(Photo., G. W. Ritchey, Yerkes Observatory.)
Nebulosities in the Pleiades as seen through the 24 inch Yerkes reflector, Oct. 19th, 1901. Exposure, 31/2hours.

On the evening of November 11th, 1572, Tycho Brahe was astounded to see a brilliant star-like point of light shining in the constellation Cassiopeia. He was certain that no star had been visible before, in that exact position. Although he had no conception of the terrific intensity of the explosion that we now know had taken place, he realized at once the supreme importance of what he saw. At first, indeed, he could not believe his own eyes, but as soon as he found that others also saw the star, he judged its appearance to be the greatest miracle that had occurred in the whole range of nature since the beginning of the world.

Similar outbursts must occasionally have aroused wonder ever since man took an interest in the starry skies, but, until comparatively recent times, the majority of these passed unrecorded. We have no account of any seen in Europe before the one which blazed out in Scorpio in 134 B.C., which is said to have induced Hipparchus to make a catalogue of the stars. The Chinese, however, tell of bright stars appearing in the sky in 2679, 2255 and 2238 B.C.

But, as far as we know, Tycho was the first to make a scientific study of such a phenomenon. He recorded all the variations in the brightness of the strange light, and proved that the object, whatever it might be, was far beyond the limits of the solar system and somewhere in the region of the stars.

Thus was introduced into astronomy a fascinating problem, to which three and a half centuries of astronomical research, with all the help that physics and chemistry can give, has failed to find a solution which has won universal acceptance.

There is, at last, fairly general agreement as to what happens during the outburst, but an extraordinarily wide divergence of opinion still exists with regard to the cause.

It was not easy to find even an appropriate name for these mysterious objects. They were at first called “New Stars” or “Novae Stellae.” But in their behaviour, during the brief period of their vivid and spectacular life, they differ completely from normal stars. The word “stellae,” therefore, has been discarded and we call them simply “Novae.”

It is doubtful to what extent even this epithet is applicable, but undoubtedly the message borne on the wings of light, though it may have been a thousand years on its way, is “news” when it reaches us. It must be read the very moment it arrives, or it will pass on and be lost for ever.

Our problem is to find out what a Nova really is, and probably the most convenient method of approach will be to consider how it resembles, and how it differs from, a normal star.

What is a Star?

In the twentieth century we have innumerable advantages that were not enjoyed by Tycho Brahe. The everyday achievements of a modern astronomer would have appeared incredible to him. It is always rash to affirm an impossibility. The philosopher who said “One thing is certain, we never can know the chemistry of the stars,” thought he was quite safe. His imagination failed to picture the magic powers of the spectroscope. When we ask to-day, “What is a star?” we get an astonishingly full and detailed answer. Astronomers in the great observatories, using giant telescopes armed with spectroscopes, interferometers and cameras, have been able to give us a surprisingly clear mental picture of what is to be found in the visible universe. One of the simplest, but most important, of the facts they tell us, is that every star is a sun, and that the great ruler of the solar system takes quite a humble place amongst the vast multitude of giant orbs that form the starry hosts.

The Sun.

We can, therefore, picture other stars most easily by comparing or contrasting them with the particular star that we know most about, our own Sun. The habitability of the earth is due entirely to the small fraction (about one 2230 millionth) of the solar energy that reaches its surface.

The Sun has a diameter more than 109 times that of the earth, and therefore, a volume 1,300,000 times as great. Although its average density is little more than a quarter of the earth's, its mass is equal to that of 333,434 worlds like ours. From the highest levels to the deepest regions observable in its atmosphere, the absolute temperature ranges from 5,000 to 7,000 degrees Centigrade. The latter is about double the temperature of the electric arc. The greater part of the interior is believed to be above a million, whilst the central regions have the inconceivable temperature of 30 million degrees Centigrade. Although the pressure at its centre must be a million tons to the square centimetre, the sun is believed to be gaseous throughout. The terrific encounters at such a pressure and temperature must denude the atoms of most of their outer electrons. The sun is 5,000 times brighter than liquid steel, and each square yard of its surface is continually pouring out radiant energy equivalent to 70,000 horse power. Its great gravitational attraction keeps all the planets in their orbits.

Other Stars.

