Chemistry and Physics. 639 



The most noticeable departure from the conventional arrange- 

 ment is the postponement of the halogen elements to a relatively 

 late chapter, which the authors consider an improvement. The 

 book appears to present the theories, facts, and applications of 

 chemistry in an excellent manner. The ionic theory is clearly 

 explained, but it is not used obtrusively. The spelling is agree- 

 ably conservative, except that " sulfur " is used for sulphur. It 

 may be regarded as one of the best of the recent works on the 

 subject. h. l. w. 



6. Simplification of the Zeeman Effect. — The "normal" mag- 

 netic resolution of spectral lines, which was predicted by the 

 theory of H. A. Lorentz, has been found experimental^ only for 

 the lines of helium and for a relatively small number of single 

 lines of cadmium, calcium, magnesium, mercury and zinc. In the 

 majority of cases, however, the spectral lines have shown widely 

 different and very complicated " anomalous " types of separation 

 in the magnetic fields used. These anomalous cases, could not be 

 accounted for by a single hypothesis, so that the subject has 

 become more and more involved and confused as experimental 

 data have been gathered. Fortunately, a great step in advance 

 has been made by F. Paschen and E. Back. These investiga- 

 tors used a special form of vacuum tube in order to obtain a 

 source of light of sufficient intensit}^ in very strong magnetic 

 fields. The material of the tubes was fused quartz. Also a 

 specially constructed Klingelfuss induction coil was required to 

 produce secondary discharges of adequate strength and potential 

 difference. With this apparatus it was possible to obtain fully 

 exposed negatives of the more intense lines of hydrogen and 

 oxygen in one hour, using the first three orders of a Rowland 

 grating of the largest size. Helium required only a few minutes 

 exposure in the fifth order. The gas pressure was higher than is 

 ordinarily used in Geissler tubes, its value being 5 mm in the case 

 of oxygen. In the second part of the paper, which deals with 

 lithium, potassium and sodium, Back employed alloys of these 

 metals. 



A typical example of the results obtained is afforded by the 

 oxygen triplet whose components have the wave-lengths 3947*438, 

 3947*626 and 3947*731, corresponding to zero field strength. 

 When the magnetic field had a strength of 2800 gauss each com- 

 ponent of the triplet widened. At 6187 gauss the three lines 

 were much broadened but still distinct^ resolved. At 7738 the 

 edge of the longest wave-length component nearer to the middle 

 line of the triplet was strengthened. At 10930 a central line 

 appeared while the 3947*626 line vanished. At 12770, 21830 and 

 27880 gauss various changes took place in the "wisps" which 

 developed from the 3947*731 line. Also the shortest wave-length 

 component of the original triplet gradually disappeared with the 

 production of wisps. Finally, at 31900 gauss the outer com- 

 ponents of the triplet had vanished altogether. There remained 

 a sharp, intense line in the position of the center of gravity of 

 the triplet. The accompanying wisps w T ere very faint. All of 



