NEW HYGROMETER OR DEW POINT INSTRUMENT. 



21 



instrument, recorded in this paper, and which may be recommended to others. 

 The proper mode will be, first, to note the instant of decided appearance of mois- 

 ture, and mark the temperature indicated by the thermometer ; then, to stop the 

 farther reduction the instant this decided deposition is observed ; next, to mark 

 the instant of disappearance, and the corresponding temperature ; and, lastly, to 

 take a mean of the two observed temperatures. In observing the point of appear- 

 ance, I notice the first apparent indication of moisture, and then see whether, 

 immediately on farther reduction of temperature, this indication becomes quite 

 distinct. I then take the former indication as marking the temperature of depo- 

 sition. In this way we can easily distinguish between a true and a false indica- 

 tion. Every one, however, as before observed, must be permitted to make his 

 observation in the manner which he thinks best. 



Since this paper was first drawn up, I have made a series of observations, with 

 the view of comparing the indications of the instrument with those afforded both 

 by Daniell's hygrometer, and by Dalton's method of transference, which last is 

 generally admitted to be the most trustworthy, although certainly very far from 

 being the least troublesome.* These observations have supported the view that 

 Daniell's hygrometer yields indications somewhat too high, although not to 

 quite so great an extent as follows from Mr Adie's observations. They shew that 

 the tendency of the instrument described in this paper is rather in the opposite 

 direction, and that its usual indications are a very little too low, but that this de- 

 viation is less on an average than 1°, and therefore quite within such limits as 

 fully to justify reliance on its results. The following table contains the observa- 

 tions ; those by Dalton's method being stated at the actually observed dew point, 

 and the others in excess or deficiency, with reference to Dalton. 



1854. March 



11. 



12. 



13. 



14. 



15. 



16. 



17. 



18. 



19. 



20. 



21. 



22. 



23. 



24. 



26. 



28. 



29. 



30. 





Dalton, . 



42J 



374 



37i 



40f 



40 



40 



35 



33 



32 



36# 



34 



42 



35J 



31 



35} 



39f 



45 



43 





Daniell, . 



-1 



+ 1} 



+ i 



= 



+ 2} 



+ i 



-1 



+ i 



+ i 



+ i 



+ 11 



-U 



+ ! 



+ i 



-2 



+ li 



+ 1* 



- i 



Average +0°29 



New Instru- ) 

 merit, J 





-1 



+ 2 



- f 



- 1 





-2J 



-2| 



-1* 



-2 



+ i 



-2 



-2} 



- 1 



-2J 



- i 



- i 



+ 1* 



Average - 0° - 95 



We thus have, in a series of observations of dew point temperatures, varying 

 from 32° to 45°, Daniell's hygrometer, in eighteen observations, giving twelve of 

 them in excess or above the dew point by Dalton's method, five of them in de- 

 ficiency or below the point referred to, and one of them shewing equality. The 

 average of the whole, however, gives only 0°-29 F. in excess. I cannot, however, 

 help saying, that when I got an observation by Daniell in deficiency, and repeated 



* In support of the great trustworthiness of Dalton's method, I may appeal to the opinion of 

 Professor James Forbes, 2d Report on Meteorology to British Association, 1832; and to Dr Thom- 

 son's opinion, there referred to. It is unnecessary to say that I had previously compared the ther- 

 mometers, and that I made allowance in the comparative table drawn up, for a slight difference ob- 

 served in their indications. 