The stars differ, one from another, in a most astonishing way in size, in density, in temperature and in luminosity, and to a much smaller degree in mass. A few examples will make this clear.

Size.—Antares, Alpha Herculis, and Mira Ceti are three celebrated giants,

(Photo., G. W. Ritchey, Yerkes Observatory.) Nebula about Nova Persei, Sept. 20th, 1901.

having diameters of three or four hundred million miles. There would be room inside Antares for Mercury, Venus, the Earth and Mars to move in their orbits, with some 66 million miles to spare outside the orbit of Mars. The volume of this giant is about 110 million times that of the sun. We cannot give examples of the opposite extreme since the smallest stars must be quite invisible through the most powerful telescopes. But we have details of a number of very small stars that happen to be specially near to us. One of these, called van Maanen's star, is said to have a diameter little over 6,000 miles, so it is considerably smaller than our earth. This gives a range, amongst known stars, of over 66 thousand times in diameter and of nearly 300 millions of millions in volume.

Density.—In spite of its small size van Maanen's star turns out to have a mass about 48,000 times that of the earth. It is nearly 300,000 times as dense as the sun, whilst the sun is four million times as dense as Antares. So the range in density is over a million millions.

Temperature.—The effective temperature of the photosphere varies greatly from star to star, and the resulting differences in the spectra have led to a useful classification of the stars. In O type, or Wolf-Rayet, stars, the elements which make their presence known are at temperatures between 30,000 and 50,000 degrees, whilst in some M type stars, such as Mira at minimum, they may be at 1,800 degrees only. These are the temperatures that rule near the surface. All stars are much more intensely heated within. In the strange companion of Sirius, Eddington calculates that the central temperature is a thousand million degrees.

Luminosity.—When the distance of a star is known, its intrinsic luminosity can be deducted from its apparent brightness. The results are often surprising. Sirius, apparently the brightest star in the sky, is in reality only 27 times as luminous as the Sun, whilst Rigel has 18,000 and Canopus 77,000 times the solar brightness. The apparent supremacy of Sirius is due to the fact that it is only 8.8, whilst Rigel is 543 and Canopus 652 light years from us.

It can readily be imagined how badly we should fare if any one of these stars was substituted for our Sun. It would be almost as disastrous in another way if our Sun were a dwarf.

Though so faint in comparison with Canopus or any one of innumerable other stars, our Sun shines 11,000 times as brilliantly as Proxima Centauri and 50,000 times as brightly as Wolf's star. This gives a 3,850 million fold range in luminosity.

Mass.—Our knowledge of stellar masses is less extensive than that of the other physical characteristics, since the mass can be determined only in the case of binary stars. The range is believed to be actually somewhat restricted, and it is often assumed that the majority of luminous stars have masses between one hundredth of, and one hundred times, the mass of our Sun. Except in the case of eclipsing binaries, it is only the minimum, not the actual mass, that can be found. In a list of spectroscopic Binaries given by R. G. Aitken, in his book “The Binary Stars,” the minimum values for one pair are 113.2 and 44.9, and for another pair 75.6 and 63.3 times the mass of our Sun. The lower limit is, of course, quite indeterminable, since, with the exception of a few that are specially close to us, all the smaller stars are invisible; but a number of pairs are known in which each star is much less massive than the Sun.

Novae.

With these few facts in mind we can realize to some extent the astounding magnitude of the changes which take place suddenly during the brief life and decline of a Nova.

Nova Aqullae.—Many will remember the night of June 9th, 1918, when Mr. G. V. Hudson, of Karori, noticed an unfamiliar point of light in the constellation Aquila. His telephone message to the Dominion Astronomer enabled enthusiasts to leave their beds and rush to the local observatories. The light that was gathered in by our telescopes carried messages that were far more marvellous than we realized at the time. All that we could do was to record the rapid changes in brightness that were taking place, to fix the position of the Nova with regard to surrounding stars by taking a few photographs, and to endeavour to identify a few of the bright lines with dark companions which were revealed when a small spectroscope was applied to the telescope. We have since learnt that the explosion, which we watched that night, had taken place 1,200 years before. During the whole of those twelve centuries the light had been speeding towards us, and spreading out equally in all other directions at a speed of about 186,300 miles per second. Photographic records show that before the explosion there was an eleventh magnitude star apparently in the place of the Nova. The brightness increased with such startling suddenness that in three days it attained a magnitude, -1.4, thus outshining every star in the sky except Sirius. Now that its approximate distance has been found, we can

