BAROMETRICAL MEASUREMENTS. 197 



extent this correction might differ from the truth, the correction produced much more uni- 

 formity in the resuUs of observations made at any one point than appeared without it. This 

 correction hkewise provides, to some extent at least, for any error which possibly may exist 

 in the ordinary valuation of the effect of the temperature of the air upon the calculated eleva- 

 tion, or of its hygrometrical state so far as it may be indicated by this temperature, which at 

 tlie superior station differs but little from T. The formula therefore used is 



h = p + (451— a) X 0-021 X (16 — T), 



in which h represents the true difference of level between the stations in metres, p this dif- 

 ference according to the tables of Oltmanns, a the upper reading of the superior barometer, 

 and T the temperature indicated by its attached thermometer. 



The additional correction which I applied to the elevations of Lake Colden and Mount 

 Marcy is 



12 ( — VI + 7n' + n — n'), 



in which m, rnf , are the first members of the conditions (1), (2) ; and n, n' , the second. 



A slight correction for capillarity, of about three-tenths of a millimetre, I deducted from 

 the correction 2*50 in the seven observations made during my first visit. 



Oltmanns' tables have been employed in these calculations, which have furnished, in the 

 cases in which I have compared them, the same results as those derived from Laplace's 

 theorem. An example of this agreement appears in the calculation of the height of Mount 

 Marcy. I have not seen the construction of these tables, but conjecture from this agreement 

 that Laplace's theorem was made their base. Humboldt, who made his calculations according 

 to the theorem, mentions the harmony of his results with those of the same heights by Prof. 

 Oltmanns, and pronounces the tables of the latter to be " of the utmost precision." 



