AND RATIO OF THEIR ATOMIC WEIGHTS. 27 



The elevation of the cistern of my barometer above mean tide level at New 

 York was during these experiments 216.1 metres, and the latitude of my laboi-atory 

 is 41" .'5(»' 15". The force of gravity there, computed by the formula 



G = G45 (i — 00259 cos 2I,) (i — .000000196 H), is 

 G = .9996423G45 



By the fornuila of Helmert, which is used in recent publications of the 

 United States Coast and Geodetic survey, 



G = 980.5934 — 2.5967 cos 2L (with the same correction for altitude), we get 



(j - 980.235. 



The Coast and Geodetic survey has made my laboratory one of their stations 

 for determining the force of gravity, and its superintendent has furnished me with 

 the I'esult 



G = 1. 000132 G (Washington). 

 The value which has been provisionally adopted for Washington is 



G (Washington) = 980.098. 



Hence 



G = 980.227. 



Dividino- this value Ijy the constant term of the formula of Helmert, \ve have 



J- = .999627. 



All the observations contained in this paper have been reduced to the values 

 which would l)e obtained at the sea level in latitude 45° by dividing by the factor 



.999627. 

 Further, the value of gravity at my laboratory, according to the same 



determination, is 



G = .999311 G (Paris Observatory). 



Ill order, therefore, to reduce the values which will be given in this paper to 

 the values which would be obtained at Paris, we must multiply them by 



lllffi = 1.0003.6. 



The scale of my mano-barometer is of glass ; its coefficient of expansion has 

 not been determined. But the length of the bar at ordinary temperature has been 

 determined by compai'ison with a bar whose coefficient of expansion is regarded 

 as well known. This is one of two bars made ])y the Societe Genevoise, which 

 have been obtained at different times by the Case School of Applied Science, at 



