Boulders as examples of two-stage forms - Landforms and Geology of Granite Terrains - page 98

(see Chapter 2). It has been argued that during cooling and crystallisation the granite becomes sta-
bilised in conditions of high lithostatic pressure, if only because of the weight of superincumbent
rocks. Some granitic masses may have been emplaced at comparatively shallow depths, but even so,
that the overlying rocks have been eroded away is evidenced by the very exposure of the granite, so
that the vertical loading has undoubtedly decreased in time. The argument is persuasive and both
laboratory work and practical experience in deep mines (Leeman, 1962), suggest that fractures
develop parallel to the surface of voids under conditions of diminished lithostatic pressure. Thus, in
deep tunnels banks of fractures aligned parallel to the voids, and therefore of arcuate shape, develop
in the intradosal zones adjacent to the tunnels ( Fig. 5.15) as a result of expansion of the bedrock.
Against the suggestion of pressure release are the occurrence of concentric structures around
corestones in such rocks as basalts that have never been deeply buried, the occurrence of the struc-
ture all round the corestone rather than preferentially on the upper sides, and the development of
flaking on the interiors of tafoni located within orthogonal joint blocks and sheet structure, in
granite, volcanic rocks such as trachyte, and sedimentary rocks such as quartzite. Also, in granites
and other plutonic and metamorphic rocks any tendency to expansion through unloading would,
surely, be counterbalanced, in part at least, by contraction on cooling (see also Chapter 2).
Some argue that corestones reflect primary petrological structures. Nodular or concentric struc-
tures were noted in the Dartmoor granites of southwestern England and some have attributed the
formation of boulders to curved joints or fissures. Curved joints certainly exist ( Fig. 5.16) , for
example in Remarkable Rocks, Kangaroo Island, South Australia or the granites of the Serra de
Xurés in southern Galicia, but they are uncommon and they have not influenced the shape of the
vast majority of boulders; primary sets of concentric or spherical fractures of a radius consistent
with the observed size range of corestones and boulders have so far not been located. On the other
hand, it is apparent that flow in magmas has determined the distribution of various minerals and
hence fractures. Thus, at many sites in the vicinity of A Coruña, northwestern Spain, fractures in the
granite exposures frequently run parallel to bands of biotite. Mineral banding, a primary petrogenic
feature, has also contributed to the development of corestones by influencing the course of weath-
ering within joint blocks. Near the Tooma Dam Site in the Snowy Mountains, New South Wales, for
example, mineral banding occurs in the marginal areas of blocks of diorite, and the shape and size
of the corestones are clearly related to these (Fig. 5.10) . Again, in the Lake Tchad region of central
Africa, corestones of granite embedded in rhyolite have been explained as a magmatic feature, with

Figure 5.15. Section through intradosal zone of tunnel (Leeman, 1962).