The serpentine muds of the various drill holes at Site 1200 are classified as silty-clay serpentine or clast-poor diamictite, depending on the grain size of the matrix and the relative abundance of clasts (see "Lithostratigraphy"). These are nongenetic terms based on grain size and aptly describe the serpentine muds from a sedimentological point of view.
Based on the internal characteristics of the serpentine mud and the geologic environment in which serpentine volcanoes form, the genetic term "tectonic mud breccia" will be used in this chapter to describe serpentine mud containing mostly angular clasts, regardless of clast abundance (in contrast to the classification scheme used to formally describe the cores at Site 1200). These breccias are the product of tectonic and geochemical processes that have changed solid ultramafic rocks to their present state. A proposed summary of the tectonic history recorded in these rocks, from earliest to latest events follows:
As the serpentine mud breccia rose from depth beneath the Mariana forearc, it incorporated large amounts of partially serpentinized ultramafic rocks, which were torn from wall rocks surrounding the conduit and were carried upward in an ascending matrix of serpentine mud. Structures preserved within the rocks recovered from Site 1200 provide information about the tectonic history of the ultramafic rocks that underlie the forearc and also about the history of their serpentinization.
The relict minerals of the original ultramafic rock (especially olivine) commonly show kink banding (see "Igneous and Metamorphic Petrology of Ultramafics"). Orthopyroxenes show undulatory extinction. These features are also typical of the serpentinized peridotites studied at Conical Seamount during Leg 125 (Girardeau and Lagabrielle, 1992) and suggest that metamorphism and shearing affected a large area of the mantle that was later to become the Mariana forearc.
Subsequent to the above metamorphism but before any serpentinization had occurred, regional brittle deformation occurred and a set of tectonic fractures developed in the mantle beneath this area. This was possibly due to flexing of the upper mantle at the initiation of the Mariana arc system (Fryer, 1996). These fractures appear to be oriented at high angles to one another, but stereographic analyses of multiple joints in different clasts failed to reveal any systematic relationships.
The above fractures likely served as pathways along which small amounts of water percolated from below, forming narrow, widely spaced serpentine veins and veinlets (those seen in serpentinite clasts are mostly <10 mm wide). These serpentine veins contain abundant fine magnetite grains, especially at their centers, which give the veins a "striped" appearance parallel to their walls (Fig. F74). Fine birefringent mineral grains (brucite?) are also present in these veins, intermixed with serpentine. Serpentine mineral crystals in these veins are typically fine grained and intergrown in mosaic patterns. The distribution of these veins in recovered serpentinite clasts suggests that little water was available at this time, as no more than 1%-2% of the peridotite was serpentinized during this event.
Subsequent to formation of the first vein set along discrete fractures, large volumes of water (likely derived from subducted oceanic crust and sediment) penetrated the obducted mantle beneath the Mariana forearc and massive serpentinization occurred. This serpentinization process was pervasive, as water percolated along intergrain boundaries and not along discrete fractures as in the first event. Samples of serpentinite cored in Holes 1200A and 1200B suggest that peridotites were 60%-100% serpentinized during this second event, typically leaving residual "bastite" crystals of disaggregated orthopyroxene in a matrix of nearly completely serpentinized olivine.
Large volume increases accompanied this second serpentinization event, as high-density olivine and orthopyroxene were hydrated to serpentine. The serpentine veins formed in the first minor serpentinization event could not increase in volume, however, and were subjected to dilatory stresses when the surrounding peridotite expanded. Expansion of the enclosing rock in directions normal to the vein planes had no effect, but dilation in directions parallel to the veins caused dilation fractures to form in an irregular boxwork pattern across vein planes (Fig. F75). Serpentine-rich fluids migrated into these open fractures, depositing narrow veinlets of uncontaminated chrysotile (Fig. F76). A similar mechanism was proposed by Fryer, Pearce, Stokking, et al. (1990) to explain the origin of these secondary veins (termed "Frankenstein" veins) in serpentinite from Conical Seamount during Leg 125. At Site 1200, the veins consisting of pure greenish white fibrous chrysotile formed by solution precipitation rather than in situ replacement and are restricted in their occurrence to within and near the first-generation banded veins. These crosscutting veinlets are <1 mm thick with fiber axes perpendicular to the veins. Thicker secondary veins must have formed by this same process elsewhere, however, as indicated by the common occurrence of chrysotile asbestos fibers up to 12 mm long in the serpentine mud breccia. Such wide fibrous veins would be zones of weakness in massive serpentinite and would not be preserved within clasts but instead form the margins of some clasts, indicating fracture boundaries (Fig. F77).
