GOULD — REDUCTION OF D AGELET S OBSERVATIONS. 



Reductions to the mean of wires. 



$ 6. REFRACTIONS. 



Readings of the thermometer and barometer were unfortunately not made by d'Agelet 

 with desirable regularity or frequency. On many dates they were omitted altogether ,• very 

 rarely were they made more than once on any night ; and the time of the readings was not 

 noted. Indeed, no clue is found to the time of observation of the meteorological instruments 

 other than the place in the printed column at which they appear, and this place seems to have 

 been dictated, in many cases at least, by the convenience or taste of the printer. 



Earnest endeavors to obtain a meteorological register for Paris, during the years 1783-5, 

 have proved unsuccessful, and for those dates on which no note of the temperature or of the 

 barometer is recorded I have been reduced to the necessity of adopting arbitrary values for 

 the coefficient of refraction, having regard, of course, to the hour of the night and to the 

 season of the year, and being guided by the tables of mean temperature at Paris. 



In computing the refractions, the tables prepared by Prof. Coffin, and published in the 

 appendix to tho Washington observations for 1845, have been employed. These are expanded 

 (8) 



