84 A New Form of Micrometer for use in Spectroscopic Analysis, 



I add the results of a few measurements made with the new 

 micrometer, designed to test the accuracy of which it is capable. 

 Twenty different readings of the point at which there is coinci- 

 dence of the lenses gave the following results : — 



8-36 8-29 8-25 8-32 8'36 831 8-38 8-37 8-34 8*37 



Mean 8*35 



8-37 8-41 8-26 8-28 8-31 8-35 8-21 8-35 8*38 8'35 



Mean 8-33 



To bring the sodium-line into coincidence with the thallium- 

 line the following displacements were necessary: — 



21-90 21-90 21-89 21-90 Mean 21-90 

 21-92 21-93 21-94 2188 Mean 21'92 



I have generally taken the mean of four readings as giving 

 the true position of a line. 



The light emitted by burning magnesium examined with the 

 spectroscope shows, besides the bright lines corresponding to b, 

 a series of bands nearer to F. I had measured the wave-lengths 

 of these bands with the greatest accuracy I could attain by the 

 use of Browning's 6-prism automatic spectroscope in conjunction 

 with a bifilar micrometer-eyepiece. The reference-lines employed 

 were the bright lines of the spark due to zinc, cadmium, iron, 

 and air, and certain Fraunhofer lines. The following results 

 were obtained for the first and second bands : — 



First band. 5006*5 from Fraunhofer lines. 

 5009-0 from zinc-line 4924 



5008*0 from air-lines < kqqk 

 5006-5 from Fraunhofer lines. 



Rnrvr** ■ r /5002 



5007*7 from air-lines < 5005 



Second band. 4995*0 from Fraunhofer lines. 



{5002 

 5005 



4997'4 from air-lines < r( ^rZ 

 \5005 



The numbers finally adopted are given below for the whole 

 nine bands; and with them are compared the results recently 

 obtained with the new micrometer. The reference-line employed 

 for these last measurements was the bright line corresponding 

 to b present in the same spectrum. 



