More to the point, Bruce, the core is warm due to a remainder of fissile isotopes. Quite warm, in fact.
This heats material at the core boundaries, which is liquid (you don't dispute that lava, for instance, is that hot, yes?), and which rises. The Crust is cooler. So the cool stuff at the crust falls.
Now the harder to explain part... It is fairly easy to show, if you want to get into the mathematics of it all, that any irregularity in the rising and falling material must DRAW IN MORE MATERIAL INTO THAT RISE AND/OR FALL.
This means that something rises over there (say at the mid-Atlantic ridge) and falls over here (say at a subduction trench).
In a very complex way, the speed of rise/fall determines to some extent the size of the rise/fall effect.
Furthermore, in the middle of large, slow currents, an irregularity can easiy start a new current that occupies a smallish part...
In other words, convection currents more or less have to arise given the situation, and they can't be entirely stable, either.
This heats material at the core boundaries, which is liquid (you don't dispute that lava, for instance, is that hot, yes?), and which rises. The Crust is cooler. So the cool stuff at the crust falls.
Now the harder to explain part... It is fairly easy to show, if you want to get into the mathematics of it all, that any irregularity in the rising and falling material must DRAW IN MORE MATERIAL INTO THAT RISE AND/OR FALL.
This means that something rises over there (say at the mid-Atlantic ridge) and falls over here (say at a subduction trench).
In a very complex way, the speed of rise/fall determines to some extent the size of the rise/fall effect.
Furthermore, in the middle of large, slow currents, an irregularity can easiy start a new current that occupies a smallish part...
In other words, convection currents more or less have to arise given the situation, and they can't be entirely stable, either.
, BJ.