FROM CURRENT LITERATURE

Gravitational shift of spectral lines on the Sun. (Charles E. St. John. Evidence for the gravitational displacement of lines in the solar spectrum predicted by Einstein’s theory. The Astrophysical Journal 67, 195, 1928). The principal consequence of the general theory of relativity—the necessity of a displacement of spectral lines toward the red side of the spectrum in strong gravitational fields—has, as is known, received decisive experimental confirmation in the spectrum of the remarkable companion of Sirius. Since then the question of the “red shift” on the Sun, debated for more than 20 years, has naturally acquired secondary importance for physics. For astrophysics, however, independent proof of the presence of a red shift on the Sun remains a cardinal question of great practical interest.

As early as 1923–1926, the well-known investigator of the red shift on the Sun, St. John1, published a careful analysis of a large group of Fraunhofer lines, as a result of which he came to the conclusion that a gravitational shift of the required magnitude exists. The effect is complicated, however, according to St. John, by radial ascending and descending gas currents in the atmosphere of the Sun.

In a new extensive communication St. John analyzes 1,537 lines at the center of the Sun and 133 lines at its limb. The analysis includes all the most reliable lines, which can be carefully compared with terrestrial sources in vacuum.

Until quite recently, one might suspect on the Sun a serious competitor to the gravitational shift, namely pressure, since the radiating layers of the solar atmosphere were assigned a pressure of 5–7 atm. But direct spectral measurements and the generally accepted theory of ionization of Saha and others have led to the conclusion that the pressure in the solar reversing layer is practically equal to zero.

The second factor that may mask or, conversely, intensify the red shift is the radial ascending and descending currents of gas in the solar atmosphere. In the lower layers there must exist ascending convection currents, which are also detected by spectral

observations on the Sun and the stars. In the upper layers, on the contrary, there are descending currents, the cause of which is still unclear.

Thus there must exist intermediate layers in which the one and the other currents practically compensate each other. It is clear that precisely these layers constitute the most favorable region for a quantitative test of gravitational displacement. In any case, for a proper analysis of solar lines it is necessary to know the atmospheric levels at which they arise. St. John gives various methods for determining these levels: from the rotation of the Sun, from vortex flows around spots, from the difference of spectra at the center of the Sun and at the edges, and from data obtained during solar eclipses. These methods lead to consistent data.

The probable error in determining the position of an individual spectral line at the center of the Sun is equal to \(\pm 0.0009\ \mathring{A}\); for a group of lines this error is reduced to \(\pm 0.0003\ \mathring{A}\). The gravitational effect that must be measured is many times greater than these errors, as can be seen from Table 1.

TABLE 1.

The results of the analysis of 586 iron lines at the center of the Sun are given in Table 2. In the first column the spectral class is indicated, in the second—the number

TABLE 2.

lines taken to obtain the mean \(\Delta\lambda\) in the 4th column; in the 5th column is given the difference \(\delta\) between the observed value and that calculated according to the theory of relativity. The shifts are indicated in thousandths of an Å. In the last column is given the height of the level of the solar atmosphere for the corresponding lines. From the table it is clear that: 1) in all cases a positive shift of the order required by Einstein’s theory exists; 2) exact quantitative agreement is obtained for the middle layers at a height of about 520 km; in the lower-lying layers the shift is diminished owing to the Doppler effect accompanying ascending streams of gas, while in the upper layers the effect is strengthened thanks to the assistance of descending streams.

The influence of descending and ascending radial streams should in practice have little effect on observations at the limb of the Sun. Table 3 gives a summary of measurements referring precisely to this case for iron lines of classes \(a, b\). It should be noted that the effect caused by the rotation of the Sun is here taken into account, since the shifts given are the means of observations for two opposite limbs at one and the same heliographic latitude.

TABLE 3.

The observational data for the limbs of the Sun are less accurate than those for the center. From the table it is seen that the gravitational effect is confirmed; the influence of descending and ascending streams is absent here, but in the upper layers there exists a certain additional effect, reaching approximately 20%. The cause of this excess “effect at the limb” (if it is reliable) has not yet been clarified; at the center of the Sun it is absent. The remaining spectral material for lines of other elements gives the same results.

This work of St. John, one may hope, settles the question of the gravitational shift on the Sun in the affirmative sense and gives astrophysicists important material on radial gas streams in the atmosphere of the Sun.

S. Vavilov.

1. Proceedings of the National Academy of Sciences 11, 382, 1925. ↩