51 



level. This weio;ht evidently depends not only upon the elevation 

 above sea, but also upon the mean temperature of the air below the 

 balloon and the amount of moisture it contains. 



The temperature and moisture conditions are easily conceived of 

 in the case of a barometer in a balloon with a great ocean of air 

 directly beneath, but when we consider the reduction for elevation of 

 barometric observations taken over extended plateaus and at great 

 distances from sea level, such, for example, as the reduction of obser- 

 vations at Denver, Colo., no clear meaning attaches to the tempera- 

 ture and density of the air column ; in fact, the air column can not 

 have any real existence, and this constitutes a considerable difficulty 

 in computing satisfactory values for the reductions for elevations*. 

 Approximate values only, therefore, are possible. Considerations 

 such as these lead us to see the advantage of making all reductions to 

 a plane, say, 5,000 feet above sea level, in which case an air column 

 actually exists, and has a definite mean temperature, humidity, etc. 



Various more or less arbitrary computations of the temperature to 

 be used in finding the reduction for elevation have been tried from 

 time to time. At present the temperature selected for the reduction 

 of the observation of the Weather Bureau is the mean of the current 

 air temperature and that of the preceding 8 a. m. or 8 p. m. obser- 

 vation. 



104. Deteimvination of height hy harometric readings.— T^h^ calcu- 

 lation of elevation above sea by barometric readings involves all 

 the principals and encounters all the difficulties of the " reduction to 

 sea level" described above. This method, therefore, of measuring 

 heights is to be used only when others are not possible, and numerous 

 observations should be made to eliminate the very large accidental 

 errors to which the method is subject, 



As the aneroid barometer is used by tourists and others so largely 

 in connection with determination of elevation, some further dis- 

 cussion of the subject is here given. 



105. Determination of heights hy the aneroid. — A reading of a ba- 

 rometer at a single station, without reference to the air temperature 

 and corresponding pressure at some adjacent points whose elevations 

 are known, gives only the crudest possible idea of the elevation of the 

 station, and the neatly constructed little pocket aneroids in morocco 

 cases on sale in shops, having their dials graduated to feet of 

 elevation above sea level, are to be regarded as extremely inaccurate, 

 especially if the scale of altitudes is graduated upon the same metal 

 plate as the scale of inches. In many aneroids the scale of feet is 

 adjustable, and on this account may serve to some advantage for 



Note. — The diminution of gravity as we go from latitude 45° to the Equator causes 

 the mercury in the barometer to weigh less, and hence for a given pressure in the atmos- 

 phere the mercury in the barometer stands higher than it would if the force of gravity- 

 preserved the uniform standard value. Therefore, the farther a barometer is removed 

 from latitude 45° the greater its correction becomes, so that at the Equator a pressure 

 that appears to be 30 inches (at sea level) is really about 29.92 inches. _ 



It should be noticed that when the barometer is thus corrected for its peculiar error 

 due to the influence on it of variations of gravity the pressure that it then gives is the 

 actual pressure of the air at each latitude expressed in terms of an absolute and not 



It is important to remember that the barometric pressure is due not only to the weight 

 of the air but also to the prevailing Avinds, the rapid heating or cooling and consequent 

 expansion or contraction of low layers of air, and to other causes. 



In the mercurial barometer we balance this elastic pressure by weight of quiescent 

 mercury ; a change of the force of gravity will change the weight of the column of 

 mercury without necessarily changing the atmospheric pressure. 



