Second Spectrum of Hydrogen. 

 Table B (continued). 



349 













Older determi- 



Series I. 



Series II. 



Mean. 



J. 



Remarks. 



nations. 



V. 



S. 



4599-3 



4597*8 



4598-5 



1 









80-0 



81-3 



45806 



4 



Hazy on both sides. 



4581 



45761 



73-3 



73-2 



4573-2 



2 









67'4 



683 



4567-8 



2 









62-7 



63-1 



4562-9 



1 









575 



58-3 



4557-9 



1 









50-7 



50-6 

 43-5 



4550-6 

 4543-5 



2 

 1 



v. 







38-6 



389 



4538-7 



1 









31-8 



33-5 



4532-6 



1 









22-8 





4522-8 



1 



V. 







06-3 





4506-3 



1 





4506 





4498-3 





4498-3 



2 







4498-5 



91-9 



... 



44919 



1 



v tolerably broad. 







59-7 





4459-7 



1 



v broad, middle. 





4454-7 



47-7 





4447-7 



1 



Feeble maximum. 







22-4 





4422-4 



1 



Feeble maximum. 







13-5 





4413-5 



1-2 



v middle. 





4414-2 



As we see, the agreement between the values in the sepa- 

 rate series is in general very satisfactory; only in the yellowish- 

 red portion there are occasionally more important deviations, 

 but yet not greater than may be easily explained in view of 

 the smaller dispersion of this portion of the spectrum. We 

 may therefore regard the accuracy of the definitive wave- 

 lengths as in general quite satisfactory, and for most of them 

 we may estimate the probable error as not more than a unit 

 of Angstrom's scale. Those wave-lengths which occur in only 

 one series of observations are included in the column of mean 

 values; because, since each of them (as of all the other determi- 

 nations) is the mean of at least three separate observations, 

 they are in fact mean observations. 



Since the identity of the spectral phenomena of hydrogen 

 from different sources may be said to be put beyond doubt by 

 the numbers of the above table, it appears to me absolutely 

 impossible to ascribe the spectrum to any foreign substance 

 whatever. This is further confirmed by the fact that a fifth 

 tube which I filled with electrolytic hydrogen, carefully obser- 

 ving all precautions, showed identically the same spectrum. 



The single circumstance which can be alleged against what 

 has been here said by those who nevertheless regard this spec- 

 trum as that of acetylene, is the always possible contamination 

 of the gas by some carbon compound from the taps of the air- 



