The Mechanics of Obh'que Slip Faulting

By M. H. P. BOTT

SUMMARY
The various mechanisms which could cause oblique slip faulting
are briefly reviewed. It is thought that such faulting may frequently
arise from the existence of preferred planes of fracture within the
rocks. The dynamics of this mechanism is studied in some detail
and an expression is obtained for the first direction of slip within
the plane under the influence of a general stress system. For a stress
system of given orientation it is found that the initial slip may
occur in any possible direction within the plane, the direction
depending on the relative values of the three principle pressures. The
theory suggests that when a pre-existing fault is subjected to a
reorientated stress system (typical or rotated) the movement after
fracture will usually be oblique. In conclusion, the general
implications of the theory are discussed.

I. INTRODUCTION

I N a recent article in the Geological Magazine, Alwyn Williams

(1958) has mentioned two possible types of mechanism which
could give rise to oblique slip faulting. The first of these depends on
a rotation of the typical crustal stress system, with horizontally and
vertically orientated principal pressures, to give an obliquely orientated
stress system ; consequently the direction of movement within the
plane of fracture would also be oblique. Williams has investigated
in some detail the direction of slip for fracture under the influence
of an oblique stress system and has applied the results to account
for the oblique slip faulting within the Ordovician, Silurian, and Old
Red Sandstone rocks near Girvan. The second type of mechanism
depends on inhomogeneity of either strength or elastic properties
within the rocks.
The dynamics of these two mechanisms are discussed in this paper.
In particular the theory of inhomogeneity of strength is developed
as it is thought that this mechanism may be of widespread importance.
A notable feature of the theory is the controlling influence of the
relative value of the intermediate principal pressure on the direction
of the initial slip within the fault plane.

II. THE MECHANICS OF ROTATION OF THE STRESS SYSTEM

A typical crustal stress system, of the type envisaged by Anderson

(1951), may be defined by horizontally and vertically orientated
principal pressures. " Rotation " of a typical stress system can occur
through combination with an obliquely orientated supplementary stress
system (Hafner, 1951).
Perhaps the commonest source of obliquely orientated stress systems
GEOL. MAG. VOL. XCVI NO. 2 11

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 110 M. H. P. Bott—

within the earth's crust is provided by major inhomogeneities of

density within or below the crust. It may be shown, for instance,
that considerable stress differences are associated with post-tectonic
granites of characteristically lower density than the intruded country
rocks (Bott, 1958); their effect is to modify the existing typical crustal
stress system in both magnitude and direction. The approximate
amount of rotation of a given typical stress system may be roughly
obtained by calculating the stress differences caused by the anomalous
" load " on the crust (Jeffreys, 1952, Chapter 6) and combining with the
typical system. The order of magnitude of the load can be obtained
from the gravity anomalies. In the investigation mentioned above it
was found that a fairly large post-tectonic granite, giving a gravity
anomaly of about — 40 mgal., could rotate a crustal stress system
large enough to initiate faulting to an extent of about 10 degrees
maximum. More extensive rotation would be expected within orogenic
belts or perhaps as a result of sub-crustal forces of obscure origin.
It would appear, then, that a large rotation of the typical stress
system as a result of interaction with a supplementary stress system
could be caused by mass distributions giving rise to major uncom-
pensated gravity anomalies or by sub-crustal forces of similar
magnitude. These conditions are likely to be found during orogenic
rather than epeirogenic movements. The main conclusion is that
normally only a limited amount of rotation would occur in regions
suffering epeirogenic deformation.

III. TYPES OF ROCK INHOMOGENSITY

Two fundamentally different types of inhomogeneity are liable to
influence the structural behaviour of rocks. These are discussed
separately below.
(a) Inhomogeneity of Elastic Properties
Elastic properties usually vary from one lithology to another, as
may be shown by a combined study of elastic wave velocities and
densities. A quantitative investigation of the influence of elastic
inhomogeneity in rocks on a stress system has apparently not been
made : it would appear, however, that some increase of stress would be
expected within the less compressible and more rigid rock and that
some rotation may occur, particularly near lithological boundaries.
Another possible cause of elastic inhomogeneity, possibly of lesser
importance, arises from preferred orientation of constituent minerals
within an otherwise homogeneous rock.
(b) Inhomogeneity of Strength
Inhomogeneity of strength depends on an inherent weakness- of the

