I was just looking at the section on medieval noon marks (illustrated in its frontispiece) in Waugh's book "Sundials." The one shown had an opaque disk with a central 1/4" hole supported out from a south-facing wall. He suggests timing the passage of the two edges of the solar image and averaging these times to get the time of noon. He estimates an accuracy of 20-30 seconds.
He also has a section on ceiling dials, in which there was some interest a few days ago. Gordon Gordon Uber [EMAIL PROTECTED] Reynen & Uber Web Design http://www.ubr.com/rey&ubr/ Webmaster: Clocks and Time http://www.ubr.com/clocks/
