METEOROLOGY AND DEEP TEMPERATURES 461 



may still be read as a model of scientific experiment and reasoning. 

 He discussed the whole question of moisture in the atmosphere, 

 including evaporation and condensation, and although the state 

 of physical science was not then advanced enough to allow him 

 to perfect a theory, he made very important practical advances. 

 He showed that temperature was the governing condition of 

 humidity, that for every temperature there was a certain amount 

 of water vapour which could be held in the air ; when that amount 

 was reached, evaporation stopped, and if the temperature fell, 

 condensation took place, or if the temperature rose, evaporation 

 could be resumed. The higher the temperature, the greater is 

 the amount of vapour which can be taken up. These facts were 

 proved by experiments of the utmost simplicity and completeness, 

 skilfully devised so as to eliminate disturbing causes and described 

 with admirable precision. There are practically only two prin- 

 ciples on which a hygrometer can be constructed for measuring 

 the humidity of the atmosphere (1) the variation in volume or 

 weight of a hygroscopic substance which absorbs moisture from 

 the atmosphere, and (2) the fall of temperature produced by 

 evaporation. De Saussure experimented with both methods, 

 but unfortunately he decided that the hygroscopic was the better. 

 Meteorologists of the next generation reverted to the temperature 

 method, and thus it came about that the dew-point hygrometer 

 and the wet and dry bulb thermometers came into use for the 

 determination both of absolute and relative humidity. The great 

 disadvantage of these is that somewhat elaborate tables have to be 

 used in order to convert the thermometer readings into percentage 

 of relative humidity or absolute humidity, as the case may be. 

 De Saussure adopted as his hygrometer a single human hair fixed 

 at one end and attached at the other to an index finger which 

 moved along a graduated arc as the hair lengthened by absorbing 

 or shortened by losing moisture. The graduations represented 

 relative humidities directly ; but each instrument required to 

 be graduated by direct experiment. The single-hair hygrometer 

 in its original form gave good results in the hands of its inventor, 

 but it was very delicate, required careful handling, and in the course 

 of time the indications became less sensitive. Since the intro- 

 duction of self-recording instruments the use of the hair hygro- 

 meter has revived, and it is now employed in the hygrograph, 

 utilising a number of hairs instead of one, and acting on a lever 

 carrying a pen which records the relative humidity on a moving 

 drum. Thus, after more than a century of neglect, the principle 

 which was so dear to de Saussure has again been adopted, and is 

 likely to continue in use. 