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 The Mechanics of Oblique Slip Faulting 111

rocks within certain preferred directions. Such a preferred direction

of weakness is likely to result from variation in lithology as different
lithologies are likely to have different strengths as well as different
elastic properties. The plane of weakness should usually lie parallel
to the bedding planes. But preferred planes of weakness may also be
found within otherwise homogeneous (or nearly homogeneous) masses
of rock in the form of earlier faults, joints, or cleavage. The existence
of these latter types should not noticeably affect the elastic properties
and therefore the stress system would remain unaffected while elastic

TEXT-FIG. 1

conditions held. These preferred planes would remain " unnoticed "
until the shearing stress within them should exceed the strength. If
homogeneous fracture had not by then taken place within the rock,
fracture would occur within the preferred plane in which the strength
was first exceeded, and the direction of the initial slip would be defined
by the direction of the greatest shearing stress within the plane.
The dynamics of simple inhomogeneity of strength within otherwise
homogeneous rocks, as a mechanism of oblique slip faulting, is
investigated in the following paragraph.

IV. THE DYNAMICS OF FRACTURE WITHIN A PREFERRED PLANE

It is evident that the direction of movement immediately after
fracture should coincide with the direction of maximum shearing
stress within the fracture plane immediately before. In this paragraph

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 112 M. H. P. Bott—

we derive an expression for the direction of maximum shearing stress

within a given plane subjected to a general stress system, which will
thus also give the expected direction of faulting.
The co-ordinate axes (x, y, x) are taken parallel to the corresponding
principal pressures (cr*, av, crz). We require to find the direction of
maximum shearing stress within a general plane having direction
cosines (I, m, n), identically connected by the relation :—
/2 + w2 + H2 = 1 . . . . (1)
In Text-fig. 1 the plane is shown intercepting the co-ordinate axes at
X, Y, and Z.
We take AXYZ to be of unit area. The component of pressure
acting in a given direction across the face is then equal to the force
acting in that direction. It can readily be shown that AOYZ = /,
AOZX = m, AOXY = n. The normal forces on AOYZ, AOZX, and
AOXY are then found to be hn m<jv, and naz respectively. Since
the system is in equilibrium the components of force on ACYZ are
(— lax, — mov, — naz), acting parallel to the corresponding co-ordi-
nate axes. Therefore the total resultant force exerted on AXYZ by
the material outside, Ft, is given by :—
Ft = - Vpo* + m*al + «2o»
the normal force on AXYZ, Fn, is obtained by resolving each component
along the normal direction as follows :—
Fn = - (/2<x* + m2ay + n2cr,)
The maximum shearing stress acting within the surface AXYZ, Tmax,
may be calculated from the relation F2, = F\ + T2max. Hence
T*max = Pal + wPa\ + n2a2z - (l2ax + m2av + n 2 a 2 ) 2 . (2)'

Combining (1) and (2),

r 2 ™* = m'(cr, - <rxy + n\oz - <rx)2
- {m\arv - ax) + n\az - ax)Y . . . (3)
For convenience we now consider the z axis to be vertical. We are
aiming to calculate the resolved parts of Tmax, the force per unit
area exerted by the material above the plane on the material below
the plane, up the line of greatest slope (Tdiv) and along the horizontal
direction YX (Tstr). These components are connected by the following
relation :—
Tml. = Tl + T d% . . . (4)
Resolving the components of force along the direction YX, it may be
shown that :—
Im (crv - crx)
l.tr = • . • • (A)

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 The Mechanics of Oblique Slip Faulting 113

Combining (3), (4), and (5) we obtain an expression for Tdlv, which
simplifies, on use of (1), to give the following relation :—
n{ m%av - ax) - (1 - n2) (az - ox)
. (6)
V l - n2

TEXT-FIG. 2.—Diagram showing the direction of maximum shearing stress

within the plane of preferred fracture under all possible stress
regimes. The preferred plane is taken as striking at 45° to the two
horizontal principal pressures and as dipping at 60°.

