# NavList:

## A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding

**Lunars by Huxtable & Bowditch**

**From:**Bill Noyce

**Date:**2002 Jan 30, 8:45 AM

I'll echo the thanks to George Huxtable for his treatise on lunars. And I would like to take him up on his offer to spell out Borda's method for clearing the distance longhand. Young's formula seems to require converting out of logs to do additions. I'd also like to thank Dan Allen for posting the pages on lunars from Bowditch. I haven't gotten as far as trying to understand why the A,B,C,D calculations work, because I'm hung up on an earlier point: I don't see where the Horizontal Parallax is adjusted by (approximately) cos(Altitude) to give (what we now call) Parallax in Altitude (PinA). The step that looks as if it ought to be doing that is the first paragraph labeled "Moon's Reduced Parallax and Refraction," which gives "Moon's Reduced Parallax." But it supposedly uses the latitude, not the altitude, and it looks as if the result is a small correction, not one that could reduce the parallax to zero for high altitudes. I guess it's possible that the rest of the formulas account for using HP instead of PinA -- that would be pretty clever. But then we need to account for simply adding "Reduced Refraction" to the HP. Oh! That table, "Reduced refraction for Lunars" has a suspicious title, and the values in it are a bit funny -- isntead of decreasing to zero at an altitude of 90 degrees, they're still about a minute. Perhaps they represent Refraction/cos(Altitude), so that they can be added to HP (= PinA/cos(Altitude), approximately) and enter into the same calculation. I was struck by the level of precision in the Bowditch tables -- tenths of seconds of arc! Do the other tables go this far? And he includes the "other corrections" George Huxtable alluded to. It seems nice that Bowditch's A,B,C,D calculation starts from the relatively small values of (Parallax+Refraction), rather than the way Young's formula depends on cos(m+s)*something - cos(M+S) to nearly cancel out. Perhaps this is why Bowditch can get away with only 4-digit logs for his tables? It was interesting that the second paragraph of Art. 305 seems to use "time" unqualified to mean "local time" -- that which you can deduce from measuring the altitude of the Sun in the east or west.