260 Dr. J. Miiller on the Determination of the Length of 



front. of the slit, and the flame thereof was coloured red by means 

 of lithium. 



The telescope was now so directed that the cross-lines were 

 brought to bear on the red lithium line of the first spectrum to 

 the left, and then on the red lithium line of the first spectrum 

 to the right, the reading of each of these positions being taken. 



Designating the red lithium line of the first side-spectrum by 

 Li u 1, the position of the nonius was for 



Li a 1 right, 72° 41', 

 Li a Heft, 55° 28'; 



consequently the angle of deviation of each of the red bands 



Li a 1 from the centre of the image of inflection gives # = 8° 6> 36 , -5. 



For the length of wave, A,, of the red lithium light we obtain 



X = b sin x = 0-002 . 0-149679 = 0"'*00029936. 



The deviation of the red lithium band Li a 2 from the centre 

 of the inflection-image was measured in the same manner, and 

 its value, y, was found to be =17° 27'*5. In conformity there- 

 with, the length of wave of the red lithium light is ascertained 

 to be 



X=h !%£ = Q.0Q2 sin ( 17 ° 2y '- 5 > - =000030000. 



2 £ 



As a mean of these two measures, we therefore obtain as the 

 length of wave of Li a the value 



\=0'"-0002997, 

 or 



X=0-0006733 millimetre. 



A similar measure was next carried out for the yellow sodium 

 line. The result arrived at was for 



Na«l a?= 7 32-5, and therefore \=00002626 



Na«2 ?/=15 12-5 „ ,, \=00002623 



Na«3 *=23 8 „ „ X=00002619 



Consequently the mean is \= 0*00026227 



or \ = 0*0005918 millim. 



From Fraunhofer' s measures we obtain for the dark line D of the 

 solar spectrum, which, as we know, coincides with Na u, a length 

 of wave of 0*0005888 millim. The result of my measures agrees 

 therefore satisfactorily with those of Fraunhofer, when one takes 

 into consideration that the division on the goniometer does not 

 admit of an accurate reading of less than 1 minute. The result 

 arrived at by me differs from that of Fraunhofer by about \ per 

 cent., and the extreme values obtained by Fraunhofer himself 

 differ by | per cent. 