The pitch of the maximum shearing stress (the angle within the plane
between the strike direction and the direction of maximum shearing
stress), 0, is given by the relation :—•
tan 6 = Tdiv/T,tr
Substituting for Tdip and T,tr from (5) and (6) :—
n ( <rs — o O
tan 6 -(1 -H2) O)
r\m* [ .
lm[ av — <TXJ

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 114 M. H. P. Bott—

Under the sign convention adopted for Taiv and T,ir an angle of zero
corresponds to a horizontal force in the direction YX exerted by the
material above the face on the material below the face. A purely
dextral strike slip movement would occur after fracture in these
circumstances. Positive angles (0° to 180°) involve a thrust component
of dip slip after fracture, and are measured clockwise. Negative
angles (0° to - 180°) are measured anticlockwise and involve a normal
component of dip slip.
The main conclusion which may be drawn from the formula of

< DEXTRAL WRCNCH - >l< T DdXTKAL GRAVrTY

TEXT-FIG. 3.—Graph showing the variation of pitch of the direction of

maximum shearing stress as the relative values of the stress differences
are allowed to vary, within the dextral wrench and gravity stress
regimes. The two stress differences are expressed as a percentage
proportion of their sum.

equation (7) is that by unrestricted variation of the relative values of

the three principal pressures the direction of maximum shearing
stress within the plane may lie in every possible direction. Thus
oblique slip faulting may occur in any possible direction for any given
orientation of the three principal pressures. The only exceptions
occur when the plane of fracture lies parallel to one of the principal
pressures.
It is convenient to discuss the general behaviour of the direction
of maximum shearing stress (and thus also of initial slip) in terms of
tectonic stress regimes (Harland and Bayly, 1958). The concept of
tectonic regimes has been mainly concerned with orogenic deformation
and the movement needs to be orientated with reference to the orogenic
belt so giving the primary and secondary regimes of Harland and Bayly.
For faulting by itself, however, the fault plane provides the only

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 The Mechanics of Oblique Slip Faulting 115

convenient reference, and the primary and secondary regimes are

better replaced by dextral and sinistral regimes (related to the plane of
fracture). As an example a plane dipping at 60° and striking at 45°
to each of the horizontal principal pressures is taken for study. The
disposition of the direction of maximum shearing stress for all possible
stress regimes for this model is illustrated diagrammatically in
Text-fig. 2, which shows a plan diagram of the surface XYZ. The
variation of pitch of the direction, under dextral wrench and dextral
gravity regimes, is illustrated graphically in Text-fig. 3.
The general behaviour of the direction of maximum shearing stress
under different stress regimes adapted from Harland and Bayly is
summarized as follows, Text-fig. 2 providing a specific example :—
(i) Gravity radial regimes, az> ay = ax. 8 = — 90°, resulting in
dip slip normal faulting.
In
(ii) Dextralgravity regime, a, > a, > ax. — 90° < 6 < — tan"1—,
m
resulting in oblique slip faulting with normal dip slip and small
dextral strike slip components.
(iii) Dextral axial regime, <rz = ay > ax. 6 = — tan - 1 —, with a
movement pattern similar to (ii).
(iv) Dextral wrench regime, a, > az > ax. When (crz — crx)/
(av - ax) = m"/(l — «2), 0 = 0°, resulting in horizontal dextral
movement. If the ratio of stress differences is greater than
wVO — «2) there will be a normal component of dip slip,
and if less there will be a thrust component of dip slip.
mn
(v) Dextral radial regime, a, > az = ox. 0 = tan"1—, resulting in
oblique slip faulting with thrust dip slip and dextral strike slip
components.
(vi) Dextral thrust regime, av>ax>az. tan"1— < 6 < 90°, result-
ing in thrust dip slip with a component of dextral strike slip.
(vii) Verticalaxial regime, a•„ = ax > az. 6 = 90°, resulting in simple
dip slip thrust faulting.
In the summary above ay > ax, so that the component of strike slip is
always in the dextral sense. If the numerical values of ax and cr, are
interchanged a corresponding pattern of movement with sinistral
components of strike slip are obtained, giving a further five regimes.
Thus a total of twelve regimes correspond to regimes 2 to 13 inclusive
of Harland and Bayly.

