MADE AT HUDSON OBSERVATORY. 47 



clock a short interval before the transit over the first wire, and preserve the 

 counting by listening to the beats. Having recorded the observation for the 

 first wire, I look again at the clock, and so on for each of the five wires. The 

 equatorial intervals of the wires in their orders for stars above the pole were 

 found to be 185.456; 185.419; I85.I8O; 185.374. The reduction to the central 

 wire is, consequently, O5.II2 x secant of declination; positive above the pole, 

 and negative below. 



The pendulum of the clock, as it came from the maker, was found to be 

 over-compensated. At three different times a portion of the mercury has been 

 removed, namely: about two ounces, Nov. 30th, 1838; three ounces, Feb. 12th, 

 1839; and five ounces, March 5th, 1839. At each of these dates the rate of 

 the clock was, of course, changed. I am of opinion that the pendulum is still 

 over-compensated, though in a very slight degree. The column of mercury is 

 now 6.3 inches high. Since March 5th the clock has not been stopped, nor 

 the pendulum touched. The inequalities of the clock's rate, as shown in the 

 accompanying list of moon-culminating stars, are to be ascribed to imperfect 

 compensation; to a change in the adjustment of the pendulum; to errors of 

 observation, the rate having commonly been determined from a small number 

 of stars, and to other causes of a more uncertain character. Fortunately, from 

 the nature of the observations, the results deduced from them cannot be greatly 

 affected by the small uncertainty in the clock's rate. 



I. Latitude of Hudson Observatory. 



For the determination of my latitude, I have made repeated observations of 

 the pole star, near the meridian, both directly and by reflection from the sur- 

 face of mercury. The three microscopes were read at each observation; the 

 observations were reduced to the meridian by the usual method, and corrected 

 for refraction by Bessel's Tables. The mean latitude deduced from sixteen 

 culminations, nine below, and seven above the pole, allowing each culmination 

 a weight proportioned to the number of reflected observations, is 41° 14' 33".7. 

 This is the mean of all the observations I have made, and supposes them all 

 entitled to equal confidence, which is far from being the case. In the first 

 observations the reflected image was quite indistinct, owing chiefly to the mer- 

 cury being placed too near the telescope ; and I have reason to believe that the 



