In response to John Whalley's comments, for what it's worth I've given below
my present understanding of the essentials of plate tectonics. Hopefully any
comments will be kind as I've been rather out of touch with the literature
recently!
Brian Walton.

Plate tectonics is possible on Earth because a relatively thin, quasi-rigid
lithosphere overlies a ductile asthenosphere. Plate separation is probably
often initiated by the rise of plumes from the core-mantle boundary and by
the slow circulation of two major, mantle-wide convection currents at present
rising beneath the centre of the Pacific and central Africa (Pavoni, 1993).
These major convection currents may be semi-permanent and possibly controlled
by the Earth's rotation, in view of their equatorial location.
The orientation of transform faults shows that plates tend to move
approximately latitudinally near the equator and longitudinally near the
poles (Doglioni, 1993). This may be the result of Earth's rapid easterly
roation so that lithospheric plates tend to lag behind the underlying
asthenosphere.
Consequently, there is a tendancy for plates to move slowly westwards
relative to the asthenosphere and this is most effective near the equator..
Plate movement is also facilitated by the tendency for oceanic lithoshere to
sink into the asthenosphere at subduction zones due to the slightly higher
density of the former, and also by the pressure exerted by magma intruded at
oceanic ridges. These effects may locally reverse the direction of plate
movement relative to the asthenosphere e.g. the Nasca plate.
Although spreading ridges probably originally form above rising plumes, e.g.
Iceland, they generally become passive features which migrate as oceans open.
Likewise, subduction zones migrate as oceans close.
References : -
Doglioni, C. 1993 Geological evidence for a global tectonic polarity. Jour.
Geol. Soc. London, V.150, 991-1002.
Pavoni, N 1993 Pattern of mantle convection and Pangaea breakup as revealed
by the evolution of the African plate. Jour. Geol. Soc. London, V. 150,
953-964.