The ascent and flow of fragment-laden serpentine mud is a tectonic process, and examination of impregnated mud thin sections at finest microscopic scales reveals that these muds are laden with microscopic angular fragments (see Fig. F9). For this reason, they are referred to as "tectonic breccias" in this chapter. The muds commonly display subparallel alignment of silt-sized serpentine platelets. These zones appear to be highly irregular and may reflect internal flow processes that indicate a swirling of mud around suspended clasts. These same zones, however, could possibly be a consequence of core disturbance.
The South Chamorro serpentine mud volcano formed by the upward migration (protrusion) of low-density, low-viscosity serpentine mud breccia (see "Physical Properties") through obducted mantle and onto the seafloor. This serpentine mud breccia contains angular and subangular rock fragments at all scales—from microscopic to blocks more than a meter across. These fragments (also called "clasts" elsewhere in text) mostly consist of serpentinized and partially serpentinized peridotites but also include mineral fragments from the original peridotites (olivine, orthopyroxene, magnetite, and chrome spinel) as well as sparse clasts of chlorite schist and other aluminosilicate metamorphic assemblages. The shapes of these clasts vary systematically with their grain size and are important characteristics of this unique tectonic breccia.
The initial disintegration of massive serpentinite appears to have been controlled by the presence of weak asbestos-rich serpentine veins formed along an older joint set, and many smaller clasts are seen to have their shapes initially determined by these joints (Fig. F78). The external shapes of the large serpentinite blocks cored in Holes 1200A and 1200B cannot be known directly, but the common occurrence of chrysotile veins on the external faces of many clasts (Fig. F77) suggests that these blocks were also bounded by veins.
Fragmentation of larger blocks into smaller ones likely occurs during contact between blocks during ascent and results in an abundance of angular clasts <25 mm in maximum dimension (Fig. F79; also see the "Appendix").
Clasts within the breccia have a seriate size distribution, from fractions of millimeters to more than a meter in maximum dimension. Below a few millimeters in size, grains mostly consist of individual minerals rather than rock assemblages.
There are two measures of shape that need to be considered when evaluating the morphology of clasts within the breccia: roundness, which is the degree of angularity of grain corners, and sphericity, which measures the equality of the length axes of a clast. Thus, a perfect cube would have low roundness but high sphericity, whereas an elongate clast with well-rounded corners would have high roundness but low sphericity. Sphericity was measured using methods outlined in Krumbein (1941).
Roundness was determined by visual reference to the "roundness charts" given by Carver (1971). In Figure F79, sphericity is plotted against maximum clast length values measured for 250 clasts from Site 1200. Although there is no apparent correlation between clast size and sphericity, the degree of roundness (shown by symbols for each clast) does vary with clast size, as smaller clasts tend to be more angular than larger ones. This apparent correlation between roundness and size is also shown in Figure F80, where the shapes of randomly selected clasts of different sizes are compared, although it should be noted that some clasts may have been rounded by the drilling process.
The distribution of clast shapes in the serpentine mud breccia is quite different from that in most clastic sedimentary rocks, where clasts are in grain-to-grain contact with one another and most clasts are rounded to a similar degree, regardless of clast size. The distribution of clast shapes in these tectonic mud breccias can therefore be used as a "signature" for these deposits and may be useful in establishing correlations with related deposits found elsewhere. Sedimentary serpentinite deposits, found worldwide where deep-sea continental margin accretionary terranes are exposed on land (Lockwood, 1971), have been suggested to be derived from serpentinite protrusions such as those of the Mariana forearc (Fryer, 1996). If they are related, analyses of clast morphologies in subaerial deposits might be useful to establish or to reject correlations with the Mariana protrusions.