SHEARING MOVEMENTS IN THE FAULT-ZONE. 37 
A half mile southeast of San Andreas Lake the pipe crosses the fault for the last time 
(M1), from left to right at an angle of 65° (vol. 1, p. 100); it was crusht and shortened 22 
inches; 100 feet to the north it was crusht again, the compression there being 1 foot. 
The total shortening, 2 feet 10 inches, corresponds to a shift 6.75 feet; as the shift at 
the fault-line was only 20 inches, a part of the shear must have been distributed. 
Near the northwestern end of Crystal Springs Lake the 44-inch Locks Creek pipe crosses 
the fault-zone from left to right at an angle of 65° (fig. 22, O, and vol. 1, p. 101, fig. 39); 
it was crusht at four points, and pulled apart 3 inches at one point; the total shortening 
was 59.25 inches; this corresponds to a total shift of 11 feet 8 inches, the greater part 
of which was distributed. 
The shifts indicated by the changes in length of the pipes must be lookt upon in many 
cases as smaller than the true shift, for many other ruptures occurred, which are noted 
in the report of the chief engineer of the water company, but of which no details have 
been given. 
EFFECTS ON OTHER STRUCTURES. 
The two best examples of combined shortening and stretching are furnisht by the gate- 
well on the shore of San Andreas Lake and by the flume and the waste-weir at the 
southeastern end of the lake; they show the existence, at the same place, of shortening 
and stretching in different directions, altho there is no indication of a compression or 
extension at right angles to the fault. The gate-well was stretcht in a direction N.79° W. 
and shortened at right angles to the stretching (vol. 1, pp. 98, 99, fig. 35). The direc- 
tion of the fault-trace is about N. 35° W., so that the directions of greatest stretching 
and shortening make angles of practically 45° with the directions of the fault. From 
the scale of the figure the stretching is found to be 3 feet 4 inches, which corresponds to a 
shift of 4 feet 8 inches. This is less than the shift in this part of the fault and confirms 
the evidence, furnisht by cracks in the ground, that the shear was distributed over a 
greater width than 18 feet, the projection of the diameter of the well (25 feet) in the 
direction of greatest stretching upon a line at right angles to the fault-trace. 
The Locks Creek flume, a 44-inch wrought-iron pipe, crosses a part of the sheared zone 
from right to left at an angle of 15° (vol. 1, pp. 99, 100; fig. 36); it was pulled apart 
4 feet, corresponding to a shift of 4 feet 2 inches. If the pipe had entirely crost the 
sheared zone, it would have indicated a greater shift, which could not have been less 
than 7 feet at this point, according to the displacement of a fence shown in the same 
figure; the flume passes thru a concrete culvert and continues to San Andreas Lake; as 
this part of the pipe and culvert were parallel with the direction of the shear, they were 
uninjured. 275 feet from the break in the flume a strongly built brick waste-weir 
tunnel crosses the sheared zone from left to right at an angle of about 57°; its great 
strength prevented it from being entirely destroyed, but it was crusht at the fault-line 
and shortened, tho the amount was not measured. 
The examples given show very clearly that the shortening and stretching of lines in 
the fault-zone was not due to any general expansion or compression causing changes of 
area, but to shear; and the character and amount of change in length of any particular 
line depended on the direction in which it crost the fault-zone and the angle it made with 
the direction of shift; so that, in some instances, of two lines crossing the fault-zone at 
the same point but from opposite sides, one was lengthened and the other shortened. It 
is quite possible that there were, in places near the surface, slight expansions or com- 
pression at right angles to the fault-line. As pointed out (vol. 1, p. 73) the fault-plane 
can not be considered a mathematical plane, and the movement must have caused a 
slight separation of the sides in places near the surface, which may be indicated by the 
trench-form of the fault-trace. It is difficult to understand how the two sides of the 
