´) have to be calculated, subtracting the measured pore pressure (
u) from vertical (1) and horizontal (3) stresses.
Different modes of failure were observed; most firm and hard samples failed by shear and some of the softer samples showed plastic broadening (see Plates P1, P2, P3, P4). Different modes of brittle failure occurred, as there are near-vertical extension fractures, slightly inclined hybrid-extension shear fractures, and shear fractures. Proof for the realistic conditions of the triaxial tests is the perfect parallelism of healed original fractures and artificial shear planes. The fact that a new shear plane was generated instead of reactivating a preexisting discontinuity (Plate P3, fig. 1) has an important implication—the inhomogeneity does not influence later deformation in terms of being a location of mechanical weakness. In conclusion, it follows that apart from the need to find intact parts of the cores for preparation of test specimens, there seems to be no need for the samples to be free of healed fractures.
For the evaluation of the results, the effective stresses (´) have to be calculated, subtracting the measured pore pressure (u) from vertical (1) and horizontal (3) stresses.
On Plates P1, P2, P3, and P4, the stress paths ([1´-3´]/2 vs. [1´-3´]/2) are presented. From the linear (elastic) parts of the 1´-3´ vs. 
1 curves, a stiffness modulus for triaxial compression could be calculated by
(1´ - 3´)/1
with
(1´ - 3´) = increment of differential stress (kPa); and
1 = increment of strain (percent).
The stiffness modulus is a measure of the amount of elastic deformation the material can sustain before irreversible plastic or brittle deformation occurs. Figure F6 shows the stiffness moduli vs. depth at both sites. The stiffness modulus depends on the differential stress evolved during the compression stage. The values increase with depth and, disregarding the Section 186-1151A-107R-2 samples, show an approximately linear rise.
Under the SEM, the slightly polished fault plane of the sample from Section 186-1151A-51R-1 shows a kind of striae or groove-oriented downdip (Plate P4, fig. 4a, 4b). An overview of the fracture plane is given in Plate P4, fig. 3. The striae can also be observed at this low magnification. Remarkably, at the end of many striae, a fossil impressed into the matrix can be found (Plate P4, fig. 4b). Consistently notable is the size of open pore voids, probably leading to a high permeability (Plate P4, fig. 5). Observation of the alignment of the clayey minerals was inhibited by the "cemetery" of fossils in the samples. Nevertheless, when we observed clayey minerals on the fault plane, they were present cumulatively (Plate P4, fig. 6) or in unique scales on fossil frameworks.
