A.— MATHEMATICAL AND PHYSICAL SCIENCES. 17 



and a spectroscopic laboratory came to be regarded by them as an 

 indispensable adjunct to an astrophysical observatory. The methods of 

 physical laboratories were freely utilised in these investigations and the 

 chemical interpretation of celestial spectra made rapid progress. The 

 introduction of photographic methods of observation by Huggins led 

 almost at once to the discovery of new lines apparently belonging to 

 hydrogen in the spectra of Sirius and other white stars, which were subse- 

 quently of great value in the establishment of Balmer's law of the hydrogen 

 spectrum. Perhaps the greatest contribution of early astrophysics to 

 our stock of knowledge, however, was that which so clearly pointed to the 

 essential identity of matter throughout the universe. 



With the discovery that the spectra of certain elements were modified 

 by varying the character of the exciting source, chemical analysis of the 

 sun and stars was supplemented and eventually overshadowed by 

 investigations of the physical conditions which prevail in those bodies. 

 The sun and stars thus came to be regarded as natural experiments on 

 generally similar masses of matter at various high temperatures — experi- 

 ments ready prepared for observation and always in operation. Thus 

 many laboratory researches were directly instigated by astrophysical 

 observations. To take one instance, the fragmentary observations by 

 Lockyer and by Liveing and Dewar of what were afterwards called enhanced 

 lines were extended and systematised through an attempt by Lockyer, 

 in which I myself took part, to interpret the spectrum of the solar chromo- 

 sphere as photographed during the total eclipses of the sun in 1893 and 

 1896. The immediate result was an important correlation of the changes 

 in the laboratory spectra of the elements with the succession of types 

 in stellar spectra, ^ from which it appeared that enhanced lines were 

 especially characteristic of stars which, on other grounds, were believed 

 to be hotter than the sun. These investigations laid the foundations for 

 a true interpretation of the spectra of the hotter stars, and led to the 

 more extended studies of enhanced lines which have proved of such great 

 importance in the development of the theory of the origin of spectra and 

 the structure of atoms. 



On the other hand, it may be remarked, astrophysics owes much to 

 laboratory experiments which were undertaken without regard to the 

 sun and stars. One of the most notable examples is Zeeman's famous 

 discovery of the splitting of spectrum lines when the source is placed 

 in a magnetic field — a discovery which was afterwards appHed with such 

 brilliant success by Hale to the detection of the magnetic fields in sun- 

 spots and of the general magnetic field of the sun. 



In one way or another the spectrum has thus become much more than 

 a key to chemical composition ; it has become also a key to the physical 

 conditions under which the corresponding radiation is excited ; and, as 

 some of the earlier workers clearly anticipated, a key to the problem of 

 atomic and molecular structure. 



The remarkable developments of modern spectroscopy in the direction 



of atomic physics have resulted from discoveries relating to regularities 



in spectra. Such regularities were suspected as long ago as 1869, and 



were actually revealed about ten years later by the admirable experimental 



1 Lockyer, Boy. Soc. Proc, vol. GO, p. 475 (1897). 



1 92(5 C 



