3 is Dr. Brinkley’s results of the application of 
turb the observation. A displacement of the intersection of 
the cross wires of the collimator of only T ~ 5 of an inch, 
might occasion an error of 1" in the inclination ; so that it 
appears really wonderful that the collimation of the collimator 
has not for weeks together suffered perceptible derangement. 
The deviations from the mean are only such as may be 
safely attributed to the errors of observation and slight effects 
of change of temperature, except perhaps in one instance, the 
27th July. 
The determination of the inclination is independent on the 
position of the axis of the circle, and therefore also furnishes 
a criterion of the degree of exactness to which observations 
with the collimator may be made. 
Explanation of, and remarks on Table III. 
The index corrections given in this Table are those, by 
help of which the mean index corrections used for Table I. 
have been determined. It will be seen, that while the circle 
was frequently reversed (often many times in the same day), 
the corrections appear more variable than when the instru- 
ment remained for a length of time without being reversed. 
This was to be expected. An inexact adjustment of the ver- 
tical axis of the circle affects the index correction ; and this is 
much more likely to take place when the circle is frequently 
reversed, than when not reversed. Indeed, from the reversals 
made for the purpose of trying the collimator, a source of 
error arose that was pointed out in the first instance by the 
collimator, and in so doing it furnished a strong testimony 
for itself. 
It will be remarked, that in July and August the index 
