Notable progress has occurred in the study of the role of microstructure such as grain-size and preferred orientation and of chemical environment especially of water (or hydrogen). The importance of grain-size sensitive creep in crustal and mantle rocks has been increasingly recognized (e.g., Tullis and Yund, 1991; Karato and Li, 1992; Karato and Wu, 1993). Grain-size sensitive creep is important because it provides a mechanism of significant rheological weakening upon grain-size reduction, leading to shear localization. Also, the extent to which grain-size sensitive creep dominates deformation properties controls the extent of weakening effects due to partial melting (Kohlstedt, 1992; Hirth and Kohlstedt, 1994a,b).
There has been renewed interest on lattice preferred orientation (LPO) and its effects on rheology. This interest is motivated by the progress in seismology to detect seismic anisotropy down to the center of the earth. Mainprice et al. (1990) discussed the effects of phase transformation in mantle minerals on seismic anisotropy. Gleason et al. (1993) investigated the mechanisms of LPO in quartzites. Meade and Silver (1993) conducted high pressure and room temperature deformation experiments on silicate perovskite and found no preferred orientation. In contrast, Zhang and Karato (1994a) found, based on high temperature shear deformation experiments, a strong preferred orientation of perovskite in the dislocation creep regime but not in the diffusion creep (or superplastic) regime. Zhang and Karato (1994a) suggested that the absence of seismic anisotropy in the lower mantle is a strong evidence for grain-size sensitive creep. Zhang and Karato (1994b) studied the relation between LPO and deformation conditions in olivine based on simple shear deformation experiments under high temperature and pressure. A LPO will result in plastic anisotropy and hence, in some cases, strain localization. Theoretical studies on plastic anisotropy and LPO development during deformation include Takeshita (1989), Ribe and Yu (1991) and Wenk et al. (1991). Kronenberg and his co-workers' work on micas includes detailed microstructural observations to delineate the mechanisms of strain localization and plastic anisotropy (Shea and Kronenberg, 1992, 1993; Kronenberg et al., 1990; Mares and Kronenberg, 1993).
The role of chemical environment on creep, particularly that of water has been investigated both in olivine and in quartz. For olivine, the most significant progress was the identification of the crystallographic mechanisms of water incorporation through the study on the effects of oxygen fugacity and oxide activity on the solubility of water (Bai and Kohlstedt, 1992, 1993). Mackwell et al. (1994) showed a large effect of water on the creep strength of diabase and discussed its possible implications for the difference in tectonic style between earth and Venus.
The role of partial melting is sensitive to the wetting relation between melt and matrix. In crustal rocks, Tullis and Yund (1991) found that the role of water (or water-rich fluid) to wet grain-boundaries is dramatically different between quartz and feldspar. Water wets feldspar to enhance diffusion creep through grain-boundary mass transport (Coble creep), whereas it does not wet grain-boundaries in quartz (Watson and Brenan, 1987). Similarly, basaltic melt usually does not wet olivine grain-boundaries and hence its effects on creep is moderate when the melt fraction is small (Kohlstedt, 1992; Hirth and Kohlstedt, 1994a,b). However, the complexities in melt geometry as a result of anisotropy in surface energy and of differential stress have been realized in recent studies, which might drastically alter rheological and melt migration behaviors (e.g., Waff and Faul, 1992; Hirth and Kohlstedt, 1994a,b; Jin et al., 1994).