(Photo., G. W. Ritchey, Yerkes Observatory.)
Nebula about Nova Persei, Nov. 13th, 1901. Compare with illustration above and note how the light spread out in the nebula in less than eight weeks.

page 30 page 31 look at this in another way. It means that, before the explosion, it was shining with four times the intensity of our Sun. In three days its intrinsic luminosity rose to 400,000 times that of the Sun. Then in the next 18 days it lost 98 per cent. of its maximum brightness, or 392,000 times that of the Sun. In eight months it became invisible to the naked eye. Spectrograms, taken during its vivid stage, show that ionized gases were rushing out from the scene of the explosion at a speed of more than a thousand miles a second. After six months a faint gaseous shell became visible. When photographed by Dr. Hubble on April 25th, 1927, this shell had grown to 18 seconds in diameter.

A similar sequence of events had been observed in connection with Nova Persei 1901, but with one exceptional feature. There seems to have been a nebula already in existence around the scene of the explosion. Successive portions of this were lighted up by the dazzling glare as it spread outwards in all directions. The apparent rate of growth of the illuminated shell, combined with the known velocity of light, gave the distance of the Nova. Expanding shells of gas were detected later, and these increased in diameter at rates which agreed with spectroscopic measures of the velocity of the outrush.

The Frequency of Such Explosions.

Until recently, bright Novae were believed to be very rare, but five have already appeared since the beginning of the twentieth century. The last three were Nova Cygni 1920, Nova Pictoris 1925, and Nova Herculis 1934.

It has been estimated that, if the whole heavens could be photographed each night, at least twenty Novae would be discovered in our Galaxy every year, whilst many more might be found in the nearer Spiral Nebulae. It is very significant that in the great Nebula in Andromeda, which appears to be a younger and more condensed galaxy, Novae are about twice as numerous as in ours, but in that in the Triangle, in which the stars are widely scattered in long spiral arms, Novae are comparatively rare.

What is a Nova?

When watching such an outburst as that of Nova Persei or Nova Aquilae, or even when reading about it, one is impelled to ask “What can have caused such a stupendous explosion?”

We have to account for hundreds of thousands of times the energy of our Sun being liberated in a few hours, for temperatures of hundreds of millions of degrees being suddenly produced, and for velocities, often exceeding a thousand miles per second, being found in the outrushing gases.

No theory which fails to explain such things, need be considered for a moment. But this is by no means all. The sudden fading of the star-like point of light, implying the dissipation of astounding quantities of energy, provides a still more searching test. If any theory survives this, it still has to face the evidence afforded by the succession of changing spectra, and the later development of the planetary nebular stage.

Professor A. W. Bickerton's diagrams of a stellar partial impact. The illustrations depict (from top) pair of stars distorted and coming into impact; pair of stars in impact; stars passing out of impact, and formation of third body; showing entanglement of matter in each body; two variables and a temporary star.

The Utter Inadequacy of Most Theories.

Now if you examine a hundred of the latest works on Astronomy, you will be amazed at the suggestions that still remain current. You will find that, of all the innumerable theories that have been proposed, two alone suggest any reasonable source for the amazing quantities of energy released. Of these two theories, one depends on the annihilation of matter, either during the formation of helium and other elements from hydrogen, or when protons and electrons rushing together are supposed to be changed from mass into radiant energy. But these rather hypothetical processes are said to take place at a more and more rapid rate as the temperature rises. It is difficult, therefore, to envisage any stopping place. If such a change were once started in a star it should go on at an increasing rate until the star is annihilated. This does not agree with the normal history of a Nova. Whenever one appears where a star has been photographed before, the final state is found to be slightly brighter than the original.

Bickerton's Explanation.

