APPENDIX. 



From the refult of this tabic, the time gahied by the 

 pendulum in twenty-four hours of mean time, after de- 

 du<Sling the acceleration on account of the contraction of 

 its rod by the cold, is feventy-three feconds, and fix hun- 

 dredths of a fecond ; which is one fecond, and two hun- 

 dredths of a fecond more than by the refult of the obfer- 

 vations of the i6th and 17th of July. But although the 

 rate of going of the watch from the 15 th to the 18 th 

 days of Auguft, was afcertained by a mean of fifty-five al- 

 titudes of the fun, I am inclined to give the preference 

 to the obfervations of July, where the exaft period of 

 twenty-four hours was determined by a revolution of the 

 fun, obferved with a telefcope whofe magnifying power 

 was fixty. And notwithftanding that the height of the 

 thermometer during the time of obfervation in Ano-ufl: 

 was remarkably uniform, and that the watch was found by 

 the comparifons with the pendulum to have loft durino- 

 the whole time as uniformly as could reafonably be ex- 

 pected ; yet a fmall irregularity in its rate of goino- near 

 the beginning or end of the obfervation, might occafion 

 the difference of this refult from the former. 



As the time corrected by the mean of fix altitudes of 

 the fun taken on the i6th and 17th July, differed 

 only one fecond and a half from that obferved by the 



E e revolution 



177 



