98 A MANUAL OF TOPOGRAPHIC METHODS. 



substituted barometers reading more minutely in order to meet the require- 

 ments of the work of this Survey. 



In field work, barometers should be read at each camp hourly during 

 the daytime, if practicable, or, if not, at such hours as to correspond with 

 the readings at the base station and with readings made by the topographer 

 in the course of his work, having in view the use of the camp as a sort of 

 secondary base station. The topographer or his assistant should read the 

 barometer on all stations, and at all important points the heights of which 

 cannot be more easily obtained by vertical angles. 



Measurements of height made with cistern barometers are subject to 

 periodic and accidental errors. The periodic errors are probably due to 

 imperfections in the formulas and constants used in the reduction. Many 

 attempts both from theoretical and practical points of view have been made 

 to remedy these defects, but thus far without success. The accidental errors 

 are due to errors of observation and to local differences in the pressure of 

 the air at the points at which observations are made. Where the hori- 

 zontal distance between the two stations compared is great, such differences 

 may be correspondingly great, and the same is true where there is a con- 

 siderable difference of elevation between the two stations. 



Under favorable circumstances barometric observations should give the 

 height within a score of feet. Where the circumstances are unfavorable — as, 

 for instance, where there is a great difference of elevation between the two 

 stations or a great horizontal distance between them — the error may be large, 

 reaching 100 feet, and even in extreme cases 200 feet. 



REDUCTION OF BAROMETRIC OBSERVATIONS. 



The pressure of the atmosphere at the sea level is approximately 15 

 pounds per square inch, or is equivalent to that of a column of mercury 

 30 inches in height, With elevation the pressure diminishes, but not in a 

 simple ratio to the altitude, as would be the case if all the strata had the 

 same densitv. The density is proportional to the pressure, and as the 

 pressure upon each layer is produced by the body of air above it, it follows 

 that each succeeding layer of air is less dense than that which underlies 



