January 25, 1918] 



SCIENCE 



97 



. I 



i;Ci=i - 



■| ! 



I I ~f S '~i I 



while the absorption region begins at X2100, 

 has a maximum between X2000 and A1900 and 

 extends to A1800. 



The uranyl salts can also be stimulated by- 

 light of a wide range of wave-lengths and in 

 the main Stokes's law is followed, although 

 the fluorescence and absorption regions over- 

 lap. Like Professor McLennan's spectrum, 

 the spectra of the uranyl salts appear to be un- 

 aflfected by the mode of excitation, and while 

 both spectra have been carefully tested for 

 resonance, both have failed to show the phe- 

 nomenon. For the above reasons it was 

 thought that possibly the spectrum of the 

 iodine vapor could be resolved into series of 

 the same simple type as those found in the 

 uranyl salts. If the wave-lengths of Professor 

 McLennan's bands are converted into fre- 

 quency numbers and plotted it is easy to dis- 

 cover series having constant intervals. Spec- 

 trum No. 1 in Fig. 1 shows the bands without 

 any attempt at classification. It will be noted 

 that the spectrum is in two sections because of 

 its great length. 



Professor McLennan notices several groups 

 of bands which are spaced approximately 20 

 units apart. Such groups are present at 

 1/A=2400 and at 1/A = 3100, but the series 

 designated in spectrum No. 2 of Fig. 1 pos- 

 sesses much longer intervals. In this plot the 

 members of the same series are given the same 

 letter, and in addition a few have also been 

 designated by long brackets. Such series as 

 A, V, N and M are given brackets and the 

 average value of the interval placed over the 

 bracket. Here, as has so often been observed 

 in the study of the uranyl spectra, a given 

 series has a constant interval, but the various 

 series have slightly different intervals. The 

 value assigned to series A is 160, but this is 

 an average value for the series, the actual in- 

 tervals being given under Series A in Table 

 I. as varying between 161 to 159 units. If the 

 reader inspects the other series he will observe 

 that the differences are generally unequal, but 

 show no systematic deviation from_ a mean 

 value. There are a few gaps in the series 

 which may be caused by the presence of an ex- 

 ceedingly strong mercury line in the region 



