Further problems arise when the circulation current system is extended to describe plate movements over the spherical surface of the earth. Dewey 10, van Andel 40, and recently Davies 8discuss the geometrical aspects of tectonic movement using Euler’s Theorem, which states that the displacement of a plate over a spherical surface from one position to another can be regarded as a simple rotation about a suitable axis through the centre of the sphere. This basically implies that in the case of the South American plate, the angular velocity will vary along its length. However as the South American plate has not been distorted laterally along its length in it’s movement away from the African plate, the circulatory forces must therefore have been evenly distributed from the large equatorial diameter to the very much smaller polar diameter.
If the west-east convection currents were / are localised along a south-north axis within the upper mantle (Fig.3c), then taken in isolation, a case for the movement of the South American plate may be made. However the African plate has been relatively stationary, and north-south convection currents must have been present to move the present Indian sub continent into the Eurasian plate. This implies that the heated convection currents must have been and still are omni directional and by definition very complex.
It is interesting to note that Davies 8 states that as the plate near the pole of rotation may be rotating about a vertical axis relative to the mantle, it would be inaccurate to think of the mantle motions in terms of simple roll cells of convection. In a spherical shell, the flow may need to connect globally in a complex manner. In his book Davies 8 also summarises other contemporary work which suggests that the ‘return flow’ from sub-duction under the north-west Pacific back to the East Pacific Rise may pass under North America approximating a great circle path, and the flow under North America may have a southerly component that would not be inferred from the local part of the plate system
A further difficulty arises when trying to understand how the convection-based ‘slab–pull’ forces, which moved Pangea northward from its southern polar position in the Permian era, changed direction in the Jurassic era to cause the break up of Pangea in a predominately west – east direction. Nor can the existing current convection hypothesis reconcile the variation in the velocity of the different plates as illustrated by Park 24 and Hamblin16. This applies in particular to the eastward anti-clockwise rotational movement of the Australian ‘continent’ and the extremely rapid movement of what is now the Indian sub–continent in a northeast direction. At present these two ‘continents’ are linked on the same plate.
It follows that it is mechanically difficult to reconcile the sustained unidirectional movements of the various Continental plates, from their positions as part of Pangea 225 my ago to their present individual positions, with the obviously omni-directional convection current flow patterns. Fig 4 shows the author’s attempt to illustrate some of this complex current movement from the data available at the surface showing the relative movements of the various plates.
Fig. 4. This figure shows that the directional convection currents presently attributed to creating the necessary forces to move the continental crusts are omni-directional and thus form a very complex pattern. The convergent and divergent symbols show (a) the northward movement of Pangea in the Permian era, followed by (b) the westward movement of South America, (c) the northeast movement if India (d) the south-east movement and counter clockwise rotation of Australia and (e) the present break-up of the African plate at the rift valley in an eastward direction. It is almost imposible to depict these movements in a 3-D global model