DYNAMIC METEOROLOGY. 387 



" Hence the uumerical value of the coefficient of the friction of the air 

 (or viscosity), as found experimentally, can in nowise be used for the 

 computation of the influence of friction in the atmos])here. Rather 

 will the number that for brevity is here called 'friction coefficient,' be 

 much larger than the above-mentioned coefficient. 



" The influence of the earth's surface on the air streaming over it can 

 be easily expressed according to this theory of friction. According to 

 the nature of the earth's surface the neighboring particles of air are 

 either forced to adhere to it or else move over it with greater or less 

 retardation. 



"The density of the air as is well known depends upon the pressure 

 and the temperature. For increasing distance above the earth's sur- 

 face the density diminishes rapidly, and at an altitude of 20 kilometers 

 it has a density of only one-tenth of that which prevails at the earth's 

 surface. The consideration of this circumstance would greatly increase 

 the difficulty of the calculation. I have temporarily ignored the dimi- 

 nution of density at great altitudes. But in the explanation of the final 

 result one would have to take into consideration this point. Since as 

 above mentioned, the density of the air is inap})reciably small at alti- 

 tudes which are slight in comparison with the horizontal dimensions of 

 the atmosphere, therefore the movements that occur at those altitudes 

 can in general exert only a slight influence on the currents in the lower 

 stratum. We can therefore from a dynamic point of view consider the 

 atmosphere as bounded at a moderate altitude by a spherical surface 

 on w^hich the air slides freely. 



" The calculation executed on the basis of this assumption leads to the 

 following result : If ice imagine a definite distribution of temperature over 

 the eartWs surface or over a special portion of it to continue for a long time, 

 then permanent currents of air exist, that can be calculated from the 

 distribution of temperature if it be considered as given. 



" Now the contrast in temperature of the hot and cold zone is the 

 l)riucipal force that, with slight modifications in the different seasons 

 of the year, causes a system of currents that includes the greatest part 

 of the earth's surface. 



"A form of expressing this distribution of temperature that is very 

 convenient for computation consists in the utilization of the following 

 expression : 



T=A+B (1-3 cos2 6) 



In this ^ is the angle that the line connecting any point of the atmos- 

 phere with the center of the earth makes with the axis of the earth so that 

 <)0o —H is the geographical latitude of the point in question. Under this 

 assumption analysis furnishes the following values for the three com- 

 ponents oC the currents at any point of the atmosphere, viz, V for the 

 vertical component taken positive upwards, N the horizontal compo- 



