192 A. Pedler — On the use of tJie Radiometer as a Pliotometer. [August, 



illumiuatiug power of that gas flame would be only 8-6 candles, or about 

 two-tbirds of the illuminating power as measured by the old process. 



It is, I think, evident from these experiments that it would be impos- 

 sible to say that because a Radiometer rotated sixteen times as rapidly with 

 one flame as it did with another, that the former flame possessed sixteen 

 times the illuminating power of the latter ; for it must be seen that in work- 

 ing mth either a very high or very low rate of revolution, there appears to 

 be considerable disturbance due to the friction of the instrument. It is 

 I believe possible, and even probable, that much better results will be ob- 

 tained, by working the radiometer always to a fixed number of revolutions 

 (say about 30 or 40 quarter-revolutions per minute) ; and by altering the 

 distance of the flame until such rapidity is obtained ; in this way the fric- 

 tion of the instrument would be reduced to a constant quantity, and the 

 comparative luminosities could be judged by the squares of the distances. 

 These observations to be conclusive will take a considerable time to carry 

 through, but I hoj^e at some future j)eriod to lay them before the Society. 



Through the courtesy of the Eev. Father Laf ont I have been able to 

 test a second radiometer of a similar construction, having blackened discs 

 of an equal size, which are suspended in the same way on a glass j)ivot. 



I have found that it is a much more sensitive instrument than my 

 own., but that the relative sensitiveness varies according to the velocity of 

 rotation. Some of the comparisons are instructive, and are given in the 

 table below ; No. 1, Radiometer being the one used in the former experi- 

 ments, and No. 2, the instrument belonging to Father Lafont. 





Source of Light. 



Hadiometer, 

 No. 1, 



Quarter- 

 revolutions 



Eadiometer, 

 No. 2, 

 Quarter- 

 revolutions 







per min. 



per mm. 



Standard Sperm Candle, burning 126 grains per hour, .. 

 Gas Jet 13-76 candle power at 10 inches distance, 



„ „ „ „ 10 „ with alum cell inter- 



Tinsfifl 



19 

 169 

 74 

 35-5 



44-5 



17 



18-5 



7 



34 

 221 

 119 

 62 



73 



„ 9-56 



10 „ 



35 



32-5 



9 



It will be seen that these comparative experiments confirm what has 

 been said before that radiometers at high and at low velocity of rotation 

 give somewhat uncertain indications ; for instance at a high rate of revolu- 

 tion No. 1 radiometer is about one-third less sensitive than No. 2 ; at a 



