TABLE OF STANDARD WAVE-LEN T GTHS OF THE SPECTRAL LINES 555 



of the grating used. The spectrometer was poor, and the errors of 

 the grating undetermined. 



Miiller and Kempf used four gratings, evidently of very poor quality, 

 as they give results which differ 1 in 10,000. 



The result of Angstrom was a marvel at the time, but the Nobert 

 gratings used by him would now be considered very poor. Taking 

 Thalen's correction for error of scale, we have for the mean of the E 

 lines 5269-80, which gives, by my table of relative wave-lengths, D = 

 5895-81. It is rather disagreeable to estimate the relative accuracy 

 of observations made by different observers and in different countries, 

 but in the interest of scientific progress I have attempted it, as follows : 



o Wt - 



Angstrom 5895-81 1 



Miiller and Kempf 5896-25 2 



Kurlbaum 5895-90 2 



Peirce 5896-20 5 



Bell 5896-20 10 



Mean, 5896-156 in air at 20 and 760 mm. pressure. 



This must be very nearly right, and I believe the wave-length to be 

 as well determined as the length of most standard bars. Indeed, fur- 

 ther discussion of the question would involve a very elaborate discus- 

 sion of standard metres, a question involving endless dispute. I think we 

 may say that the above result is within 1 in 100,000 of the correct value, 

 which is very nearly the limit of accuracy of linear measurements. This 

 should be so, as the probable error of the angular measures affects the 

 wave-length only to 1 in 2,000,000, 10 and hence nearly the whole accuracy 

 rests on the linear measures. 



RESUME OF PROCESS FOR OBTAINING RELATIVE WAVE-LENGTHS 



1. Determination of about 20 lines in the visible spectrum by coin- 

 cidences by Koyl." 



10 Is not a grating and spectrometer thus the best standard of length, and almost 

 independent of the temperature? Gratings of 10 cm. length can now be ruled on 

 my new engine with almost perfect accuracy, as seen in the calibration of Grating 

 IV in Dr. Bell's paper, and it seems to me the time has come for their practical use. 



11 These observations of Mr. Koyl were finally given no weight, on account of the 

 inferior apparatus used. They serve a useful purpose, however, as checks on the 

 other work. 



