xlviii 



INTRODUCTION. 



The side argument is the mean temperature of the air column (6) given 

 for intervals of i° from o° to ioo° F. The top argument is the approximate 

 difference of altitude Z obtained from Table 51. 



For temperatures above 50° F., the correction is to be added, and for 

 temperatures below 50° F., the correction is to be subtracted. It will be 

 observed that the correction is a linear function of Z, and hence, for exam- 

 ple, the value for Z = 1740 is the sum of the corrections in the columns 

 headed 1000, 700, and 40. 



In general, accurate altitudes cannot be obtained unless the tempera- 

 ture used is freed from diurnal variation. 



Table 53 gives the correction for gravity, and for the effect of the vari- 

 ation of gravity with altitude on the weight of the mercury. When alti- 

 tudes are determined with aneroid barometers the second factor does not 

 enter the formula. In this case the effect of the latitude factor can be ob- 

 tained by taking the difference between the tabular value for the given lati- 

 tude and the tabular value for latitude 45° 29'. The side argument is the 

 latitude of the station given for intervals of 2°. The top argument is the 

 approximate difference of height Z. 



Table 54 gives the correction for the average humidity of the air at 

 different temperatures. In evaluating the humidity factor as a function 

 of the air temperature, the tables given by Prof. Ferrel have been adopted 

 {Aleteorelogical researches. Part in. — Barometric hypsometry and reduction 

 of the barometer to sea level. Report, U.S. Coast Survey, 1881. Appendix 

 10.) These tables by interpolation, and by extrapolation below o°F., give 

 the following values for jS: 



For Fahrenheit temperatures, 



This correction could have been incorporated with the temperature 

 factor in Table 52, but it is given separately in order that the magnitude 

 of the correction may be apparent, and in order that, when the actual hu- 



