Stellar Astrophysics After 1900 ( Major Trends In The History of Astronomy)
The Draper catalogues provided astrophysicists with the massive numbers of spectra needed for statistical studies of the distribution of stellar types, which in turn might provide clues on stellar evolution. Two independent studies of particular interest, one by E.Hertzsprung, a Danish astronomer, the other by the American astronomer H.N.Russell, resulted in one of the corner-stones of modern astrophysics – a diagram now named after both of them (the Hertzsprung-Russell diagram] which describes the distribution of stars as a function of their luminosities and temperatures.
In 1905 Hertzsprung used a statistical method for obtaining distances to examine some of the stars on A.Maury’s list. He found that stars at the beginning of the classification scheme, now called O, were extremely luminous, and also that the red stars of classes K and M contained two types – one of faint absolute luminosity, the other, very high luminosity. In addition these distinctions in luminosity were correlated with the sharpness of the spectral lines noted by Maury. Hertzsprung proposed the names dwarf and giants for the two types of stars. At about this time Russell, engaged with obtaining distances by photographic methods, found that there was a definite correlation between luminosity and spectral type. These results suggested to him an evolutionary sequence similar to Lockyer’s with the luminous red stars at the young end of the scale, the faint red stars at the other extreme. On hearing of Hertzsprung’s discovery of two luminosity classes, Russell believed his theory to be confirmed. Hertzsprung, on the other hand, considered that the results indicated two parallel series of evolution.
In 1911 Hertzsprung found that a plot of the apparent magnitude against colours for stars in the Pleiades and Hyades clusters yielded a narrow band distribution with stars becoming redder with decreasing brightness. Russell produced a similar plot (rotated by 90°) for all stars, which he discussed in 1913 at the Royal Astronomical Society and later that year at the American Astronomical Society. The distribution of the majority of the stars in a narrow band across the diagonal of the plot (now called the main sequence), demonstrating the distribution of spectral classes, was agreed by most astronomers to be the result of varying surface temperatures. However, there was no general agreement over the causes of the luminosity difference between the horizontal line (giant branch) and the diagonal band, that is between the dwarfs and giants, as named by Hertzsprung. Russell’s view was that density was the main cause of the difference: the order of evolution was one of increasing density from the giant branch through the O stars along the main sequence to the faint red stars with the temperature first rising, and then, along the main sequence, falling. This picture was finally overthrown in the 1920s with the large-scale use of photoelectric photometry (introduced by Stebbins about 1910) to examine the colour-magnitude diagrams of stars in open and globular clusters. These studies revealed that the straightforward linear picture of evolution had to be significantly modified.
An important by-product of the work on luminosity differences was the method of spectroscopic parallax. Hertzsprung, by 1911, had suggested that if the luminosity of a star could be obtained from its spectrum its distance could be deduced. However, he did not pursue the idea. In 1914, the American astronomer Adams and the German astronomer A.Kohlschutter announced results which led to the introduction of the method in 1916. They found that giants and dwarfs of the same spectral class produced lines of dif¬ferent intensity. Using stars with known distances, they were able to calibrate intensity ratios of certain lines so that the observation of these same lines in another star provided its absolute luminosity, which, on knowing the star’s apparent magnitude, provided its distance. This method, much more powerful than the method of trigonometrical parallax, led to a rapid increase in the number of known stellar distances.
Somewhat earlier, the determination of the actual sizes of stars had been calculated by combining the laws of thermal radiation with estimates of stellar temperatures. These results showed that the names giant and dwarf could be interpreted literally to refer to the relative sizes of the two classes. An interferometer mounted on the 2.5-m telescope at Mount Wilson was used by A.A.Michelson and F.G.Pease in 1920 actually to measure directly the sizes of certain supergiant stars; their measurements confirmed earlier calculations. Another method for obtaining the dimensions of stars was developed by H.N.Russell and H.Shapley at Princeton from a detailed study of the light variations in eclipsing binaries; their study meant that the sizes of stars of all classes could be obtained.
In addition to the spectroscopic and eclipsing binaries, variable stars, particularly Cepheids, provided new insight into the nature of stars. Shapley had shown in 1914 that the Cepheid phenomenon could not be explained by a binary system, as had been supposed up to this time. He suggested that the light variations were due to actual pulsations. A pulsation theory was soon developed by the British astrophysicist A.S.Eddington, from which it was possible to derive the sizes of Cepheids. His theory also provided an expla¬nation for the correlation between the period and magnitude of Cepheids just discovered by Henrietta Leavitt. Later, Eddington developed a theory of stellar interiors which still forms the basis for such studies. In a series of investigations between 1918 and 1924 he presented the theory of radiative equilibrium of stars, a theory which requires the centre of stars to be at extremely high temperatures. At the time, no energy mechanism wan known which could retain such high temperatures for long periods of time. A solution was only reached in 1938 when H.Bethe in the United State and C.von Weizsacker in Germany discovered a nuclear reaction in which hydrogen is transformed into helium. The discovery of this reaction meant that for the first time estimates of the energy -source of the Sun were sufficient to satisfy modern estimates of the age of the Earth.
This concludes our brief survey of the development of stellar astrophysics. In many ways this subject must be seen as one of the great triumphs for theoretical researchers. Even though only one star can be examined in great detail, and despite the fact that we cannot see inside any stars, physicists have learned much about the basic physical processes through the application of the laws of thermodynamics and nuclear physics as deduced on Earth. We are next going to look at the sequence of events that led to an under¬standing of the nature of the non-stellar objects – the nebulae. Here we see less direct application of physics. Our interest mainly lies in the fact that the extragalactic nebulae – galaxies as we now term them – hold the key to determining the large-scale structure of our Universe. We begin our story with William Herschel, one of the finest observers, who made the first detailed listing of the nebulae. He was not able to discriminate the gaseous nebulae of the Milky Way from the external galaxies; that step was not in fact made until the present century, and it is at that point that our description ends.