314 SCIENTIFIC RECORD FOR 1884. 



nual maximuiu takes place ou the 25tii of July, and the minimum ou 

 the 20tli ofJanuary. He finds a negative evaporation during the win- 

 ter months, which must be an error in his instruments unless, iudecdj 

 the deposit of fog and dew is extraordinarily heavy. He finds the 

 ratio between the evaporation in perfectly free spaces and that from 

 an evaporimeter established in an ordinary thermometer shelter to 

 A'ary between one and three. {Z. 0. G. 31., xvii, p. 242.) 



200. [The relative amounts of evaporation from snow, ice, fresh and 

 salt water, in full sunshine and cloud, calm and wind, is a matter that 

 still needs to be well determined.] 



201. Dr. J. Hann, in his new edition of Jeliuek's Instructions for tak- 

 ing Meteorological Observations, has introduced tables for the con\enient 

 computation of the quantity of vapor contained in any atmospheric strata 

 as well as its influence on barometric hypsometry. He first states that 

 the formula published by him ten years ago still represents more recent 

 ob>ervati()us in the higher atmospheric strata. He then applies this 

 equation to the computation of the probable moisture at elevated sta- 

 tions in Austria, Italy, Ceylon, Java, England, Switzerland, and shows 

 that wherever we have observations the prediction by means of this 

 formula is well verified, so far as regards monthly' and annual or even 

 weekly averages. 



This reliability justifies us in using the computed values of vapor ten- 

 sion when actual observations are missing. The etfect of this moisture 

 upon the barometric computation of altitudes amounts to as much as 17 

 meters for an altitude of 2,000 meters in a moist, warm climate like Cey- 

 lon, but oidy G meters in a cooler climate, such as Switzerland ; in both 

 cases the correction for vapor gives higher altitudes. 



The effect of vapor in diminishing the density of the air, thus giving 

 rise to ascending currents, is sensible, but very slight in comparison with 

 the effect of ordinary changes of temperature; thus the entire vapor in 

 a mass of air 2,000 meters deep is only equivalent to the influence of a 

 change of 1°.3 C. in temperature, and, as the temperature changes may 

 easily amount to ten times this figure, they are relatively much more 

 important. 



By integrating the i^roper equation Hann deduces the formula for 

 computing the weight of water in a column of air of any height, whence, 

 of course, the amount contained in the whole earth's atmosphere can be 

 deduced. {Z. 0. G. M., xix, p. 128.) 



202. W. K()i)peu describes the j)rocess of growth of a cumulus cloud 

 in the warmest part of the day. He finds that the highest point moves 

 forward more rapidly than the lower part of the cloud --, that this disap- 

 pears little by little, while a lower portion of the cloud has risen up to 

 the same height ; this again in its turn disappears, to be replaced by the 

 upgrowth of a next lower portion, and so on. He explains this as due 

 to the fact that the cumulus is really the head of an ascending mass of 

 moist air. and is driven along by winds that have a greater velocity 



