﻿EQUIPMENT AND WORK OF AN AERO-PHYSICAL OBSERVATORY. 13 



A line drawn through the underscored values for the dates in question 

 will be found to almost coincide with the path of the storm. Minus 

 values indicate the retrogression of the disturbance. 



Suppose further, however, that, starting with the fundamental equa- 

 tion pv = RT, we were able to follow an}^ given air wave as it is propa- 

 gated in a manner similar to that in which an ordinary sound wave is 

 followed. Lord Rayleigh (" On the Vibrations of an Atmosphere," Phil. 

 Mag., Feb., 1890) has given a numerical example of the high degree of 

 rarefaction necessar}^ before there is a change of sign for a period of one 



hour. In C. G. S. measure, n= and "a" (the velocity of sound) 



obUU 



= 33 X 10*, g = 981 ; then the ratio of the density at a given height to 

 the density at the ground comes out 1/290. This, of course, is for an up- 

 ward wave ; but for the case of " a swaying of the atmosphere from one 

 side of the earth to the other " Rayleigh deduces a period of 23.8 hours. 

 He remarks, however, that the suitability of the value of "a" is very 

 doubtful, and, further, that the suppositions of his paper are inconsistent 

 with the use of Laplace's correction to Newton's theory of sound propa- 

 gation. Moreover, can the heat and cold present in atmospheric vibra- 

 tions be supposed to remain constant? But the near approach of this 

 period to 24 hours he considers to be of more than passing interest and 

 possibly connected with the diurnal and semi-diurnal variations of the 

 barometer. 



Now, the forecaster has to deal with a succession of atmospheric waves, 

 and it is just the gain or loss of heat accompanying the propagation of 

 the slower waves that he attempts to forecast. Therefore we think it to 

 be of prime importance to introduce into our forecast work as far as 

 possible numerical values for atmospheric vibrations. 



Atmospheric Electricity. 



We have thus far discussed the known properties of atmospheric air 

 chiefly in connection with aqueous vapor. There remains another equally 

 important line of research intimatel}^ related to the vapor conditions 

 and likewise of great importance in weather prevision, viz., atmospheric 

 electricity. 



At the outset we advocate the introduction of that unstable and seem- 

 ingly lawless element, the electrical potential of the atmosphere, on the 

 synoptic weather chart. To the graphic representation of air pressure, 

 temperature, and in a crude way, air motion, let us add, although it does 

 seem unpromising, the electrical potential, corrected for temperature, 

 elevation, quantity of vapor present (see further on Exner's experiments 



