John;

I agree that things should be as you said, but the data don't show it.

I went thru the NEIC catalog from 1973-2006 and extracted all 6+ quakes.

Then I compared each quake to those following, looking for a quake over 1000 km away. When I found it, I computed the time interval between them and added one to the tally for that interval. Then start over. Thus no multiple uses, no aftershocks, just time to the next distant quake.

If quakes are unrelated the tally should be a fairly flat bell curve centered on the average time between mag 6+ quakes. I don't know what effect the 1000km requirement should have. But it isn't. It's an exponential, with zero the highest and extending beyond 30 days (where I stopped plotting.)

This says to me there IS such an effect even though there shouldn't be.

Roger