386 RECORD OF SCIENCE FOR 1887 AND 1888. 



" From all this it seemed to me worth the trouble to investigate to what 

 results a new treatment of the problem of atmospheric movements in the 

 most general analytical method would lead. 



"The starting point necessarily is the equations of motion of hydro- 

 dynamics, as to whose reliability, so far as is known to me, no manner of 

 doubt has been expressed. The ultimate cause of the movements con- 

 sists in differences of density of air that is subject to the attraction 

 of the earth. 



" The influence of the rotation of the earth, according to the laws of 

 mechanics, can be expressed by a deviating force, so that after its intro- 

 duction the earth can be considered as at rest. Furthermore, frictional 

 forces are to be introduced, since without them the atmospheric currents 

 under the continuous influence of accelerating forces will attain infinitely 

 great velocities. The resistances opposing the atmospheric movements 

 in all cases consist of a series of influences of various origins, and in part 

 also, variable in amount. If a rapidly moving mass of fluid penetrates 

 into a quiet fluid it experiences a sensible resistance almost like that of 

 a solid body moving in the quiet fluid. If, on the other hand, a definite 

 portion of a fluid comes continuously under the influence of a motive force, 

 then " fluid jets'' are formed, or, as they were first called by Helmholtz, 

 "discontinuous currents," which glide through the quiet fluid with rela- 

 tively slight friction.* Two neighboring currents running parallel to 

 each other, with different velocities, will always affect each other in such 

 a way that the more rapid is retarded, the slower current is accelerated. 

 This interchange is much favored when between the two larger main 

 currents smaller side currents are produced by means of local causes, 

 which bring about a partial mixture of the two streams. All these 

 phenomena, especially that last mentioned, indicate in my judgment 

 that we shall take account of the most important resisting influences 

 if we follow the assumption first expressed by Newton, that the mutual 

 influence of the rapidly moving parts of a fluid is proportional to the 

 differen(;e of their velocities. Hence it at once follows that a current, in 

 which all parts move parallel and with equal velocities, experiences in 

 general no friction. On the other hand, wherever currents of different 

 velocity occur near each other the friction has an effect which is especi- 

 ally strong at the boundary of the currents. 



" Since I also adopt the just mentioned law of friction for the atmos- 

 pheric currents, I would expressly designate this as an hypothesis. I 

 thereby in nowise assume that the process is so simple as it is in the 

 experiments for the determination of the coefficient of friction, i. e., 

 viscosity or internal friction, iu which everything is so arranged inten- 

 tionally that the velocities of the fluid remain very small. 



* H. von Helmholtz, Ueber discontiunirliche Flussigkeitsbewegnngeu, Berlin, Mo- 

 natsb., 1868. Reprinted in Helmholtz, Wissenschaftliche Abhandlnngen, Vol. i, p. 

 146. Compare also Oberbeck, Wiedem. Ann., ii, S. 1, ou the same eubject. 