Fortunately, an explanation, which depends only on the established principles of chemistry and physics, was thought out in New Zealand sixty years ago. Professor A. W. Bickerton, of Canterbury College, was induced to consider the problem by the appearance of a Nova in Cygnus in November, 1876. He realized at once that the usual explanations, such as the combustion of hydrogen, or the eruption of a volcano, on a dead sun, were absurdly insufficient. No event less than the encounter of two stars seemed capable of liberating suddenly such vast stores of energy. On the 4th of July, 1878, Bickerton read, before the Philosophical Institute of Canterbury, a remarkable paper in which his theory was ably elaborated. This paper was followed by many others as the theory was found to throw light on the life histories of all kinds of celestial objects.

Partial Impact.

Stated very briefly the theory is that a Nova is caused by the partial, or grazing, impact of two stars, drawn together by their mutual gravitation. Each having some original velocity, they do not meet directly, but whirl in hyperbolic orbits around their common centre of gravity. If they come so close as to graze one another, the parts that meet are struck off and coalesce to form a fiery whirling unstable mass with extraordinary characteristics. This “Third Body,” or “Cosmic Spark,” as Bickerton called it, is found to furnish the key to the enigma. The wounded stars pass on, and make little show in the spectacular display. The “Third Body” is a twisted spindle shaped mass, with the lightest elements at the centre and the heaviest at its ends. It is intensely heated, but in a most unusual way. Since all have had the same onward motion transformed into atomic agitation or heat, the different elements are at widely different temperatures. Initially the helium is four times and the lead 207 times as hot as the hydrogen. But the average temperature being hundreds or thousands of million degrees, the body, with its compartively small mass, is thermodynamically unstable. Its atoms have more than the critical velocity of escape. Its disappearance is not page 32

(Photo., G. W. Ritchey, Yerkes Observatory.)
The Great Nebula in Andromeda as seen through the 24 inch Yerkes reflector, Sept. 18th, 1901. Exposure, 4 hours.

due to cooling. The mass is dissipated into space because it is too hot to hold together. It depends on the fraction struck off whether the wounded stars, each with a long lake of fire on its surface, escape from one another, or whether they are wedded into a binary system by the attraction of their brilliant but short-lived son.

The elements in the turbulent “Third Body” try to adjust the distribution of energy so that each atom has an equal share. To do this the heavy atoms must give heat to the lighter ones. Thus hydrogen soon leads the outward rush, with helium following at about half the speed. The brilliant nucleus becomes surrounded by expanding luminous shells of gas, whose light, when analysed by the spectroscope, tells what they are made of, and how fast they are flying. The atoms coming directly towards us absorb some of their appropriate radiation, so each bright line becomes fringed by a dark border on the violet side.

The peculiarities of each particular Nova depend on the characteristics of the colliding stars and on the depth of the impact.

The Tragic Neglect of a Fertile Working Hypothesis.

This wonderfully prolific theory has never been examined critically by anyone who speaks with authority. The reason usually given for this neglect is that stars are so far apart that they do not collide. But with streams of thousands of millions of stars interpenetrating one another, and each pulling every other more strongly as the distance decreases, it would be strange if no encounter ever took place. And it seems mathematically certain that, if one star does graze another, the clash must give birth to a Nova. One of the outstanding advantages of this theory is that is is not founded on vague surmises but on arithmetical calculations. One of the losses from its neglect is that it has allowed an immense amount of work to be spent elaborating theories that are arithmetically absurd.

The Star-Nebula Theory.

To give a single instance of this we may notice that even to-day the most popular theory of Novae is that each is caused by a star entering a nebula. The star is supposed to be stopped in a few days, or even in a few hours, by the resistance of the nebula. This theory leaves the sudden fading completely unexplained, but probably this is of no importance, for, with the accepted average density of a star and a nebula respectively, we are asked to believe that each atom of the latter is able to stop suddenly the headlong rush of 31/2 millions of millions of times its own mass.

What Happens When Star Meets Star.

When advocating this theory we may reasonably be asked to prove that the energy liberated when stars collide is actually sufficient to account for the observed behaviour of a Nova.

If a small mass is drawn from an immense distance to the surface of the Sun, it attains a speed of 386 miles per second. The same speed would be required to enable any body to escape from the Sun. This speed is, therefore, called the critical velocity of escape.

To leave the earth a projectile would have to start with a speed of at least seven miles per second. But this would not enable it to escape from the Solar System. It would either fall into the Sun or describe an orbit round it, according to the direction of the start.

