46 ASTRONOMICAL OBSERVATIONS 



seconds of time. A tangent screw, with a long handle, gives a slow motion, 

 and enables an obsersrer to keep an object readily in the field of view. The 

 declination circle is also twelve inches in diameter, graduated to ten minutes, 

 and reads by two verniers to ten seconds of arc. 



The clock was made by Molineux, of London. It has a mercurial pendu- 

 lum, the cistern for the mercury being of glass, and the cylinder is terminated 

 by a steel point, which indicates the arc of vibration upon a fixed scale. It 

 loses no time in winding, an operation which I perform every Monday morn- 

 ing. It is regulated to sidereal time, and its rate is tolerably uniform. It is 

 suspended by a stout iron hook, which was inserted in the north wall of the 

 transit room as the building was erecting, and which passes through the oak 

 back of the clock case. It is rendered steady by two screws, which pass 

 through the back of the case, near the bottom, and enter a timber inserted in 

 the brick wall. The case does not touch the floor. An opening in the side 

 wall, between the transit and equatorial rooms, allows the clock dial to be 

 easily seen from the platform of the dome, and thus one clock is made to serve 

 two instruments. 



The instruments were first placed in the observatory, September 8th, 1838, 

 and I at once applied myself diligently to their adjustment. Having verified 

 the line of coUimation of the transit and levelled the axis, the telescope was 

 brought into the plane of the meridian approximately by high and low stars, 

 and, subsequently, by repeated observations of Polaris, both above and below 

 the pole. In noting the transits of Polaris, I do not attempt, by a single obser- 

 vation, to estimate the time when the star is bisected by a wire. The uncer- 

 taint}^ of such an observation I have found to amount to several seconds. The 

 star, in approaching a wire, appears to make an indentation upon it; and, also, 

 itself suffers a partial eclipse. After passing the wire, the indentation appears 

 upon the other side; and the deficiency, also, appears upon the other side of 

 the star. When the light of the star is faint, as when the sun is several hours 

 above the horizon, the star is entirely occulted for three or four seconds. At 

 such times I note the instants of the star's disappearance, and of its reappear- 

 ance ; the mean I consider the instant of the star's passage over that wire. At 

 other times I note the two instants M^hen the star makes equal indentations upon 

 the two sides of the wire, or suffers an equal loss of brilliancy, taking the mean 

 of the two observations. For all transit observations, I take a second from the 



