460 C. PIAZZI SMYTH ON MICROMETRICAL MEASURES. 



APPENDIX IV. 



SEE THE THIRTY-ONE PLATES, FOLLOWING AFTER THE 

 PRINTED MATTER. 



Viz. 29 Plates, each opening to 18x11 inches, and showing what they contain on a 40 

 foot spectrum length from A to H ; 



1 Plate, folding out to 42 x 11 inches, giving approximate and contracted views only, of 

 whole spectra, 26 inches long from A to H ; 



And 1 Plate, folding out also to 42x11 inches; but showing what it contains on a 

 spectrum length of 220 feet from A to H ; a length erroneously printed in earlier pages herein 

 as 120 feet only. 



APPENDIX V. 



ON THE NUMERICAL "WAVE-NUMBER" SPECTRUM SCALE 

 OF ALL THE PLATES. 



The regularly altering number of theoretical Waves of Light at each part of the spectrum, 

 contained in a certain constant unit of length, and called for shortness " Wave-number," has been 

 adopted here, rather than the successive lengths of each of such waves, or "Wave-length," as a 

 practical scale for each of our spectrum pictures — because it gives a most desirable mean between 

 the oppositely exaggerated views of Prisms on one side, and Gratings on the other. And the Inch 

 was at the same time employed as the unit of absolute length referred to, because it is not 

 only British, but nearly Earth-commensurable in the best way ; viz., as the 500 millionth of 

 the length of the Earth's Axis of Rotation ; and it furnishes also a convenient series of numbers 

 for the memory. 



The method is, moreover, in the direction of its increase of figures, combined with the 

 universal European mode of writing from left to right, — exactly suitable to Fraunhofer's now 

 nexpugnable order of lettering the chief lines of the Solar spectrum from Rod A as the beginning, 

 to Violet H as the end ; or from lowest to highest, or Earthly ordinary, to Solar transcendental, 

 temperatures. Hence "Wave-number " always goes conformably from Fraunhofer's A to his 

 B, C, &c, and from his so-called b l to his V 1 , b s , &c. While the "Wave-length" method, 

 with its reversed numbers, leads the Spectroscopists who adopt it — whether in terms of 

 French or English measures of length — to do despite to the memory of their great predecessor 

 by goiug backwards with his letters, while forwards with their own numbers ; or by boginning 

 the visible spectrum with H and ending it with A, in a manner so confusing to the rest of 

 the world, accustomed long since to invariable procedure from A to II ; and also from b 1 to b*, 

 in place of the opposite arrangement so recently introduced by the French metricalists. 



C. P. S. 



ER I! A T A. 

 1'. 410, lim- 29 ; p. 420, line li) ; p. 420, line 12 "/, imo ; p. 421, line 20 ; p. 442, line 19; and back of Plate 



Ixxvi. line 4 ab inm, for 120, read 220, foot spectrum length passim. 

 1'. 442, line 12, for trebling read trebling. 

 I'. 4 18, lim- 9, for bright-lines pectra, read bright-line spectra. 



1'. 443, line 21, i<,r sped pe I spectroscopi 



P. 447, footnote, erase the latter half. 



