It is also obvious that the micrometer may be turned through 180° and still have its wires parallel to any particular line. The position of the stars also depends upon the star fixed on for the centre round which our degrees are counted; for in the case of two stars just one over the other in the field of view, if we take the upper one as centre, then the position of the system is 0°, but if the lower one, then it is 180°; in the case of two equal or nearly equal stars, it is difficult to say which shall be considered as centre, and so the position given by two different persons might differ by 180°. There are also generally two verniers on the position circle, one on each side, and these of course give readings 180° different from each other, so that 180° has often to be added or subtracted from the calculated result to give the true position. All that is really measured by the position micrometer is the relative position of the line joining the stars with the N. and S. line. In order, therefore, to find, whether 180° should be added or not, a circle is printed on the form, with two bars across for a guide to the eye, and the stars as seen are roughly dotted down in their apparent position—in the case in point about 150°. Our readings being now made, we first take a mean of those of position, which is 169°·8, nearly, and the zero is 109°·8; deduct 90° from this to give the reading of the N. and S. line 19°·8, then we deduct this from the mean of position, 169·8, giving us 150° as the position angle of the stars. It often happens that the observed zero is less than 90°, and then we must add 360° to it before subtracting the 90°, or what is perhaps best, subtract the observed zero from 90°, and treat the result as a minus quantity, and therefore add it to the mean of position readings instead of subtracting as usual. The observations of the second star give a case in point: the zero is 88°·9, and subtracting this from 90°, we get 1°·1; we put this down as -1°·1 to distinguish it from a result when 90° is subtracted from the zero; it is then added to the mean of position readings 81°·1, giving 82°·2, but on reference to the dots showing the approximate position of the stars, it is seen that 180° must be added to their result, giving 262°·2 as the position of the stars.