Some thoughts on Paul Eggert's proposed Israel DST rules for =>1999. The proposed rules of the third Friday in March to the third Sunday in September are very reasonable. It seems that the dates of the change are always set to avoid Jewish holy days/periods and the Jewish Sabbath, sunset Friday to sunset Saturday, and Paul's proposed rules do that. Passover, Nisan 15 in the Jewish calendar, is the first full moon after March 27 in the Gregorian calendar in this century. A third-Friday in March beginning prevents the beginning of DST from running into Passover. (Back when the Jewish calendar was determined on the basis of observation, Passover was supposed to be the first full moon after the vernal equinox (which survives in Christian Easter rules). I was informed that a group of rabbis met in Jerusalem to discuss reform of the Jewish calendar, but were unable to come to a conclusion (although the person who told me this suggested the rabbis' real reason for doing nothing was to avoid a schism such as the one that happened in Christendom in 1582 when the Gregorian calendar was implemented and the Catholics adopted it and the Protestants and Greek/Russian Orthodox did not.)) Although the Jewish New Year currently can fall during the date range 15-21 September, it can never be on a Sunday. Neither can Yom Kippur (which can occur prior to September 21) be a Sunday, so a third Sunday in September rule avoids that as well. (Tishri 1 can never fall on Sunday, Wednesday, or Friday because Yom Kippur must avoid Friday or Sunday and Succoth (Tishri 21) must avoid Saturday). However, New Years is celebrated for two days -- Tishri 1 and 2 -- and if Tishri 1 falls on Saturday, the clock change might be moved from Sunday to the preceding Friday. Also, the third Friday in March to third Sunday in September satisfies the current Israeli legal requirement of =>150 days of DST per year. Side note: there are always 154 days between the third Friday in March and the third Sunday in September and they are always the same dates in their respective months (the 15th in 1996, the 21st in 1997). I know that this is a bit of a digression, but I do think that Paul's patch does have some sound reasoning behind it. Chris Carrier