38 report — 1878. 



sinusoid series accurately accompany each other throughout their whole 

 journey. 



But the event is different if the light encounters an optical agent which 

 acts differently on waves of different periods. Of this kind are the prisms 

 and diffraction-gratings of our spectroscopes. Here Nature herself effects 

 the decomposition which is indicated hy the theory. Waves of different 

 periods are compelled to travel in different directions, and thus the several 

 terms of the sinusoid series appear under the form of lines in the spectrum. 

 The wave-lengths corresponding to each position in the spectrum have been 

 determined with great care, and these when corrected for the dispersion of 

 the air are proportional to the corresponding periodic times, which thus 

 become known. Moreover, the intensities of the lines may be observed, and 

 will give the coefficients to be applied to the corresponding terms of the 

 sinusoid series. Hence by a discussion of the observations we may expect 

 to learn much with regard to the original disturbance caused by the source ot* 

 light. Non-periodic disturbances of the second class will be indicated by 

 continuous spectra, while the other two classes of disturbances will be dis- 

 tinguished by spectra which consist of separate rays : and a careful study of 

 the positions and intensities of the rays may give valuable information as to 

 the periodic time, and sometimes even as to the particular form of the original 

 disturbance. 



Hence in the present state of science it is of importance to facilitate this 

 inquiry as much as possible ; and it is hoped that aid will be given to the 

 student of nature by the Table now published, in which the oscillation- 

 frequencies of the principal rays of the visible part of the solar spectrum 

 have been computed from Angstrom's admirable determinations of their 

 wave-lengths in air, combined with Ketteler's observations on the dispersion of 

 air. Such a table and its accompanying map afford the most assistance that 

 can be given towards the detection of harmonic relations ; for rays that are 

 harmonically related are therein represented in the simplest form that is 

 practicable — in the Table by an arithmetic series of the same type as the series 

 of natural numbers, where the common difference is equal to the first term ; 

 and in the Map by a series of equidistant lines. 



While this theoretic advantage has been the guiding aim of the Committee, 

 they have also kept constantly in view the convenience of observers. A map 

 of Oscillation-frequencies offers peculiar facilities for this, as its red end is 



o 



less extended when compared with its blue end than in Angstrom's map and 

 more extended than in Kirchhoft's. It thus delineates the spectrum with an 

 appearance intermediate between that of a diffraction spectrum and that of 

 a prismatic spectrum, and does not distort either spectrum too much for 

 practical use. It may thus be employed without inconvenience by observers 

 with either of the two great classes of spectroscope. The Committee are 

 accordingly occupied in preparing such a Map to accompany the Table. 



The Committee Q were of opinion that it would prove a boon to observers to 

 have Kirchhoff's, Angstrom's, and the new numbers in reference to each ray 

 brought together in one horizontal line. Before this could be accomplished it 

 was necessary to make a systematic comparison of Angstrom's numbers and 

 maps with Kirchhoff's, and of both with the actual solar spectrum, in order 

 to identify the rays wherever practicable. 



The Committee therefore felt that it was desirable that this work should be 

 undertaken ; and it has been satisfactorily accomplished by Charles E. Burton, 

 Esq., B.A., F.R.A.S., who has made all observations and computations required 

 by the Committee. He has, moreover, inserted (in brackets) in column 2 the 



