NavList:

I have followed this discussion with interest, and with some trepidation regarding the apparent nastiness involved from time to time - it is not my intent to involve myself in the discussion of air navigation, although I have on several occasions been forced to, let us say, assist in this area, but rather to contribute something to the subject of celestial navigation. 

The matter of amplitudes has been well presented and discussed, however. I believe one important point, as respects surface navigation, at least, has been neglected. By definition, azimuth is the angle at the zenith between the vertical circle passing through the north point (or for that matter any cardinal point) of the horizon and the vertical circle passing through the body observed, as measured on the plane of the horizon. As such, all instruments utilized in the measurement must be absolutely horizontal or vertical, as the case may be - the resultant error due to a deviation from the horizontal or vertical by the instrumentation employed being at a minimum for a body in the horizon and increasing with the altitude of the body. This is one of the more important reasons for the use of amplitudes in determining compass error and a reason for the apparently long forgotten adage to restrict the use of bodies at an altitude greater than 30 degrees for the determination of compass error, if at all possible.

All azimuth measuring instrumentation, except perhaps the shadow pin, comes equipped with a spirit level which should be used to prove the horizontal while taking azimuths. This can be tricky business in a seaway, hence the advent of the amplitude which serves to at least mitigate any resultant error to the extent possible. It was always amazing to me to see how often this practice was neglected by navigators, both merchant marine and navy, with the consequent erratic compass error record.  

Regards,

Henry