VOL. LXX.] PHILOSOPHICAL TRANSACTIONS. 743 



But on trying this experiment various times, placing the thermometers at differ- 

 ent distances from the flame of the lamp, and making the proper calculations 

 ao-reeable to those distances, it appeared, that the intensity of the heat did not 

 decrease exactly in the duplicate proportion of the distances from the flame of 

 the lamp, but showed a very odd irregularity. It seemed to decrease faster than 

 the duplicate proportion of the distances for the space of 2 inches and a half or 

 3 inches, after which it decreased much slower. Whether this effect may be at- 

 tributed to some different state of the air's purity at different distances from the 

 flame of the lamp, or to the vapours proceeding from the flame, he could not 

 determine. 



The above-mentioned experiments gradually induced him to try the effect of 

 the light of the sun and of a lamp on thermometers whose balls were painted 

 with different colours. Dr. Franklin's experiment with the pieces of cloth set 

 upon snow that was exposed to the sun is very well known. The doctor found, 

 that those pieces of cloth, whose colour was darker, sunk deeper in the snow 

 than the others, by which it appears, that they became hotter. Mr. C.'s view 

 was to examine those different degrees of heat imbibed by different coloured sub- 

 stances with precision, to observe if they kept any proportion to the spaces occu- 

 pied by the prismatic colours in the prismatic spectrum, or if they followed any 

 other discoverable law ; but those attempts met with many difficulties, the 

 greatest of which was the choice of colours. The water colours that are com- 

 monly used, as carmine, sap-green, &c. are of so different a nature from each 

 other, that when the balls of the thermometers were painted with them, their 

 surfaces were not equally smooth, which occasioned great difference in the effect ; 

 for two thermometers, whose balls had been painted with the same colour, but 

 the paint laid smoother on one than on the other, showed different degrees of 

 heat when they were both exposed to the rays of the sun. 



He attemped to make thermometers with tubes of differently coloured glass; 

 but when a ball was formed with any of those tubes, the substance of the glass 

 in the ball, being much thinner than in the tube, differed very little from clear 

 coloured glass. To include the thermometers in close boxes, in which the rays 

 entered through coloured glasses, was also found ineffectual ; not only because 

 the colours so transmitted were far from being homogeneous, but especially be- 

 cause some of those glasses are much more opaque than others, even of the 

 same colour. 



The least ambiguous method therefore, was that of painting the balls of the 

 thermometers with water colours, taking care to lay them as equally smooth as 

 possible. In this manner Mr. C. repeated several experiments, using sometimes 

 a dozen of thermometers at once, whose balls were painted with various colours, 

 and were exposed to the sun ; and from a vast number of experiments, and after 



