A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Sean C
Date: 2015 Aug 28, 08:26 -0700
To be clear: the purpose of viewing cylinders (or the arc) through the index mirror is to check perpendicularity error of the index mirror. Perpendicularity error is adjusted with the screw on the back of the index mirror. Side error is checked either by sighting a distant vertical object and moving the sextant left and right horizontally to see if the direct and reflected images are aligned, or by sighting the horizon and rotating the sextant about the axis of the telescope/sight tube to see if the images of the horizon remain aligned. (Either way, the sextant should be set at 0°0.0'.) Side error is adjusted with one of the two screws on the back of the horizon mirror. On an Astra, I believe it is the screw which is furthest from the frame (at about the 3 o'clock position when looking at the back of the horizon mirror).
As Frank has pointed out, side error is really not a source of error in measurements. It may or may not be a nuisance if it is large enough, but some members of NavList claim that a small amount of side error actually helps them differentiate between the two images when doing things like checking index error using a star or overlapping images of the sun.
I hope that made sense.