EXPERIMENTAL EVIDENCE THAT REACTION OF KYANITETO FORM SILLIMANITE IS AT LEAST BIVARIANT

Mixtures of sillimanite and kyanite starting materials used in two different investigations of sillimanite-kyanite equilibrium (E. Althaus, 1967, S. W. Richardson, P. M. Bell, and M. C. Gilbert, 1968) were placed side by side at 750 C and 10 plus or minus 0. 5 kb partial pressure H//2o; after five days, kyanite in Althaus mixture had nearly disappeared while in Richardson, Bell, and Gilbert mixture sillimanite had nearly disappeared; differences in sillimanite-kyanite transformation reported in original papers are not due to experimental error or to differences in experimental technique but must be attributed to differences in starting materials themselves.