On Solar Radiation. 203 



diation from the scale of the thermometer. I therefore have 

 found it best to estimate the power of the sun's rays as a 

 function of the time, in the manner directed by Sir John 

 Herschell for the actinometer. In this manner I have found 

 the effect during one minute to vary from 3° of Farhenheit 

 to 7°. But my experiments have not as yet been satisfactory. 

 I find also that sometimes the moisture in the air contained 

 in the case is deposited like dew on the inside, which of 

 course interferes with the passage of the sun's rays ; perhaps 

 this may be neutralized, by leaving the door a little open, 

 and the heating of the air in the case may perhaps be pre- 

 vented in the same way. 



The only objection to the use of common thermometers is, 

 that they are not sufficiently delicate in their indications ; 

 but perhaps it may be found, that small tubes with balls of 

 an inch in diameter, may answer the purpose in the manner 

 of air thermometers, the end of the tube being immersed in 

 some fluid at the time of observation. 



If the naked thermometer is exposed, it may be conveni- 

 ently done by putting the scale into a small pair of spring 

 forceps, the forked end of which is worked into a conical 

 screw, which may be readily fixed into a tree, a post, or any 

 piece of wood at hand, so as to keep the thermometer at 

 some distance from any thing else. It is known that large 

 blackened masses of metal, exposed to the sun's rays, acquire 

 in India a high degree of heat ; but what it may be I have 

 never had an opportunity of observing. It has been a 

 guard room jest to remark, that the guns on the ramparts of 

 Fort St. George were hot enough sometimes to fry a beef- 

 steak upon. I do not suppose that the experiment was ever 

 put to the test; but I remember that they were often much 

 too hot to bear the hand upon. The temperature thus 

 tested could not however be very high, for the boiling point 

 of spirits of wine, or 180° Farhenheit, is a great deal too hot 

 to bear the finger in, and the conducting power of metal 



