METEOROLOGY AND OTHER SCIENCES 735 



been furnished from observations made with the aid of balloons and 

 kites. Such work is in prospect for the near future. 



Formerly, the investigation of the atmosphere in the immediate 

 neighborhood of the earth's surface was thought to be sufficient, 

 but now the opinion is becoming more and more prevalent that, 

 for a deeper insight into the nature of the properties of the atmo- 

 sphere, an accurate knowledge of the higher layers of the atmosphere 

 is also necessary. For example, in the practical carrying-out of cal- 

 culations, the circulation theory presupposes a knowledge of the 

 pressure and density of the atmosphere at various altitudes. 



In the last few years, quantitative observations have led to the 

 conclusion that the temperature of the air in the lower layers de- 

 creases, in the mean, by about 4 C. per kilometer increase in altitude. 

 Further up, the rate of decrease becomes still greater, so that at 

 altitudes between 5 and 10 kilometers the rate of decrease is 

 somewhere about 8 C. per kilometer. 



This is explained as due to the adiabatic expansion of the masses 

 of air with their vertical displacement. A mass of dry air is, through 

 expansion, cooled in rising by about 9.8 C. per kilometer. The pre- 

 sence of moisture in consequence of precipitation cloud-formation 

 causes a decrease in this cooling effect, and the somewhat lower 

 figures derived from observation are thus explained. The influence 

 of this precipitation of water is particularly strong in the lower 

 regions of the atmosphere up to about three kilometers, where 

 the air contains much water-vapor, which gives rise to the forma- 

 tion of huge clouds. This application of physics leads to the con- 

 clusion that probably in the higher, more nearly water-free air 

 layers, the temperature sinks still more rapidly with increasing 

 altitude. This conclusion is, however, not borne out. For, with a 

 decrease of 8 C. per kilometer, the temperature of the air at an alti- 

 tude of about 35 kilometers would sink below absolute zero. In 

 other words, higher up no air could exist. But observations on the 

 heights of meteorites, made with the aid of their glow, as well as 

 upon the heights of auroras, indicate that there is an atmosphere 

 of considerable density at a height of 100 kilometers. The decrease 

 of temperature with altitude must, therefore, be very much smaller 

 than previously assumed. This conclusion, founded on astronomical 

 and physical observations, has been recently confirmed through 

 direct temperature measurements a* high altitudes by Teisserenc de 

 Bort, and Assmann. They found that at great altitudes some- 

 where about thirteen kilometers the decrease in temperature 

 with the height is extremely small, practically vanishing. 



This cannot be otherwise explained than by the assumption that 

 at these altitudes the vertical circulation of the air is, in comparison 

 with other factors, too insignificant to be considered. The factors 



