216 PROCEEDINGS OF SECTION A. 
and the third the same correction in terms of R. The corrections 
for readings from 20 to 35 will be the same as those below, but 
in reverse order :— 
122. u R. u 
5 + 3°150 + 0°054 13 + 0°686 + 0012 
55 2°940 050 13°35 O88 ‘010 
6 2°744 047 14 ‘504 ‘O69 
6°5 2°548 043 14°5 *420 007 
ai 2°366 “O40 15 “300 006 
75 2°184 037 15°5 280 005 
) 2°016 034 16 224 004 
85 1°848 ‘O31 165 168 003 
9 1-694 029 17 126 002 
9°5 1°540 026 17°5 ‘084 ‘001 
10 1°400 “024 18 056 ‘001 
10°5 1260 021 18°5 028 “C00 
11 1134 ‘019 19 ‘O14 000 
11°5 1008 017 19°5 “O04 000 
12 0°S96 015 20 ‘000 ‘000 
12°5 0-784 013 
It will be unnecessary to draw attention to the magnitude of 
these corrections ; they are sufficiently startling. They are, how- 
ever, fully borne out by the observations. For instance, to return 
to the case of star 6703 for an example, the observations of this 
star show that ten (10) revolutions of the screw in its different 
parts represent the following angular intervals : 
10 R. 10 R. “1OR:. 
34 -—24 588 62 27°5-17°5 589°53 21°5—11°5 585*66) ~ 
17 
33 —23 588°63 PH lly 589-01 21 —I1 585°39 
32°5—22°5 591-08 26:°5-16°5 588°36 205-10°5 58552 
382 —22 589°68 26 -16 588-49 20 -10 584°87 
31°5-—21°5 590°18 25°5-15°5 588-08 19'5— 9'°5 585°36 
31 -21 590°57 25 —15 586°58 19 - 9 58552 
30°5-20°5 588-76 24:5-14°5 58746 185-— 85 586:17 
_ 30 -—20 588°88 24 -—14 586°82 18 - 8 583°85 
29°5-—19°5 590-05 23°5-13°5 585-92 17°5-— 775 584:22 
A) il, 58849 23 =13 587:07 lyf = of 584°87 
28:5-18°5 58746 22°5-12°5 586°68 16°5— 6°5 585°27 
28 -18 589°53 22 -—12 586 04 16 - 6 585°00 
The meaning of the above is that the result of measurement of 
an angle by ten revolutions of the screw at one part of its length 
differs by as much as 5 seconds from the result when another 
part of the screw is used. 
The effect of this error of screw on the observations ordinarily 
made with this micrometer (zenith distance observations for lati- 
tude) will of course not be so considerable as that just referred to, 
as it is usual to set the telescope as nearly as possible at the mean 
zenith distance of each pair of stars, so that the observations are 
made pretty symmetrically with regard to the middle reading 
(20 R) of the screw. This cannot always be done though, and in 
