388 SHORT MEMOIRS ON METEOROLOGICAL SUBJECTS. 



It' we compare the mean pressure before, during, and after the rain 

 with the annual mean pressure, we obtain the deviations : 



Since the mean daily range of the barometer in Batavia amounts to 

 2.7 millimeters, it follows that the influence of the rain is insignificant 

 in comparison therewith, and we have a plain confirmation of the state- 

 ment that in the tropics the rain does not disturb the diurnal oscillation 

 of the barometer. 



I make the following extract from a letter of Blanford, to whom I 

 communicated the result of my computation of the Batavia observations 

 with the inquiry as to whether he could perhaps refer to opposing obser- 

 vations in India: "The result you have obtained from the Batavia obser- 

 vations, viz, that the barometer is lower before rain than during or after 

 it, is completely in accordance with my own observations at Calcutta." 



I consider the slight increase of pressure here demonstrated during 

 and after the rain not as a primary efl'ect of condensation of vapor, 

 which therefore would be operating in a contrary direction to that as- 

 sumed by Espy and Eeye, but only as a secondary effect due to the 

 cooling of the lower strata by the rain-drops descending from higher, 

 colder strata and by evaporation. Perhaps, also, the impact of the fall- 

 ing water and the air dragged down with it contributes somewhat td 

 the initial rapid rise of pressure, but a reason to be given later can also 

 cause the barometer to rise somewhat before the rain. 



We believe ourselves now to be justified in the conclusion that the con- 

 densation of aqueous vapor in the atmosphere has no appreciable influ- 

 ence on the change of atmospheric pressure. The low barometers of the 

 storm centers can therefore not be explained by the precipitations, 

 although these latter may probably be an occasion for their develop- 

 ment, to which point we will subsequently return. 



We will next consider the solution of the contradiction that exists 

 between the theory which assumes a considerable fall of pressure under 

 a precipitation and the above-given results of observation, which re- 

 veal nothing of the kind. 



The computation of Reye on the magnitude of the influence on the 

 pressure of the condensation of a given quantity of atmospheric aqueous 

 vapor is based upon the following train of thought : If the vapor in a 

 mass of moist air is condensed, the expansive force of the air, and there- 

 with its volume, is at first diminished, but the air is at once so strongly 

 warmed, and thus again expanded, by the liberated latent heat of the 

 condensed vapor, that the original contraction is many times exceeded. 

 Reye computes that the expansion at 30° C. is somewhat more than five 

 times the initial contraction : at — 10° C. it is somewhat more than six 



