202 Mr. S. P. Langley's Experimental Determination of 



i. e. to determine its deviation by trials with the bolometer at 

 intervals sufficiently close to avoid the possibility of missing 

 it altogether. According to Briofs formula, the deviation 

 should be 45° 21/, and in the preliminary search the circle 

 was accordingly set to this reading. Beginning at this point, 

 and exposing the bolometer at every five minutes of deviation, 

 it was found that the maximum effect was obtained nearer 

 45° 15'. The approximate position having thus been found, 

 the slit Si was narrowed to 2 millim., and the following 

 measurements taken, the horizontal line giving the mean 

 results of a series of thirty exposures of the bolometer, as it 

 moved through the spectrum. 



Table I. — Determination of the Refrangibility of 

 feeble Heat-rays. 



Prismatic deviation 



Means of galvanometer-readings. 



45° 02' 



4-6 



45° or 

 5-6 



45° 10 r 

 6-0 



45° 15' 



5-8 



45° 20' 



2-7 



The maximum reading at 45° 10 / corresponds to a coinci- 

 dence of the 2 millim. bolometer-aperture with the 2 millim. 

 invisible image of the slit whose position is sought. From a 

 subsidiary curve drawn through the points whose coordinates 

 are respectively (a=45° 02',y=4-6), (#=45° 07', y=5'6), 

 (^=45° 10', y = 6 # 0), &c, it was concluded that the deviation 

 of rays whose wave-length is 1^'767 is 45° 10'; and each 

 point in this determination being obtained from the mean of 

 five observations, the result is partly free from irregularities 

 caused by changes in the state of the sky, and minute 

 instrumental variations from extraneous causes, which here 

 become of great relative importance owing to the feeble heat 

 measured. 



Subsequent determinations like the preceding gave for the 

 deviation of the same ray 45° 06 / and 45° 07'; and from a 

 consideration of all, the deviation adopted was (instead of 

 45° 21/, as given by Briot's formula) 45° 08', corresponding 

 to a refractive index of To549. 



By means of measurements like the one described above, 

 the deviations of various obscure rays of known wave-lengths 

 were determined. The indices of refraction were then com- 

 puted by the usual formula n= - where a=62° 



sin ^ a 



34' 43". The results are contained in the followng table, 

 where, however, only the final results of sue cessfu idays are 



