Streck et al. (2007) concluded that the high-magnesian andesite (HMA) from Mt. Shasta represents a mix of dacite, basalt, and underlying Trinity ophiolite. The authors present two mixing models calculated to reproduce the major element composition of the HMA (average of samples 85–41a–d; Baker et al., 1994). The two models mix a Mt. Shasta dacite (sample 83–58; Grove et al., 2005); a theoretical harzburgite, thought to come from the Trinity ophiolite; and a Mt. Shasta basalt, either a high-alumina olivine tholeiite (HAOT; sample 85–38; Baker et al., 1994) for model 1 or a calc-alkaline basalt (CAB; sample 85–1a; Baker et al., 1994) for model 2. These models led the authors to conclude that HMAs are not mantle-derived melts, that the HMA of Mt. Shasta is produced because of the availability of ultramafic material (Trinity ophiolite; Quick, 1981) underlying the volcano, and that the “primitive” and “adakitic” signatures of HMAs may be decoupled on a local scale (adakitic signature coming from the dacite, and the primitive signature coming from the entrained ultramafic debris). However, data available in work cited by Streck et al. (e.g., Baker et al., 1994; Grove et al., 2002, 2003, 2005; Quick, 1981; Anderson, 1973) are inconsistent with their conclusions and were overlooked in the development of the mixing models.
Primitive Olivine Not From the Trinity Ophiolite
Olivines (Fo90–94) found in the HMA are interpreted by the authors as xenocrysts from the Trinity ophiolite. Abundant spinel from these olivines have chromium numbers from 70 to 80 (where chromium number is molar [Cr/Cr+Al]; Baker et al., 1994). Spinels from the Trinity ophiolite, in contrast, have chromium numbers from 40 to 60, and the most primitive olivines reported are Fo92.6 (Quick, 1981). Further, melt inclusions in HMA olivines described by Anderson (1973) show compositions similar to that of the host HMA. These pieces of evidence indicate a magmatic, phenocrystic origin for the primitive olivine found in the HMA, not a xenocrystic origin from the Trinity ophiolite.
Mixing Models Fail for Trace Elements
Mixing models must produce the target composition for both major and trace elements and isotopes. Results of the mixing models of Streck et al. for Sr and Nd isotopes indicate that their models do not reproduce the HMA even with the large variation in the Trinity ophiolite samples (Fig. 1). Mass balance calculations show that mixing the dacite and basalts in the two models produce an overabundance of many trace elements, including Sr, Ba, Dy, Lu, Y, Yb, Pb, and Zr. This illustrates the inability of the mixing models to produce the HMA with any third component, Trinity ophiolite or otherwise.
Mt. Lassen HMA
The assertion by the authors that the HMA of Mt. Shasta was the only HMA magma in the Cascades is incorrect. The studies of Clynne and Borg (1997) and Borg et al. (1997) describe HMAs for the Mt. Lassen region. That region has no underlying ultramafic body but the HMAs still contain primitive olivine phenocrysts (Fo89–91) with abundant euhedral Cr-spinel inclusions (Cr#s 71–77; Clynne and Borg, 1997). Based on a comprehensive study of the primitive magmas in the region, the study of Clynne and Borg (1997) determined that the Mt. Lassen HMAs were produced by higher extents of melting of less fertile mantle, with a greater slab-derived component, than those of the HAOTs and CABs of the region. This is also the model proposed by Grove et al. (2002) for the production of the Mt. Shasta HMA.