402 REPORT OF THE CALIFORNIA EARTHQUAKE COMMISSION. 
portion of the stream trench. They very commonly extended for several hundred feet, 
in some instances for several hundred yards, and were characteristically arranged in 
linear series. ‘The cracks in the series in some cases overlapt en échelon, and in others 
they were in groups of parallel cracks in belts a few hundred feet wide. In no case was 
there any suggestion that they were more than purely superficial phenomena. A unique 
manifestation of surface cracks is that described by Matthes and Crandall in the vicinity 
of Livermore. (See plate 141,.) 
On the hillsides and ridge crests, at points not within the Rift zone, cracks were of 
common occurrence. Most of these were connected with landslides, as has been indicated 
in the section dealing with that subject. Roadways and artificial embankments were 
particularly susceptible to damage from such cracks. But some of the cracks had no 
apparent connection with landslides, actual or incipient, and these are of especial 
interest. The most northerly are those described by Mr. E. S. Larsen in the region north- 
west of Covelo, Mendocino County, as set forth in the record of intensity. Some of the 
cracks described by Mr. Larsen crost the crests of rocky ridges; and altho it was not 
possible to follow them for great distances, they evidently extend down into the rock. 
It is remarkable that in the district where these cracks occur, there was no evidence 
of a local rise in intensity and, therefore, nothing to suggest that they were the seat of 
a supplementary local earthquake. The probable interpretation of the occurrence is 
that they are secondary cracks of a rather exceptional kind, in ground that required 
no very severe shaking to rupture it superficially. Cracks of a similar character were 
noted by Mr. C. E. Weaver in the Clear Lake district and on the flanks of Mount 
St. Helena. 
On the San Francisco Peninsula, similar cracks were observed by Mr. R. Crandall on 
Cahill Ridge and Sawyer’s Ridge, and are described by him in his account of the dis- 
tribution of intensity in that region. In the Santa Cruz Mountains, such cracks were 
common and are described more-or less in detail in the section on the distribution of 
intensity. In general they appear to be the result of the earthquake rather than a con- 
tributory cause, although in some cases it is quite possible that they may have been local 
ruptures of the nature of auxiliary cracks and so gave rise to subordinate vibration. 
EFFECT OF THE EARTHQUAKE UPON UNDERGROUND WATERS. 
SIGNIFICANCE OF THE PHENOMENA. 
Perhaps the most interesting and significant fact which the study of the earthquake 
has brought to light, apart from the great fault along the Rift, was the general disturbance 
of underground waters. In earthquakes generally, the phenomena which appear at the 
surface of the earth have become well known and, indeed, almost commonplace in recent 
years; but what transpires in the earth’s crust below the surface, as the earth-waves 
generated at the seat of disturbance pass through it, is as yet a matter of uncertainty 
and inquiry. The effect of the shock upon the movement of underground water, as 
manifested by the behavior of springs and wells, throws light on this question. A few 
pages are, therefore, devoted to recording information of this kind. 
It appears from the reports that have come in that springs and wells were very gener- 
ally and variably affected throughout the disturbed area, indicating a sudden derange- 
ment in the normal movements of such water. This derangement could only have 
been effected by the changes in spaces in the rocks in which the waters in the subsur- 
face region are contained, whether flowing or stagnant. These spaces are of 4 general 
kinds: (1) interstitial spaces, or so-called voids, between the constituent fragments of 
imperfectly compacted rocks, such as sands, gravels, sandstones, conglomerates, tufts, 
