I'm writing about some "trivia" of yours that Grace LeMonds printed in her Mensa newsletter. She quoted "Port Of Call" for an item saying the earth's rotation is slowing 1/1000 second each day.
I wrote a short article about 10 years ago about the slowing of the rotation based on geology data from stromatolites. I thought up various calendars for the different numbers of days in the year. The article and a graph are attached.
The graph shows about 58 million years +- from the present results in 5 days difference in the length of the year. This is 49 days to change 1/1000 second, quite a difference from your source. If your source was referring to seconds of longitude, I come up with 15 days for 1/1000 second.
This is not a real big deal. I haven't found but one person in the last 10 years who got excited about Paleozoic calendars. I found one Mensan working on calendars (naming the months & intercalation days & everything) for Mars & other planets for future space colonization.
But if it isn't too much trouble I would like to have a little more info. Thanks to Grace for forwarding this to you!
It appears you already know more about the subject than I. The source for the original factoid was Discovery Magazine. I wasn't able to get a good scan of your graph and so below is what I hope is an accurate Quattro Pro generated replacement. (The Y axis is reversed from your graph - Quattro Pro wouldn't scale from positive to negative.) KP
0 days/year = 4,168,763,500 years in future
Million years = -11.414 x (Days in year) + 4168.86
Fits the stromatolite data from 133 mil. BP to present<\P>
Million years = -9.94 x (Days in year) + 3613
Fits data from 600 mil. BP to 133 mil. BP<\P>
Change shows curve in line, and predicts a higher date for rotation to stop.
Factoid: According to a new study reported in the journal Science, 900 million years ago, the day was only about 18 hours long.