When we turn to the earlier chapters in the life of a star, the story is less clear. It is at least generally agreed that the blue-white stars exhibit an earlier and hotter stage. They show comparatively little absorption, and there is an immense preponderance of the lighter gases, hydrogen and helium. They (Sirius, Vega, etc.) are, in fact, known as "hydrogen stars," and their temperature is generally computed at between 20,000 and 30,000 degrees C. A few stars, such as Procyon and Canopus, seem to indicate a stage between them and the yellow or solar type. But we may avoid finer shades of opinion and disputed classes, and be content with these clear stages. We begin with stars in which only hydrogen and helium, the lightest Of elements, can be traced; and the hydrogen is in an unfamiliar form, implying terrific temperature. In the next stage we find the lines of oxygen, nitrogen, magnesium, and silicon. Metals such as iron and copper come later, at first in a primitive and unusual form. Lastly we get the compounds of titanium and carbon, and the densely shaded spectra which tell of the thickly gathering vapours. The intense cold of space is slowly prevailing in the great struggle.
What came before the star? It is now beyond reasonable doubt that the nebula—taking the word, for the moment, in the general sense of a loose, chaotic mass of material—was the first stage. Professor Keeler calculated that there are at least 120,000 nebulae within range of our telescopes, and the number is likely to be increased. A German astronomer recently counted 1528 on one photographic plate. Many of them, moreover, are so vast that they must contain the material for making a great number of worlds. Examine a good photograph of the nebula in Orion. Recollect that each one of the points of light that are dotted over the expanse is a star of a million miles or more in diameter (taking our sun as below the average), and that the great cloud that sprawls across space is at least 10,000 billion miles away; how much more no man knows. It is futile to attempt to calculate the extent of that vast stretch of luminous gas. We can safely say that it is at least a million times as large as the whole area of our solar system; but it may run to trillions or quadrillions of miles.
Nearly a hundred other nebulae are known, by the spectroscope, to be clouds of luminous gas. It does not follow that they are white-hot, and that the nebula is correctly called a "fire-mist." Electrical and other agencies may make gases luminous, and many astronomers think that the nebulae are intensely cold. However, the majority of the nebulae that have been examined are not gaseous, and have a very different structure from the loose and diffused clouds of gas. They show two (possibly more, but generally two) great spiral arms starting from the central part and winding out into space. As they are flat or disk-shaped, we see this structure plainly when they turn full face toward the earth, as does the magnificent nebula in Canes Venatici. In it, and many others, we clearly trace a condensed central mass, with two great arms, each apparently having smaller centres of condensation, sprawling outward like the broken spring of a watch. The same structure can be traced in the mighty nebula in Andromeda, which is visible to the naked eye, and it is said that more than half the nebulae in the heavens are spiral. Knowing that they are masses of solid or liquid fire, we are tempted to see in them gigantic Catherine-wheels, the fireworks of the gods. What is their relation to the stars?
In the first place, their mere existence has provided a solid basis for the nebular hypothesis, and their spiral form irresistibly suggests that they are whirling round on their central axis and concentrating. Further, we find in some of the gaseous nebulae (Orion) comparatively void spaces occupied by stars, which seem to have absorbed the nebulous matter in their formation. On the other hand, we find (in the Pleiades) wisps and streamers of nebulous matter clinging about great clusters of stars, suggesting that they are material left over when these clustered worlds crystallised out of some vast nebula; and enormous stretches of nebulous material covering regions (as in Perseus) where the stars are as thick as grains of silver. More important still, we find a type of cosmic body which seems intermediate between the star and the nebula. It is a more or less imperfectly condensed star, surrounded by nebular masses. But one of the most instructive links of all is that at times a nebula is formed from a star, and a recent case of this character may be briefly described.
In February, 1901, a new star appeared in the constellation Perseus. Knowing what a star is, the reader will have some dim conception of the portentous blaze that lit up that remote region of space (at least 600 billion miles away) when he learns that the light of this star increased 4000-fold in twenty-eight hours. It reached a brilliance 8000 times greater than that of the sun. Telescopes and spectroscopes were turned on it from all parts of the earth, and the spectroscope showed that masses of glowing hydrogen were rushing out from it at a rate of nearly a thousand miles a second. Its light gradually flickered and fell, however, and the star sank back into insignificance. But the photographic plate now revealed a new and most instructive feature. Before the end of the year there was a nebula, of enormous extent, spreading out on both sides from the centre of the eruption. It was suggested at the time that the bursting of a star may merely have lit up a previously dark nebula, but the spectroscope does not support this. A dim star had dissolved, wholly or partially, into a nebula, as a result of some mighty cataclysm. What the nature of the catastrophe was we will inquire presently.
These are a few of the actual connections that we find between stars and nebulae. Probably, however, the consideration that weighs most with the astronomer is that the condensation of such a loose, far-stretched expanse of matter affords an admirable explanation of the enormous heat of the stars. Until recently there was no other conceivable source that would supply the sun's tremendous outpour of energy for tens of millions of years except the compression of its substance. It is true that the discovery of radio-activity has disclosed a new source of energy within the atoms themselves, and there are scientific men, like Professor Arrhenius, who attach great importance to this source. But, although it may prolong the limited term of life which physicists formerly allotted to the sun and other stars, it is still felt that the condensation of a nebula offers the best explanation of the origin of a sun, and we have ample evidence for the connection. We must, therefore, see what the nebula is, and how it develops.
"Nebula" is merely the Latin word for cloud. Whatever the nature of these diffused stretches of matter may be, then, the name applies fitly to them, and any theory of the development of a star from them is still a "nebular hypothesis." But the three theories which divide astronomers to-day differ as to the nature of the nebula. The older theory, pointing to the gaseous nebulae as the first stage, holds that the nebula is a cloud of extremely attenuated gas. The meteoritic hypothesis (Sir N. Lockyer, Sir G. Darwin, etc.), observing that space seems to swarm with meteors and that the greater part of the nebulae are not gaseous, believes that the starting-point is a colossal swarm of meteors, surrounded by the gases evolved and lit up by their collisions. The planetesimal hypothesis, advanced in recent years by Professor Moulton and Professor Chamberlin, contends that the nebula is a vast cloud of liquid or solid (but not gaseous) particles. This theory is based mainly on the dynamical difficulties of the other two, which we will notice presently.
The truth often lies between conflicting theories, or they may apply to different cases. It is not improbable that this will be our experience in regard to the nature of the initial nebula. The gaseous nebulae, and the formation of such nebulae from disrupted stars, are facts that cannot be ignored. The nebulae with a continuous spectrum, and therefore—in part, at least—in a liquid or solid condition, may very well be regarded as a more advanced stage of condensation of the same; their spiral shape and conspicuous nuclei are consistent with this. Moreover, a condensing swarm of meteors would, owing to the heat evolved, tend to pass into a gaseous