1875.] A. Pedler — On the use of the Ruliometer as a Pliotometer. 191 



of the dark heat rays by the cell of alum solution ; there is, of course, a 

 certain loss of light by the use of the cell of liquid, and, in order to ascer- 

 tain the amount of this, some experiments were made with the Bunsen pho- 

 tometer, which gave as an average of several determinations, that a total 

 loss of 8'6 per cent, of light occurred in the jmssage thi-ough the alum cell. 

 This shews then that from two-thirds to three-fourths of the mechanical 

 effect in the radiometer, is, in the case of a coal-gas jet, due to dark heat 

 rays, and not to light rays at all. 



In this series of observations we again see a general agreement of the 

 average of the results at the varying distances, with the law of inverse 

 squares ; but when the experiments are examined in detail, it will be found 

 that there are somewhat serious divergences from the theory, and that the 

 discrej)ancies are still greater when the rapidity of the rotation is compared 

 with the illuminating power of the gas-jet on the separate occasions. The 

 differences between the illuminating powers and the velocity of revolution 

 are more marked in this table than in the last, and I cannot help thinking 

 that the value of the radiometer as a photometer has been much overesti- 

 mated. 



Apparently from these experiments, Avhich, however, are fewer in num- 

 ber than I should have wished, we must either beheve that the old jDrocess 

 of jihotometry cannot be thoroughly depended upon, or that the radiome- 

 ter does not yield absolutely constant results. Dm*ing the experiments I 

 have made, there have been some instances in which I obtained some very 

 cui'ious alterations in the rapidity of rotation of the radiometer with 

 scarcely any apparent alteration in the external circumstances, these, how- 

 ever, I cannot at present satisfactorily explain. Since making the above 

 experiments I have received a paper by Mr. Crookes (published in the Pro- 

 ceedings of the Royal Society, vol. XXIV, p. 27G) in which he proves, that 

 different parts of the S2:)ectrum have very different actions on the rotation of 

 the radiometer ; and as the light of coal gas varies from white to yellow, 

 it is jjossible that the origin of the discrepancies between the radiometer 

 and photometer may be due to the differences in colour of the light. There 

 are, however, other photometrical instruments such as, " the Sugg Jet 

 Photometer," and it is my intention to compare this instrvimcnt with the 

 radiometer. During these experiments, I have of course not neglected to 

 test the radiometer with the standard sperm candles, and even here I 

 obtained somewhat discordant results. As an average of my determina- 

 tions, I found that my radiometer, when placed at a distance of 10 inches 

 fi'om a candle burning 120 grains of sperm per hour, made IS'2 quarter 

 rotations per minute. The avei'age of the radiometer under the same cir- 

 cumstances with a gas flame of 12-candle jxjwer was 15702 quarter-re volu- 

 tion 3 according to the radiometer, thercfure, imdcr these conditions, the 



