adjustments as a consequence of removal of material along the “solution ways.” Deformation of a rock possessing a strong preferred orientation of phyllosilicates (parallel to bedding or to a previous cleavage) may result in the development of a crenulation structure and thenec a differentiated crenulation cleavage. Mechanical re-orientation will in this case be a more important element in formation of cleavage, although mass transfer and crystallization are still necessary to produce a domainal cleavage.