350 Mr. Louis Bell on the Absolute 



The fifth group consists of the remaining lines of carbon, 

 together with the lines which were at first attributed to carbon 

 by Liveing and Dewar, and were afterwards shown by Hartley 

 to be produced by silicon. The author's analysis leads him to 

 think that the latter lines belong both to carbon and silicon, 

 and are due to a common element in those substances. The 

 wave-lengths of the whole group, when multiplied by 5/4, 

 become wave-lengths of lines of the water-spectrum ; and 

 when multiplied by 25/16 give a group of wave-lengths which 

 satisfies the criteria II. for b as it occurs in hydrogen as com- 

 pletely as any of the preceding groups so far as the water- 

 spectrum is concerned, and somewhat more completely than 

 them so far as the observed lines of hydrogen and oxygen 

 are concerned. The factors 5/4 and 25/16 are interpreted to 

 indicate the condensation of b, as in former cases. It is 

 noteworthy that 25/16 x 2506-6 = 3916-5 = 2/3 x 5874% the 

 wave-length of the helium-ray D 3 . 



The conclusion is that carbon contains c in the same state 

 as that in which it occurs in oxygen and in magnesium, and b 

 in four different states : (1) in which it is less condensed than 

 in hydrogen in the ratio 3:5; (2) in which it is in the same 

 state as in free hydrogen ; (3) in which it is in the state 

 in which it occurs in the hydrogen in water-vapour, more 

 condensed, in the ratio 5 : 4, than in free hydrogen ; and 

 (4) in which it is still more condensed, in the ratio 25 : 16, 

 than in free hydrogen. In the second and third of these 

 states b occurs also in magnesium. 



XL VII. The Absolute Wave-length of Light.— Part II. By 

 Louis Bell, Fellow in Physics in Johns Hopkins Uni- 

 versity*. 



[Continued from p. 263.] 



THIS continuation of my previous paper contains the 

 angular measurements and the details of the measure- 

 ment and calibration of the gratings, together with the final 

 results. In addition I have endeavoured to point out the 

 probable sources of error in some recent determinations of 

 absolute wave-length. 



Angular Measurements. 

 In my former paper (this Journal, March 1887) the work 

 with glass gratings was described in detail, so that it will only 

 be necessary to summarize it here. 



* Communicated by the Author. 



