I'EBRUABY 11, 1916] 



SCIENCE 



187 



ever, and the discovery, although pub- 

 lished, was for the time forgotten. More 

 than two hundred years later (1777), 

 Scheele made images imprint themselves on 

 paper that had been saturated with a solu- 

 tion of silver chloride, but these images 

 disappeared when exposed to light. Fi- 

 nally, Niepce, in 1827, produced perma- 

 nent photographic pictures on metal, and 

 Daguerre improved the method in 1839. 



Becquerel and Draper in 1845 independ- 

 ently photographed the Fraunhofer lines, 

 this being the first application of photog- 

 raphy to scientific research. Since that 

 day photography has become one of our 

 important new senses and an indispensable 

 instrument of research. Even the long- 

 sought color photography is now a reality. 



Spectroscopy presents a good illustra- 

 tion of our subject. That rainbow colors 

 are produced by edges of glass plates was 

 known from the beginning of the Christian 

 era. Glass prisms were manufactured in 

 the seventeenth century, and attempts 

 made to explain the production of the col- 

 ors resulted in the solution given by New- 

 ton, in 1672. 



Wollaston, in 1802, observed seven dark 

 lines in the solar spectrum, but Fraun- 

 hofer by making larger and better prisms 

 and by using a telescope was enabled to see 

 ' ' countless ' ' numbers of them. He also dis- 

 covered the bright line spectrum of sodium, 

 superposed, however, over the continuous 

 spectrum of the heated carbon particles 

 present in the flame. He was also the first 

 to make and to use the diffraction grating 

 and to measure the wave-length of sodium 

 light. No explanation of the dark solar 

 lines was given, however, for forty years. 

 After the invention of the Bunsen burner 

 in 1857 many substances could be easily 

 vaporized, and spectral lines were obtained 

 free from the continuous spectrum which 

 confused previous experimenters. Thus 



spectrum analysis was developed, and the 

 true nature of the dark lines in the solar 

 spectrum soon afterwards demonstrated. 

 The possibility of detecting the motion of 

 stars by the shifting of the spectral lines 

 was considered. This, however, could not 

 be done with the instruments then avail- 

 able. Nothing more could be accomplished 

 until diffraction gratings were much im- 

 proved. 



The original gratings had been con- 

 structed of wire, and later they were made 

 with scratches on glass. These were soon 

 perfected sufficiently so that in 1868 Hug- 

 gins detected the shifting of the dark lines 

 in stellar spectra, beginning a new era in 

 the study of astronomy. The further per- 

 fection of these gratings has been most re- 

 markable and the results obtained with 

 them of the highest importance. Even now 

 we feel the need of a still higher resolving 

 power. The problem was, and still is, to 

 get the lines equally spaced with sufficient 

 accuracy for a large number of successive 

 lines. In about 1870, the Nobert gratings, 

 previously used, were replaced by those of 

 L. M. Rutherfurd (1816-1892), who finally 

 made gratings on speculum metal, with a 

 resolving power of 10,000. Rowland (1841- 

 1901) then succeeded in making gratings 

 with a resolving power of 150,000, which 

 advance revolutionized spectroscopy. The 

 gratings now made with the Rowland en- 

 gine have a resolving power of about 400,- 

 000, and the 10-inch Michelson grating, 

 600,000. These recent improvements in the 

 ruling of the grating, with the added aid of 

 photography, are extending far the limits 

 of a fertile field of research and amassing 

 valuable data for the ultimate demonstra- 

 tion of atomic structure. The time taken 

 through a long number of years to con- 

 struct an accurate screw was most prof- 

 itably spent. The same amount of time 

 employed in the taking of observations, in 



