PRECIPITATION ON MOUNTAIN SLOPES— POCKELS 8q 



that on the other hand, the larger drops that carry down with them- 

 selves the water condensed in the lower strata of clouds will fall 

 at a relatively slight horizontal distance. But now, as the numer- 

 ical computation sKows, the lower cloud strata contribute relatively 

 far more to the condensation than the upper clouds; therefore, the 

 influence of the horizontal transport will not be so very large, at 

 least with moderate winds. Moreover, this influence does not affect 

 the total quantity of precipitation caused by the flow up the moun- 

 tain side, but only its distribution on the mountain slope and it 

 consists essentially in a transfer of the location of maximum pre- 

 cipitation toward the mountain. In this sense, therefore, we have 

 to expect a departure of the actual distribution of precipitation 

 from that which is theoretically given by the computation of W 

 as a function of x, according to equation (14). This departure will, 

 under otherwise similar circumstances, be considerably larger in the 

 case of snowfall than in the case of rain. 



As concerns the upper limit y' , which is to be assumed in the 

 integration of equation (14) in order to obtain the total quantity 

 of precipitation falling upon a unit of surface, we have to substitute 

 for y' that altitude at which condensation actually ceases in the 

 ascending current of air. Theoretically, if from the beginning 

 adiabatic equilibrium prevails up to any given altitude, then the 

 condensation brought about by the rising of the earth's surface 

 must also extend indefinitely high, even to the limit of the atmos- 

 phere, since the vertical component of velocity diminishes asymptot- 

 ically toward zero. But practically, our solution of the problem 

 of flow probably no longer holds good for very high strata, and cer- 

 tainly the assumption of adiabatic equilibrium does not hold good ; 

 but even if the latter were the case, if therefore, the ascending cur- 

 rent carried masses of air from the surface of the earth up to any 

 given altitude, still, in consequence of the increasing w r eight of the 

 particles of precipitation carried up by the ascending current on the 

 one hand, and the increasing insolation on the other hand, an upper 

 limit of cloud must be formed 2 



We will therefore assume as given some such upper limit of clouds 

 at a definite altitude, and for simplicity will assume this to be the 

 same everywhere. The value of this altitude, y\ is the upper limit 

 of the integral (14). However, the altitude assumed for y' if it is 

 large, namely, many thousands of meters, can have only a slight 



2 W. von Bezold: Sitzb. Ber. Akad. Wiss., Berlin, p. 518, 1888, and p. 303, 

 1891. 



