136 ANNUAL REPOET SMITHSONIAN INSTITUTION, 1932 



tests with materials of different composition give you an idea of the 

 safe limits. Thus, with the spectroscope you can test these alloys in 

 a minute part of the time that chemical analysis requires. 



But it is in the realm of atomic physics that spectroscopy has 

 played its greatest role. Fifty years ago, Lockyer, from a study of 

 the spectra of electric arcs and sparks, and of the stars, concluded 

 that, in the spark and in the hotter stars, ordinary atoms are de- 

 composed into products which give different spectral lines. This 

 bold generalization was fully justified 40 years afterward, by the de- 

 velopment of the theory of ionization. 



About 40 years ago, series of lines were detected in many of the 

 simpler spectra, and found to be representable by formulae in which 

 the " Rydberg constant," common to all spectra, appeared. Here was 

 evidence of some uniform feature in the constitution of the different 

 atoms. The Zeeman effect, according to which a spectral line emitted 

 by a source placed in a strong magnetic field is split up into polarized 

 components, again showed features common to different atoms, and 

 suggesting the presence within them of moving electrical charges. 

 The Bohr-Rutherford theory of atomic structure — with electrons in 

 orbital motion around a nucleus — was based very largely on these 

 spectroscopic data. It accounted at once for the typical spectral 

 series of hydrogen, and accurately predicted other series in the infra- 

 red and ultra-violet. With simple modifications, it explained the 

 more complicated system of series in the spectra of the alkalies. The 

 multiple character of the terms of the series was later interpreted 

 as a result of the spin of the electron — thus increasing the " astronom- 

 ical " resemblance of the atom-model ; while the appearance of nu- 

 merous terms in the more complex spectra was accounted for by 

 differently quantized inclinations of the electron orbits. The com- 

 plex multiplets of lines found in the spectra were thus fully ex- 

 plained. In its final form (due to Hund) this theory has been 

 brilliantly successful in elucidating the structure of atoms and inter- 

 preting and even predicting the details of their spectra. Work in 

 this field has been very active, and only the most complex spectra 

 (rare earths and some heavy metals) and those of a few very rare 

 elements remain to be deciphered. 



In the case of molecules, changes in the states of oscillation and 

 rotation of the nuclei, as well as in the electronic states, are possible, 

 and the spectra are much more intricate, consisting of complex bands 

 comprised of closely packed lines. Those of diatomic molecules are 

 now well understood — with important gains in our knowledge of 

 molecular structure and the nature of chemical " affinity " — and the 

 still more intricate polyatomic molecules show signs of yelding. Dif- 

 ferent isotopes of the same element, when present in compounds, often 



