Ah - you take a few days off to go bush and everyone horns in on your special subject. Never mind - thanks for all the learned input.

Must however correct one slight inaccuracy:

"It's driven by magma being forced up between the plates in the centre of the major oceans - the Pacific and the Atlantic. The mid-Atlantic ridge is highly volcanic. New plate material is formed by the magma and is pushed away from the scene of its creation. Or something like that, anyway."

The subduction of one plate under another isn't really "driven" by the magma at the opposite side of the plate forcing its way up. As mentioned in an excellently descriptive post, the convection currents within the mantle (driven, it is postulated, by the heat generated from the radioactive decay of the minerals present) carry the crustal plates across the surface of the earth - rather like a conveyor belt.

Where they collide, the crust on one side of the impact zone buckles and heaves - building mountain chains like the Himalayas and the Andes. The other plate is "subducted" (ie pushed under the other) - the line along which this happens often forming a deep ravine ("trench") in the ocean (eg the Mariana Trench off the east coast of Japan). The crustal material that melts has a lower density than the surrounding mantle and rises slowly through it, eventually cooling at or near the surface as granitic or related rocks on the far side of the mountain range.

Crustal material delivered into the subduction zone at one side of the plate must be replaced - this happens at the Mid Ocean Ridges (MOR's). The magma that comes to the surface at the MOR's is filling the void left by the plates as they move apart along the line of the MOR. We've all seen pictures of this happening - especially the Mid Atlantic Ridge where it comes to the surface in Iceland.

Not all plates meet head on. The movement between the Pacific plate (upon which half of NZ's North Island sits) and the Australian plate is lateral and, as mentioned in the previous post, they've been locked together for a long time. Simply put, the rate of movement at the join isn't keeping up with the movement away from there. It has to release sooner or later - and all Kiwis should be hoping for a lot of smaller quakes, rather than one big mutha.

Enough of the rocks - on with the words!

stales