ATMOSPHERE IN RELATION TO HUMAN LIFE AND HEALTH. 321 



The development of large ice crystals to be attempted in the labora- 

 tory, such as sometimes form on the outside of hailstones. Electrifica- 

 tion, saturation of air, and great rapidity of movement would seem 

 needful. 



The study of the movement of convection currents over a soil or sur- 

 face heated to various degrees above the temperature of the air. Smoke 

 might to some extent show the manner in which the currents rise and 

 the height to which they reach in continuous streams. The effect of 

 wind, at some height above the surface, in promoting or retarding the 

 unbroken ascent of currents might be observed, in connection with 

 such phenomena as showers, tornadoes, and the formation of cumulus. 

 The effect of a calm above a moving air mass might similarly be shown 

 on a small scale. 



The radiation of air and of vapor, separately and together, and mixed 

 in various proportions; also the absorption. Experiment might give 

 information respecting the radiation and absorption of air and vapor 

 in respect of li^ht and of heat in general of various refrangibility. 



The radiative and absorptive power of fog or cloud. Experiments 

 might give useful results both in the laboratory and in natural condi- 

 tions. The effects of dust and smoke mixed with the fog might be 

 observed, and the comparative loss of heat in unit of time by dusty 

 or smoky and dust-free air. 



Observations are needed on the geographical distribution of thun- 

 derstorms and hailstorms, the influence of mountains, forests, and 

 local winds, and on means of forecast and warning against damage. 



The elaboration of plans for the mechanical use of wind power for 

 pumping, irrigation, factories, mills, and traction or propulsion, and 

 for the conversion of wind power into electrical energy. The geo- 

 graphical distribution of wind force, and the areas in which steady, 

 strong winds blow continually or for long periods, need to be ascer- 

 tained in order to place windmills in economically advantageous posi- 

 tions. The heights above the ground at which wind is strongest 

 should also be ascertained. 



Mr. Symons notes that the Hon. R. Abercrombie, in 1875, summed 

 up the results of a study of the oscillations of the barometer in thun- 

 derstorms, and concluded that there are two classes of storms in this 

 country — one iu which the barometer rises, in the other it falls. The 

 rise is always under the visible storm, and the greatest rise is under 

 the greatest uptake, or ascensional column of air. Dr. Fines, of Per- 

 pignan, established a Redier baragraph in 1875, and in a memoir 

 published in 1883 gave reproductions of the traces of several storms. 

 He found that before heavy rain at Perpignan there is usually (1) a 

 decrease of pressure and temperature; (2) with the rain, sudden 

 increase of wind, rapid rise of barometer, and fall of temperature; 

 (3) at the end of the storm rain, reversal of the last three phenomena. 



It appears probable that a fall of the barometer before thunder or 

 hail storms may be caused by the iucreased amount of vapor in the 

 SM 95 21 



