ii6 



NATURE 



[April 5, 191 7 



and by Lyman, respectively. His attempt to apply 

 the same method to helium led to results which are 

 still in dispute, and will be referred to later. 



In reviewing recent progress we may begin with that 

 field in which the United States has taken a leading 

 part — that of astrophysics. This domain belongs as 

 much to the physicist as to the astronomer. The 

 heavenly bodies are laboratories on a vast scale, in 

 which nature has provided conditions of temperature, 

 pressure and electrical state which we may never hope 

 to rival on the earth. The spectroscope gives us data 

 from which Jt may be possible to form some idea of 

 these conditions by comparison with our feeble labora- 

 tory imitations of celestial phenomena, and conversely, 

 the latter may aid in the interpretation of terrestrial 

 phenomena. 



One of the most fruitful astronomical applications 

 of the spectroscope is to the determination of veloci- 

 ties in the line of sight, by the Doppler-Fizeau prin- 

 ciple. A large mass of such data has been collected, 

 from which some important generalisations have been 

 derived. For example, Campbell has determined the 

 velocity and direction of motion of the solar system 

 through space, and has found a remarkable and as 

 yet unexplained relation between the velocities of stars 

 and their apparent age, the redder and presumably 

 older stars and a class of nebulae having in general 

 the greater velocities. It likewise appears that two 

 immense star streams are crossing each other in the 

 Milky Way. Many spectroscopic binaries have been 

 discovered and their orbits determined, and recently 

 there have been found remarkable displacements and 

 rotations in nebulae which may throw some light on 

 the nature and destiny of these bodies. The spectro- 

 scope has enabled astronomers to undertake the am- 

 bitious task of tracing the course of stellar evolution. 



The most ingenious and fruitful device for studying 

 the sun is the spectroheliograph, invented by Hale in 

 1892. With this instrument photographs of the dis- 

 tribution of a given constituent of the solar atmosphere 

 may be obtained by restricting the light falling on 

 the photographic plate to the wave-length of one of 

 the characteristic lines of the element. The configura- 

 tion of the hydrogen clouds in the neighbourhood of 

 sunspots led Hale to suspect vortical motions in such 

 regions. In 1908 the study of a number of plates, 

 which showed that hydrogen flocculi were actually 

 drawn into these spots from great distances, proved 

 without question that sunspots are cyclonic areas of 

 enormous extent. Thus the long-disputed question 

 as to the nature of sunspots was answered, but this 

 was not all. Vapours which emit or absorb line 

 spectra are ionised, and as the more mobile electrons 

 would diffuse more rapidly to higher levels than the 

 positive ions. Hale inferred that the immense whirls 

 of electrified vapours in the neighbourhood of the 

 spots must cause a radial magnetic field. If such 

 fields are sufficiently intense, the longitudinal Zeeman 

 effect should be produced. As a matter of fact, the 

 spectrum of light from the spots is characteristically 

 different from that of the surrounding photosphere, 

 one of these peculiarities being the doubling of many 

 lines. As Hale anticipated, an examination of the 

 state of polarisation of such lines showed them to be 

 circularly polarised, and the direction indicated that 

 the whirling vapour was negatively electrified. Hale 

 likewise sought for the more minute effects which 

 might be expected from the rotation of the solar 

 atmosphere as a whole. A study of the breadth of 

 spectral lines^ at different latitudes and the detection 

 of traces of circular polarisation at their edges showed 

 that the sun possesses a magnetic field with polarity 

 corresponding to that of the earth, but of much greater 

 intensity. Although the atmospheric conditions on 

 the earth are very different from those on the sun, 

 NO. 2475, VOL. 99] 



it is possible that these investigations may assist us 

 in solving the baffling problem of the earth's mag- 

 netism. 



One of the most impressive facts revealed by the 

 spectroscope is the substantial identity of constitution 

 of the heavenly bodies. Everywhere we find evidence 

 of the existence of such elements as hydrogen, sodium, 

 calcium, and iron. But we also find an infinitude of 

 differences in the appearance of the lines, which we 

 must attribute to differences of temjxsrature, vapour 

 density, pressure and electrical condition. It is sug- 

 gestive to find that the spectrum of some stars re- 

 sembles that of the arc, of others that of the spark. 

 We may hofve by comparing the spectra of these bodies 

 with those produced in our laboratories under varied 

 conditions to reach some conclusions regarding their 

 physical state. The Mount Wilson physical labora- 

 tory is doing much valuable work of this kind. 



In the spectra of the solar corona and of nebulae 

 and nebulous stars certain lines are found which do 

 not belong to known elements. This need not indi- 

 cate any fundamental differences between the life- 

 history of such bodies and that of the older stars. 

 Twenty-five years ago Lockyer's views regarding the 

 dissociation of elements in the stars were treated with 

 levity by most physicists and astronomers. To-day 

 such notions are held to be quite rational. The more 

 elementary forms of matter would naturally be of 

 small atomic weight, and hence would diffuse to 

 higher levels than the heavier elements, and might 

 ultimately escape into space. If it were not for the 

 fact that it is held captive in chemical combinations, 

 we should know nothing of hydrogen. Helium first 

 revealed itself to us through its solar lines, and would 

 still be otherwise unknown to us were it not for its 

 continuous production in radioactive processes. The 

 elements giving the spectra of the corona and of the 

 nebulae are presumably of small atomic weight, and 

 are possibly the units out of which more complex 

 known elements are built, in later stages of develop- 

 ment; or they may be, conversely, the results of the 

 disintegration of such elements. It is not impossible 

 that in the future we may detect traces of these 

 elements on the earth or manufacture them by some 

 powerful disintegrative process. Meanwhile, deduc- 

 tions from known relations between frequencies of 

 the spectral lines, their breadth, and the atomic 

 weight of the elements may give us some clue to 

 their atomic weights. Nicholson has succeeded in 

 constructing hypothetical atoms with given nuclear 

 charges and electron ring systems which give with 

 remarkable accuracy the positions of the lines of the 

 corona and nebulae. Rayleigh showed from kinetic 

 theory and Michelson proved experimentally that at low 

 pressures the width of lines may be entirely due to 

 Doppler displacements, which vary directly as the 

 square root of the absolute temperature and inversely 

 as the square root of the atomic weight. Buisson 

 and Fabry have verified this law and applied it to the 

 study of nebulae. The width of certain lines, deter- 

 mined from the limit of interference, indicates that 

 the temperature of the Orion nebula is about 15,000 

 degrees, and that two groups of lines are due to atoms 

 of weights 272 and between i and 2 respectively. 

 This is a remarkable confirmation of Nicholson's 

 previous conclusion that the emission centres are of 

 atomic weights 2-95 and 1-31. 



During the past ten years the boundaries of the 

 known spectrum have been greatly extended in both 

 directions. The difficulties of investigation in the 

 infra-red are very great, but by the methods of 

 reststrahlen and of focal isolation Rubens, working 

 in succession with Nichols, Wood and von Baeyer, 

 has isolated and measured certain regions of great 

 wave-length. The longest wave-length measured is 



