Daniell on a new Hygrometer. 137 



perature of the vapour had risen to 55°, the barometer still con- 

 tinuing at 30.12. 



Towards the end of the eclipse, when the disc of the sun was 

 quite free from clouds, I had a good experiment upon the 

 power of the sun's rays. I directed, upon some gun-powder, the 

 focus of a small lens, which had very little more than sufficient 

 power to ignite it in clear sunshipe ; it was exactly at three- 

 quarters past two that it took fire, 55 minutes from the greatest 

 obscuration, and thirty-two minutes before the end of the 

 eclipse. This effect would appear to be very much greater than 

 the corresponding one produced upon the thermometer ; and I 

 could not but observe that the impression upon the feeling was 

 likewise much more than would have been expected from the 

 fall in that instrument. This latter sensation was probably 

 owing to the great increase in the radiation of the body, as was 

 the former effect to the decrease of absorption, both surfaces 

 being instantly sensible to the diminution of the sun*s energy, 

 while the air only felt the influence of bodies which had been 

 primarily affected *. 

 ' . ■ ■ III 



• A practical illustration of the unfitness of hygroscopic substances for 

 the construction of accurate 'instruments, may be found in the Annals of 

 Philosophy for the last month. It consists of the following memorandum, 

 appended to the Meteorological Journal, so accurately and ably kept by 

 Mr. Howard. 



" The mean of the hygrometer is, for the latter serenteen days of the 

 month, 79.35, but the mean deduced for the like space, from a new one now 

 substituted for it, is 64.68 ; it appears, therefore, that the old one, the in- 

 accuracy of which has been heretofore stated, will require a deduction of 

 fourteen degrees from its later results. It appears to have been employed 

 from about midsummer, 1819, and the error, which has apparently arisen 

 from the stretching of a too slender piece of whalebone, has probably in- 

 creased from that time to the present.'' 



k 