Two stars like our Sun, drawn together by gravitation, would acquire before their surfaces met a relative speed of 386 miles per second. But the velocity destroyed in a slight graze would be little more than 193 miles per second. If, however, during the encounter the centres approach to within one radius of one another, the effective speed would be about 273 miles per second.

This implies an energy per unit mass 268 million times as intense as that of trains moving in opposite directions, each at 60 miles per hour. In a stellar collision this energy is transformed chiefly into heat, and is equivalent to 23 million calories per gram. The temperatures produced are most impressive.

Hydrogen molecules move about one mile per second when at a temperature of 200 degrees absolute, or 73 degrees below zero. If they move 273 miles per second, their temperature must be about

(Photo., G. W. Ritchey.)
Spiral Nebula Messier 51 Canum Venaticorum as seen through the 60 inch reflector at Mt. Wilson, April 7th–8th, 1910. Exposure, 103/4 hours.

page 33

(Photo., Mt. Wilson Observatory.)
The Great Nebula in Orion as seen through the 100 inch reflector at Mt. Wilson, Nov. 19th, 1920.

15 million degrees. All other elements will be hotter still, lead being at 3,105 million degrees. If the molecules are broken up into separate atoms the speeds will be increased.

Now, since within the “Third Body” the elements are at such widely different temperatures, at every molecular or atomic encounter the heavier element must give energy to the lighter. Hydrogen thus is enabled to get up speeds exceeding a thousand miles per second, such as are disclosed in many of the spectrograms of Novae.

The Peculiarities of Different Novae.

No two Novae are exactly alike, though all have certain essential family characteristics. The great differences observed in successive Novae are explained by the dissimilarities in the colliding stars and by the variations in the depth of the encounters.

We might expect that the collisions of gigantic stars would be more spectacular than those of stars like our sun, and that direct impacts would be grander than partial ones. Neither of these suppositions is correct. The most massive stars we know are all of enormous size and they have extremely low densities. Their encounters are slow, and the resulting temperatures comparatively moderate. Then, again, in a direct encounter the whole mass remains to restrain expansion. Professor Bickerton proved that, in a direct impact of equal stars, the energy is exactly sufficient to form, out of the two, a single star with double the diameter of either, and at the same temperature. Such an encounter, therefore, would merely double the luminosity whereas a grazing impact may multiply it hundreds or millions of times. This astounding increase in brightness is caused by the sudden expansion of the Cosmic Spark, due to its high temperature and small gravitational restraint. An impact may generally be considered a partial one if less than a third is struck from each star.

The magnitude and intensity of the explosion depend chiefly on the size and the density of the bodies involved. If the density remains unchanged the speeds developed in similar encounters are proportional to the diameters of the stars, the temperatures vary as the squares of the diameters whilst the duration of the encounter is unchanged. The clash of a pair of giants takes no longer than that of a pair of dwarfs.

If the diameters remain unchanged the temperature varies as the density, the velocity as its square root, and the duration of the encounter inversely as its square root.

(Photo., Mt. Wilson Observatory.)
Active solar prominence 140,000 miles high. The disc represents the earth.

To illustrate this let us compare or contrast an impact between a pair of giants like Antares, and another between two dwarfs like van Maanen's star, with one between two stars like our Sun. In the case of the giants, the velocity acquired is a quarter, and the temperature one-sixteenth, of those generated in a solar collision, and the encounter takes 109 days instead of 70 minutes. In a clash between the two dwarfs the temperatures would be twenty and the speeds nearly 4½ times those found in the case of Suns, and the encounter would be over in less than seven seconds. In such a collision if there was any lead in the “Third Body” it would be initially at 60,000 million degrees Centigrade.

We have considered only collisions between bodies exactly alike. A very improbable case. But the same principles apply in all encounters. Whilst variations in detail are infinite in number, all Novae have the same essential characteristics. We often hear of Novae and Super Novae. The latter are simply those in which temperatures and luminosities are thousands of times above the average, and the speed of the out-rushing gas about 4,000 instead of 1,000 miles per second.

When our telescopes are so much increased in power that we can detect the collisions of comparatively insignificant bodies, we may need a third class for those Novae that fall far below the average.

Meanwhile the theory we have advocated, and every other reasonable theory that is proposed, should be tested again and again by skilful observation and by rigorous mathematical calculation.