Compression axes inferred from calcite twin lamellae in eleven samples of Madison Limestone from both flanks of Teton anticline are remarkably similar. They indicate that the greatest principal stress (σ1) was subparallel to the dip at one time during the folding and inclined to the bedding at other times. In three of the four samples from the plunging nose, compression axes are more diffuse than are those from the flanks.
The compression axis pattern for the flank stations agrees with one of the two prominent macrofracrure patterns. The second prominent fracture assemblage indicates that at some time during the folding σ1 was parallel to the fold axis. This orientation of the compression axis is never seen in the calcite data, even though some samples were collected immediately adjacent to macrofractures of this group. However, the fabric of a flank specimen, experimentally deformed such that the compressive load was parallel to the strike direction, clearly shows that compression axes inferred from e1 lamellae are subparallel to the experimental σ1, and compression axes inferred from e2 lamellae reflect the original natural fabric. That is, the original e1 lamellae are demoted to e2 lamellae by the superposed deformation. Moreover, the twin lamellae index is much higher in the experiments than in the naturally deformed rocks. Thus, in cases of superposed deformations, the calcite technique statistically maps the compression axis associated with the largest strain. Therefore, for the specimen studied, the shortening associated with σ1 perpendicular to the fold axis probably is greater than that parallel to the fold axis.