406 Transactions of the Society. 



pinhole • 3 mm. diameter in a piece of thin brass placed in front 

 of the condenser of an optical lantern, arranged so .as to project a 

 parallel beam of light. The angular width of the bright object was 

 therefore only about 4 per cent, of the angular diameter of the 

 lirst dark ring, and the uncertainty of setting would therefore be 

 considerably less than this. The light was required as uearly as 

 possible monochromatic and very intense for obvious reasons. 

 These results were obtained by using an electric arc, the positive 

 carbon having its core bored out and replaced by finely ground dry 

 salt ; this carbon was placed below the negative carbon. This 

 arrangement gives an intense sodium light, consisting mainly of 

 the D-lines, l)ut containing also the other pairs of lines emitted by 

 sodium in the arc. The light from the latter was too feeble to 

 sensibly modify the diffraction figure formed by the much prepon- 

 derating D light present. The accuracy of the micrometer eye- 

 pieces was well within the errors entailed in setting the cross-wire 

 on the rings. 



The measured diameters were reduced to give the constant for 

 each ring for comparison with the theoretic values. Of the first 

 dark ring 7 series of 4 measurements were taken ; of the first 

 bright ring 2 series of 4 measurements each ; and of the second 

 dark ring, 3 series of 4 measurements each. The mean values of 

 these determinations are given below with Airy's values for 



comparison : — 



Airy. Present Experiments. 



1st dark ring 1-22 1-21 



Ist bright ring 1-64 1-68 



2nd dark ring 2-23 220 



The outcome of these measurements is to show practical agree- 

 ment between theory and experiment ; this is especially so in the 

 case of the first dark ring, for which the possible precision of 

 measurement is certainly the greatest. The error is greatest in the 

 case of the first bright ring, but here amounts only to 2^ per cent. 



