Defcripiion of a mw Hygrometer and Photometer. 465 



referred to the Intermediate fpace, gives the magnitude of a degree *. I have adopted, as 

 the moft natural and convenient divifion of the thermometer, that of celfius or the centigrade ; 

 fo that each degree of the hygrometer correfponds to the thoufandth part of the interval 

 between the freezing and boiling points. All the liquor is now puftied back into one of the 

 balls, and by applying the flame of a candle the {horter branch is gently bent round till its 

 ball touches the infide of the oppofite tube, and lies -J of an inch below the inflected ball. 

 The fcale, which fliould contain from 50 to 150 degrees, is next divided, f and is faf- 

 tened between the two branches with cement made of rofm and bees-wax. (See a draw- 

 ing of the infirument of the full ftze, Fig. 1. Plate XIX.) 



The inftrument is adjufted by throwing air from the one to the other ball till the liquor 

 refts at the top of the fcale. The lower ball and its annexed cylinder, are covered with 

 thin filk of the fame colour as the upper ball, and a few tfireads are likewife lapped about 

 that part of the tube which it touches. The inftrument is laftly cemented into a piece of 

 wood, either end of which admits a cylindrical cafe that ferves equally to proteft or to 

 hold it. On other occaGons, the hygrometer is inferted into the focket of a round bottom 

 piece where it ftands vertical. 



But this inftrument does not merely point out the drynefs of the air ; it enables us to 

 determine the abfolute quantity of moifture which it is capable of imbibing : for the conver- 

 fion of water into fteam is found to confume 524 | degrees of the centigrade divifion, and 

 evaporation, analogous in its efFe£t may be prefumed to occafion the fame wafte of heat. 

 If, therefore, air had the fame capacity as water, for each degree of the hygrometer it 

 would depofit as much heat as it would abftracb by diflblving the y^Vo P^"^' of i's weighc 

 of humidity. But the capacity of air is to that of water as 1 1 to 6, and confequently it 

 would require in that proportion a greater evaporation to produce the fame efFe£t. We 

 may hence conclude, that, for each hygrometeric degree, the air would require y x yj'^-g 

 or t^Vt P^''* ^y weight of water to effe£l: faturation. 



Striftly fpeaking the degrees marked by this hygrometer do not meafure the drynefs of 

 the air at its aflual temperature, but only its ftate of drynefs when cooled down to the 



• When a ftandard inftrument \% once conftrufted, others can be more eafily, though not quite ft ex- 

 aftly, graduated by comparifon. In a dry room, they may be ufed for the occafion as hygrometers, and 

 the magnitude of a degree in eaeh thus determined. If the weather is clear and fettled, it would be pre- 

 ferable to convert them into photometers, and expofe them exclufively to the direft rays of the fun, when 

 moderately elevated above the horizon. Other expedients will fuggeft themfelves. 



f If the temperature of the room happens to vary during the obfervation, it muft be allowed for. 

 When the feafen permits, the upper ball may be covered with fnow, which will render the graduatioa 

 more certain. There is always a fmall correftion to be made for the inverted pofition of the cylindrical 

 refervoir in the finifhed ftate of the inftrument: let n denote the width of that cylinder compared with the 

 bore of the tube, and a ~ the length of too degrees of the hygrometer in Englifli inches ; this fpace muft 



be diminilhed, before divifion, by the th part. The formula iseafily derived. 



J This has been ftated, but I expeft foon to ascertain the quantity with more precifion. 



ftandard 



