256 Obituary Notices of Fellows deceased. 



optical grating, which strikingly illustrates the power of mechanical 

 genius directed by theoretical ability. The construction of optical 

 gratings had already in the hands of Rutherfurd reached a high 

 degree of perfection in America ; but these instruments could not be 

 employed for photography, or for measurement without the use of 

 lenses, while owing to inaccuracies of ruling and periodic errors, the 

 spectra exhibited " ghosts," and it was impossible to realize the 

 theoretical limit of resolving power. The first difficulty was met by 

 the invention of the concave grating, which was communicated to the 

 Physical Society of London in 1882, and published in the Philosophical 

 Magazine for September, 1883. The second difficulty was met by the 

 invention of a simple method of making a perfect screw by grinding, 

 as described by him in the article " Screw " in the ninth edition of the 

 Encyclopaedia Britannica. Without a perfect screw a perfect grating 

 could not be ruled ; but the construction of a screw was after all only 

 a part and that a small one of the difficulty. The perfect screw 

 requires to be perfectly mounted, and worked automatically in such a 

 way as to rule perfectly straight lines at equal intervals ; and the 

 perfected apparatus requires to be most carefully studied, so that its 

 residual errors may be automatically corrected during the ruling of 

 the grating. In all these constructive details the mechanical genius of 

 Rowland is exhibited at its best. 



The maps of spectra which he has made it possible to produce, by 

 the admirable gratings, otherwise unattainable, which he has distri- 

 buted at a nominal price all over the world, have done more than any 

 other invention to add to our knowledge of the fundamental science of 

 spectroscopy, and form the most lasting monument of his genius. 



The greater part of the next decade was occupied by Rowland and 

 his students and assistants in working out spectroscopic and kindred 

 measurements to the limit of accuracy made possible by the construc- 

 tion of these concave gratings. This involved incidentally an elaborate 

 study of photographic processes for the perfect reproduction and 

 accurate recording of results. It has led to new discoveries, such as 

 the effect of pressure or density of the surrounding atmosphere on the 

 wave-length of a bright line, the relations connecting the frequencies 

 of the various series of lines in the spectrum of an element, and 

 analysis of a line by a magnetic field, which could not have been made 

 without spectroscopes of great resolving power, or without standard 

 tables of wave-lengths. 



In his later years Rowland gave more of his time to engineering 

 problems ; as was natural, his advice was often sought as regards new 

 departures in the inception of large electrical undertakings. He also 

 invented, and gave much time and attention to developing, a system of 

 multiple telegraphy, which was shown working at the Paris Exhibition 

 in 1900, the year before his death. 