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 116 M. H. P. Bott—

V. DISCUSSION
It has been shown in the previous paragraph that the maximum
shearing stress within a preferred plane of fracture—and thus the
initial slip direction after fracture—may lie in every possible direction
for a variable stress system of given orientation, provided the plane
is not parallel to any of the principal pressures. This conclusion is
valid whatever the orientation of the stress system. Thus it is unneces-
sary to postulate rotation of the typical stress system to account for
oblique slip originating in this way, although such slip may equally
well occur in any direction under the influence of an oblique stress
system.
There are two conspicuous .contrasts with the theory of faulting
from fracture in a homogeneous medium, as developed by Anderson
(1951). In the theory of homogeneous fracture the relative value of the
intermediate principal pressure has no bearing on the direction or
incidence of fracture : the direction is, however, controlled by the
internal coefficient of friction of the rock. In the theory of inhomo-
geneous strength as developed above, once the directions of the
maximum and minimum principal pressures are defined, the direction
of the initial slip depends only on the relative value of the intermediate
principal pressure : and the only influence of the internal coefficient
of friction would be to delay fracture—a delayed fracture may involve
an alteration of the relative value of the intermediate principal pressure
during the build-up of stresses, and thus a slightly modified initial
slip direction.
Perhaps the commonest type of oblique slip of this type is likely
to be found in renewed movement along pre-existing fault planes under
the influence of a reorientated stress system. We should expect the
first movement to conform to one of the classical classes of fault,
while subsequent movements would in general show oblique slip.
Examples of such renewed movement are common, frequently being
revealed by slickensiding.
The theory also gives us the chance of determining the orientation
of a stress system causing a given system of movement together with
the relative value of the intermediate principal pressure, provided that
the direction of oblique slip can be obtained for three or mote different
preferred directions of fracture.
Perhaps the theory may be most simply applied to subsequent
movements on vertical joint surfaces. Under a typical stress system
only horizontal slickensiding should be found according to the predic-
tions of the theory, although oblique slip should be characteristic of a
rotated system. This criterion should enable us to determine whether
or not the stress system has been rotated when slip occurred.
It is possible that faulting commonly occurs under the influence

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35

 The Mechanics of Oblique Slip Faulting 117

of a mechanism intermediate between homogeneous and inhomo-

geneous fracture. A set of circumstances favourable to such a
mechanism would occur when a preferred direction of strike for the
fracture is favoured by a basement fracture pattern without a definite
control of the hade of the fracture plane. When the horizontal principal
pressures lie obliquely to the strike, oblique slip faulting would be
expected. The hade of the fault plane would be influenced by the internal
coefficient of friction as well as by the orientation and relative magnitude
of the principal stresses. Primary oblique slip faults, either affecting
the basement or the overlying sediments, could originate in this way.
Perhaps basement control of overlying fracture patterns is of greater
importance in the development of subsequent faulting than has
generally been realized.
Apparent oblique slip faulting may sometimes occur as a resulting
of tilting, but the writer follows Williams in rejecting this as the
most important mechanism, because frequently a multitude of different
tiltings would be required to account for the variety of slickenside
directions in a region. Oblique slip faulting may also take place by
fracture within a homogeneous mass of rocks under the influence of
a rotated system. Under usual circumstances, however, it is thought
that only a limited amount of rotation can occur, and that preferred
fracture planes and directions provide the usual mechanism. In
conclusion, it seems possible that the oblique slip faults of the Girvan
district may also be explained as a single period of inhomogeneous
fracture under the influence of an approximately north-south horizontal
tension, with possible variation of the intensity of the intermediate
pressure.
I should like to express my gratitude to Dr. R. A. Smith for checking
and suggesting improvements to the mathematical side of the paper.

VI. REFERENCES
ANDERSON, E. M., 1951. The Dynamics of Faulting. Oliver and Boyd, Ltd.,
London.
BOTT, M. H. P., 1958. The uplift of granite masses: a problem of isostasy.
Paper read to Section C of the British Association, Glasgow, 1958.
HAFNER, W., 1951. Stress distributions and faulting. Geol. Soc. America
Bull, lxii, 373-398.
HARLAND, W. B., and M. B. BAYLY, 1958. Tectonic regimes. Geol. Mag.,
xcv, 89-104.
JEFFREYS, H., 1952. The Earth, its Origin, History, and Physical Constitution.
Third edition, London.
WILLIAMS, ALWYN, 1958. Oblique-slip faults and rotated stress systems.
Geol. Mag., xcv, 207-218.
DEPARTMENT OF GEOLOGY,
SCIENCE LABORATORIES,
DURHAM.

12

http://journals.cambridge.orgDownloaded: 13 Mar 2015 IP address: 138.251.14.35