520 THE POPULAR SCIENCE MONTHLY. 



owing more itself to cliemistry than to physics, in the latter 

 especially has it been of assistance to the spectroscope, so that the 

 experimenter is not dependent upon the observations of the mo- 

 ment to make his comparisons. The most considerable work of 

 this kind has been done by Prof. Rowland within the last half- 

 dozen years, in making remarkably large and detailed j)hotographs 

 of the solar spectrum, the spectrum itself, in its perfection and 

 beauty, being due to the matchless gratings constructed under 

 Rowland's directions. Photography has proved to be an unas- 

 sailable recorder for all the natural sciences, and is likely to be- 

 come more and more firmly established as such. Disputes over 

 priority in discovery will become less frequent since investiga- 

 tions made in solitude will appeal to their photographic record as 

 a safe witness, impartial and indisputable. 



Another subordinate problem is to determine the intensity 

 of sound in absolute measure. Acoustics has been studied with 

 reference to the energy involved less than other branches of phys- 

 ics, although we easily recognize some transformations of such 

 energy into mechanical in the phonograph and electrical in the 

 telephone. But most determinations of the intensity of sound 

 have been relative, by comparison of different sounds, or else the 

 same sound at different distances or in different media. They 

 have not been expressed in absolute units. Absolute values of 

 radiant energy, in the form of heat and light, have been deter- 

 mined, but the methods have not been sufficiently simplified to 

 make them readily applicable in experimenting. Temperatures 

 are still given in arbitrary degrees, and intensity of illumination 

 has no acceptable basis expressible in terms of the fundamental 

 quantities mass, time, and distance, although several methods 

 have been suggested in which the direct, subjective estimate of it 

 by the eye plays no part. 



This brings us to a consideration of the great service rendered 

 to scientific investigation by an absolute system of units and 

 measurements. Such systems were instituted by Gauss and 

 Weber between the years 1834 and 1850, and their introduction 

 was especially fruitful in the study of electricity. The mechanic 

 was enabled by that means for the first time to compare the elec- 

 tric forces produced with the mechanical ones employed, and 

 gained thereby for the first time a just estimate of the former. 

 The adoption throughout the scientific world of the centimetre- 

 gramme-second absolute system for all branches of science is by 

 no means the least valuable outcome of the development which 

 electrical science has undergone since 1850, for in the possibility 

 of tracing back all natural phenomena to the three mechanical 

 units of space, mass, and time, science received new evidence for 

 the inherent unity and the mechanical character of all forces of 



