APPENDIX NO. 8. TEST OF A TRANSIT MICROMETER. 
4 71 
Table VI. 
Declination for group 
Mean value of 
sec S (=colli- 
mation factor) 
Mean residuals, without regard to sign 
J. F. H. with 
rapid speed of 
driving heads 
J. F. H. with 
slow speed of 
driving heads 
Other 
observers with 
rapid speed of 
driving heads 
Other 
observers with 
slow speed of 
driving heads 
O 
0 
5 
5 
S 
2 tO 
—22 
I. O 
0.03 
O. 04 
O. 09 
O. 07 
3 to 
16 
• I. O 
O. 03 
O. 04 
O. 08 
O. 08 
17 to 
29 
I. I 
0. 04 
O. 04 
0. 08 
O. 09 
30 to 
38 
I. 2 
O.05 
O. 06 
O. 07 
O. 08 
39 to 
46 
1-3 
O. 07 
O. 07 
O. 09 
O. II 
47 to 
52 
1.5 
O. IO 
O. 07 
O. 08 
O. IO 
53 to 
57 
i-7 
*0. 17 
O. 08 
O. 12 
O. 06 
58 to 
64 
2. i 
O. 08 
O. 09 
O. II 
O. 09 
65 or more 
3-4 
O. II 
O. 08 
O. 12 
O. 08 
Table VI indicates that the residuals of J. F. H. with the transit micrometer for 
slow-moving stars are much larger in proportion to those for equatorial stars than for 
key observations. Thus, for stars of declination about 6o° the residuals are fully double 
those for equatorial stars, whereas for key observations they would be but 50 per cent 
greater. On the other hand, for all other observers than J. F. H. with the transit 
micrometer, the accuracy is much more nearly the same for stars of different declinations 
than for key observations. 
Each observer who was just beginning to use the transit micrometer usually showed 
by his remarks, and by the character of the record on the chronograph sheet, that he had 
some difficulty in keeping up a steady motion of the driving heads when observing an 
equatorial star, even after the change to the slower speed of the driving heads had been 
made. On the other hand, little difficulty of this kind was encountered after one became 
expert in the manipulation of the driving heads. This is the apparent explanation of the 
relatively large errors for equatorial stars of the other observers as compared with J. F. H. 
For J. F. H. and with the slower speed of the driving heads the relative magni 
tudes of the residuals for different declinations correspond closely to a constant angular 
error, as shown in Table VII. * 
* There was but one residual in this group; hence this value is uncertain.