( -ATT ) 



The lines 15105,8 fP. 2) 11028 (Pl,5), 9520 (B.), 6908 (B.) and 

 8500 (B.) form a B.S. The first two lines give a differential vibration : 



40115,5 (P. 10) 

 (2,B.S.) — (1 ,B.S.) =: 2474,1 ') h = 40420 K = 4,04 fi (Moll) 



Summational vibrations of the lines of B.S. with those of the 1 S.S. 

 have not been observed. 



The line 4642,5 (S.) or 4642,35 (R.) may be considered as a 

 summational vibration of the first line of the principal series with 

 (1,S.S.). We get, viz. : 



(1,P.S.) 4- (1,1 S.S.) = 2 J 540 h = 4642,4. 



Rubidium. Saundkks gives as first doublet of the P.S. 7947,6 and 

 7800,2. The frequencies of the differential series are 11138,3 and 

 10978 (I,D.S.), 15259,3 and 15056,4 (2,D.S.j. The corresponding 

 wave-lengths have not beeji observed. These lines form again com- 

 binations with the first lines of the sid)ordinate series. H.M.Randall") 

 and F. Paschkn'-) have measured this si)ectrum ; the former below 

 28000 A.U. As first doublet of the first subordinate series Ramdall 

 finds 15290,3 and 14754,0; Bergmann measured before 15410 and 

 14850. RiTZ corrected these values by comparing (he measurements 

 of potassium by Bekgmann with those by Paschen, and found 15265 

 and 14727. According to my earlier calculations') 15185 is the 

 component with the greatest wave-lengih. Randall gives as first 

 doublet of the IIS.S. 13667,7 and 13237,0. For the following calcu- 

 lations we have made use of Randall's measurements. 



(1,D.S.) -(1,1 S.S.) -4360,5 .^^ 22933,2 ^^ 22936,7 (R. 1) - 



(1,D. S.) —(1,1 S.S.) — 4438 z^:^ 22532.4 ;,,.== 22533,0 (R 0,8) ) ^^^^ ^'^^ 



(2, IS.S.i) — (1,D.S.) 3^1912,2 /^^ 52296 >.^,,z_52313,4 (P. 6) 



(3, IS.S.) — (1,D.S.) —4973,3 /^^ 20107 ^' . 



(3, 1 S. S.) - (1, D. S.) :^- 4898,3 y^ ^ 20415 ( "°^ observed 



1) It should be specially noted here, thai A/, is followed by a +^, and the 

 possible, though not liie lirohahle value of ^ is the sum of the possible errors of 

 observation in the two fust lines of the series. This indication agrees with 

 4aP — ^p of Rrrz. Paschen doubts the validity of this difTerential vibration. 

 Ann, de Phys. Vol. 33-1910, p. 733. 



2) P. Paschen, Ann. d. Phys. 33, 1910 p. 717. 



H. M Randall, Ann. d. Phys. 33, 1910 p. 739. 

 3j Diss. Amsterdam 190G. 



