[ ^ 1 
VI. Corrections to the Computed Lengths of Waves of Light published in the Philosophical 
Transactions of the year 1868. By George Bidbell Airy, C.B., Astronomer Royal. 
Received October 2 , — Read November 16, 1871. 
In a paper communicated to the Boyal Society in the year 1867, and printed in the 
Philosophical Transactions for 1868, I attempted the computation of the Lengths of 
Waves of Light for all the lines which Kirchiioff had observed in the Solar Spectrum, by 
adopting an algebraical formula of the fifth order, and substituting in it for every spectral 
line the value of Kirchhoff’s measure for that line, the numerical bases of the formula 
being derived from Fraunhofer’s and Ditscheiner’ s measures of the wave-lengths for 
six principal lines. Subsequently 1 obtained the means of comparing many of my com- 
o 
puted results with measures of wave-lengths by Angstrom and Ditsciieiner, and I found 
that the discordances were far larger than I had anticipated. I remarked, however, 
“ By means of the comparison there is no difficulty in computing for any other line 
the correction that ought to be applied to the wave-length in the principal Tables, in 
order to exhibit the true wave-lengths on Ditscheiner’s scale, without appreciable 
error.” 
Want of leisure long prevented me from entering upon the examination necessary for 
preparing, in a form easy for applications, the correction which my numbers required. 
Lately, however, I have taken it up ; and I have constructed a Table of corrections to 
the numbers of my Table generally, and I have applied them, both to the general Table 
of wave-lengths and to the values of wave-lengths for the spectral lines of the atmo- 
sphere and several metals (the accurate exhibition of which was, in fact, the first object 
of my computations). I now offer these corrections and corrected numbers for the 
acceptance of the Royal Society. 
The work of comparison and correction was conducted by a graphical process. For 
this, I refer to the diagram (Plate XII.), premising the following explanations: — The 
abscissa-measures are the computed numbers for Wave-lengths in the Philosophical 
Transactions, 1868. The Ordinate-measures are the corrections required to make 
these computed numbers agree with observed wave-lengths. The crosses represent the 
. o 
corrections required by Angstrom ; and the dots represent the corrections required by 
Ditscheiner. 
My first step was, adopting my computed numbers as a line of abscissae, to mark the 
values of the discordances (“ Angstrom — computed numbers” and “ Ditscheiner — com- 
puted numbers”) as ordinates. The points thus determined for the two experimenters 
were placed on the same sheet of paper, but were distinctively marked. The result of 
mbccclxxii. x 
