Geology Reference
In-Depth Information
century regional planation by the wind was favoured by workers such as Keyes (1912) and Jutson
(1914). But even in arid lands the idea of widespread aeolian erosion is inconsistent with the field
evidence, and the concept enjoyed only brief and limited favour, both generally and in respect of
inselbergs. And though recent work suggests that some extensive playa depressions, for example
in northern Iran, are substantially of aeolian origin, these features are e xcavated in weak uncon-
solidated materials. There is no evidence that wind can erode at a re gional scale and in resistant
country rock.
That the plains adjacent to or sur rounding bornhardts have been eroded by rivers was suggested
by Bornhardt (1900), Falconer (1911) and many others. Like Passarge (1895) they implied differen-
tial erosion but emphasised fluvial action. Rivers have undoubtedly been responsible for erosion of
the regolith and for the extensive exposure of the weathering front in etch plains (Chapter 4). Thus,
in specific instances exposure of the bornhardt masses can, in whole or in part, reasonably be attrib-
uted to wave action, or to glacier ice, or to nival processes, but rivers have been most widely active.
6.3.2
The scarp retreat hypothesis
Fluvial erosion of a par ticular type is invoked by many writers, and par ticularly by the late King
(1949, 1968). They interpret inselbergs as Fernlinge, monadnocks de position, or remnants o f cir-
cumdenudation remaining after long-continued scarp retreat and pedimentation (
Fig. 6.9a).
Vertical
fractures determine the location of major rivers, and hence, indirectly, divides or interfluves, and
scarp recession proceeds from the valleys thus determined.
The idea of scar p retreat can be traced back to F isher (1866, 1872) in the mid Nineteenth
Century. The mechanism was invoked by Dutton (1882) in their interpretations of the landscapes
they explored in the American West, Holmes and Wray (1912) and Holmes (1918) explained some
of the gneissic inselber gs of Mozambique in ter ms of scar p recession. But it w as King (1949,
1968) who applied the concept to landscape evolution both in a broader sense and as a basis for a
general theory of inselberg development. Most proponents of scarp retreat argue that the process
is restricted to caprock situations in arid and semi-arid lands. This assertion is incorrect for scarp
retreat occurs in any environment in a caprock situation, as for example in sedimentary terrains in
central Labrador. Nevertheless, King (1949, 1968) while passionately urging that scarp retreat and
pedimentation are dominant wherever running water is active (i.e. everywhere save in glacial areas
and the dune deserts), nevertheless conceded that the process attains optimal effects in the semi-
arid tropical and subtropical lands.
Certainly, inselbergs are well-developed and preserved in such regions. Rates of geomor pho-
logical activity are low in desert lands. Scarp-foot weathering and erosion lead to the steepening
and the recession of slopes and to the formation of a pronounced piedmont angle (see Chapter 9).
Furthermore, granite is resistant under dry conditions, but is very susceptible to alteration when in
contact with water (Chapter 3). In hot, arid and semi-arid lands not ony is the contrast in erosional
vulnerability between high and lo w levels in the local topo graphy more pronounced than else-
where, but the dry granite effectively acts as a caprock so that, e ven in granitic terrains with no
duricrust development, the formation and maintenance of escarpments is theoretically feasible.
Jessen (1936) interpreted the inselbergs of Angola in terms of scarp retreat, but suggested that
the bounding slopes became steeper late in the cycle (Fig. 6.9b). This could be explained in terms
of a diminishing rate of scar p recession, which, as in plateau ter rains (Willis, 1936), would allow
more time for basal weathering and scarps being regraded to a maximum inclination commensurate
with stability.
Proponents of the scarp retreat hypothesis attribute the rounded form of bornhardts to differen-
tial weathering under epigene attack. This explanation had earlier been applied to boulders (see
Chapter 5) and in respect of larger residuals by Mennell (1904) who pointed out that outstanding
edges of fracture-defined blocks are readily removed. Falconer (1911, p. 246) thought that the insel-
bergs of Nigeria “naturally assume that configuration of surface which afforded the least scope for
the activity of the agents of denudation”. The domes of the southeaster n Piedmont of the USA
have also been construed as due to rounding by weathering (granular disintegration).
