[Duffy]

Roger; 

I respect any objection you have, but I am afraid it is my turn to be a little confused here, because I am trying to understand your definition of twilight. I've been through my recent post and can find no mention of twilight anywhere.  The trigger idea I posted relates to the quake epicentre and its opposite time longitude. Nowhere does it relate to the crosses on my map, apart from the Crete event at the beginning, which gave me the idea in the first place.    I have been using sunrise / sunset and the morning/ night transition periods, which you refer to as dawn /dusk in your country.  I used a 40 arc minute margin during the test because the sunrise / sunset bands range from 2 to 3 minutes depending on latitude (which I mentioned at the start of the test), I allowed a +/- 1 degree margin for sun and moon position relative to the longitude selected (which I mentioned at the start of the test), and the data is in analogue time which is less reliable than digital, so an acceptable allowance of a 30 second discrepancy is included.   

This will not happen more than chance allows because it is periodic, it is already loosing influence because the last 5 quakes have had no contact with the crosses.  I would be interested to know what period you used for your test cases, because my period was 11 days before new moon ... 153 quakes in 11 days is a lot of 5's per day !

I would appreciate it if you could still provide me with your chosen selection, so I am able to run the test again. I will carry it out under the same conditions stated above, and as you now appear to have means of verifying what I post, you can monitor any discrepancies.

I know you just do this for mental exercise, but I really appreciate your effort. It works for me too, because I just want to find the right answer ... stops me running down loony highway to the press, every time one of my "chance" predictions are correct !


Duffy