54 DESCRIPTION OF THE SUCCESSIVE SPECTRA AND THEIR LINES. 



image of that fissure was seen ; its sides were perfectly sharp, and of the same appear- 

 ance as though the grating had not intervened ; on the right and on the left of this 

 image, two equal spaces, completely dark, and, beyond each of these spaces, a series 

 of solar spectra, each having its violet extremity pointing inward towards the central 

 image, and its red extremity outward. Of these spectra, the first on each side is per- 

 fectly insulated, but the violet of the third projects upon the red of the second, and in 

 the same way each successive spectrum is overlapped by those coming after it. These 

 spectra are situated symmetrically on each side of the central image, and with a tele- 

 scope, or when other proper means are used, the fixed lines are plainly visible in them. 



188. In fig. 121 these phenomena are represented: A is the central white image; 

 BCD.... the successive spectra on the right hand ; E F G, the symmetrical ones on 

 the left. 



189. If, now, we measure the distance of any one of the fixed lines, for example H, 

 from the middle of the central image in the successive spectra on one side, we shall 

 find that the distances which separate that centre from this ray in the consecutive spec- 

 tra on one side H' in the first, H" in the second, H'" in the third spectrum are in 

 an arithmetical progression ; A H" is double, AH'" is triple of A H'. The angular sep- 

 aration, measured by the instrument, between a given ray and the middle of the cen- 

 tral image, is termed the deviation. 



190. " FRAUNHOFER has measured thus the deviations of the principal lines, making 

 use of different gratings ; that is to say, of gratings the lines of which were more or less 

 close, and in which the opaque groove was more or less large compared with the trans- 

 parent interval. He has proved, in this manner, that the deviation of the same line, or 

 of the same colour, does not depend on the ratio of the width of the groove to the 

 width of the transparent interval, but on the sum of those two magnitudes; that the abso- 

 lute value of the deviation is in the inverse ratio of this sum; that is to say, that in 

 multiplying the measured deviation by the known sum of the width of a groove and a 

 transparent interval, we obtain a number which is constant for the same ray, whatever 

 may be the grating which we use. FRAUNHOFER has calculated upon exact and numer- 

 ous experiments the values of this constant for the seven principal lines of the solar 

 spectrum, and finds that these values are precisely equal to the lengths of undulations of 

 colours corresponding to those rays, such as FRESNEL had obtained by other processes." 

 (LAME'.) . 



191. Of these phenomena the undulatory theory gives a rigorous account. It 

 farther shows, that the deviation of one of the colours in the first spectrum, multiplied 

 by the sum of the breadth of one of the grooves of the grating, and of its correspond- 

 ing transparent part, is equal to the length of an undulation of that particular colour. 

 This product, therefore, is constant for all gratings, and gives an easy means of de- 

 termining the length of an undulation. 



192. In the same spectrum, also, the deviations of any two colours are to one an- 

 other as the lengths of their respective undulations. For this reason, therefore, the 

 violet ray is situated nearest to, and the red farthest from the optical axis. 



193. From a note attached to M. MELLONI'S memoir (Comptes Rendus, Jan., 1844, 



