754
ASTRONOMY
More recently Ritter has developed the subject with great fulness in a series of papers published in the Leipzig Annalen der Physik und Chemie. As we have not room for even an abstract of these researches, we can state only what seem to be the most obvious conclusions. That the sun and stars are masses of gas is shown not only by their density, so far as ascertained, but by the fact that solid or liquid masses would cool with such rapidity that they would cease to shine at the end of a few centuries. The superficial portions would first cool and eventually become solid. It was shown by Lane that a mass of gas, held in equilibrium by the mutual attraction of its parts, grows hotter through the loss of energy by radiation, because the temperature gained by the resulting contraction exceeds that lost by the radiation. The effect is analogous to that of a resisting medium upon a planet. The direct result would be to retard the motion of the planet, so that it would fall nearer to the sun, but the velocity gained by the fall would exceed that lost by the resistance. As a planet would go faster by being resisted, so the temperature of a gaseous mass like a star rises as it contracts in volume. This rule ceases to hold when the attraction has become so great that liquefaction takes place, and then the radiation of light by the body is nearly at an end. Another result of the theory is that the greater the mass of a star the higher will be its temperature in a given stage of its development. There may, therefore, be a wide range among the temperatures of the stars, the older and the larger ones generally being the hotter. The volume of a star is so great that its interior, though gaseous, is not transparent, and it therefore radiates heat only from its superficial portions. These are thus subject to a rapid cooling process; they probably liquefy or solidify, and then by their superior gravity sink into the interior and are replaced by gaseous masses from the interior. Thus a circulation is constantly going on, the hot matter of the interior rising to the surface, and the cool matter of the surface sinking to take its place. A star cannot radiate heat more rapidly than the heat is supplied by this process; consequently, the superficial temperature is limited by the rapidity of circulation, which cannot go on at more than a definite rate, depending on the temperature and the nature of the gas. A necessary result of this view is that each star is constantly contracting in volume, and must continue to do so until the interior becomes liquid or solid. At this stage the increased friction resisting the circulation will result in a continuous fall of the temperature of the surface, which in process of time must cease to radiate light. Recent discoveries give colour to the hypothesis of evolution implied in this view of the physical constitution of stars in general. Their agglomeration into systems is what we should naturally expect as the result of the condensation of irregular nebulous masses. One of the most suggestive conclusions of recent science is that the stars differ greatly in density, and that, in the Density large majority of cases, the brighter ones at least of the are much rarer than the sun, probably even of stnrs. gaseous density. In the case of the Algol type of variable stars we can, by the comparison of the duration of the eclipse with the periodic time, form an approximate idea of the size of the star. If the elements of the orbit can be determined, the mass will also become known, and thus we may form an idea of the density. So far as this method has been carried out, the conclusion is that already stated. Revolving double stars, when the elements of the orbit become known, afford evidence that points in the same direction. Knowing the parallax of a star and the elements of the orbit of a companion moving around it, we can determine the mass of the two
bodies. Their absolute brilliancy compared with that of the sun is also known from the parallax and the apparent brilliancy; and it is thus found that, while the mass of Sirius is a little more than double that of the sun, it emits some thirty times as much light. It is now. known that a relation between the mass and brilliancy of a double star can be derived from the elements of the apparent orbit without a knowledge of the star’s distance. Were all these bodies of the same density and the same intrinsic brilliancy per square mile of surface as the sun, then to every star of a given apparent magnitude, whatever its distance, would correspond a certain relation between the periodic time and the apparent mean distance of a companion revolving around it. Assuming the distance of the companion to be 1", its period on this hypothesis is shown for stars of various magnitudes in the following table:— Magnitude. Feriod. Magnitude. Period. yy4 14-1 o 0-9 28-2 1 1-8 5 56-2 2 3'6 6 112-0 3 7-1 7 If the apparent mean distance is different from 1" the time of revolution is given by Kepler’s third law in the form T2 = Tj2 S3, S being the mean distance in seconds, and Tx the period in the above table. In a large majority of double stars whose orbits have been approximately determined it is certain that the time of revolution is much greater than would result from this rule. It follows that the stars in general are either much less dense, or much more brilliant, than the sun. For reasons that we cannot state here it is likely that the former alternative is more usual than the latter. It is remarkable that, in the case of those bright stars the mass of whose faint companions has been determined, the brilliancy of the latter deviates in the opposite direction; their masses are disproportional to their brightness. Thus the mass of the companion of Sirius is nearly half that of Sirius itself, although the latter gives several thousand times as much light as the companion. The same is probably true of the companion of Procyon, and, in a less degree, of rj Cassiopeia. It would seem that in these cases the companion, being of the same age as the bright star, has cooled off more rapidly, and perhaps condensed to a solid or liquid. The variety of spectra among the stars also emphasizes the diversity of their physical constitution. Sir W. Huggins {Publications of Sir W. Huggins's Observatory, vol. i. page 75) shows that the evidence of the spectroscope agrees with the general evolutionary theory. The spectra of the stars may be made to fall into line, so as to form a series, leaving little room for doubt that the actual differences between them represent in the main successive epochs of star life rather than original differences of chemical constitution. When the spectra of close pairs of stars are examined, the difference between the two sets corresponds to the view that the fainter star of the two is in a more advanced stage of condensation. The problem of the structure of the heavens, if structure they may be said to have, may be regarded as the ultimate one of sidereal astronomy. Branches of this problem are the questions of the physical constitution of the stars, their possible separation stars. into systems, the stability of those systems, and the duration and stability of the universe itself. The central question around which all the others may be grouped is that of the actual distribution of the stars in space. If we gould determine the distance of a star as readily as we do its direction from us, this question would be immediately settled, and a geometric model of the universe could be constructed showing its form and
