NOVA. A nova may be defined as a variable star which brightens up for a short time to a luminosity far exceeding that of its normal condition, and does so only once over a long interval of time. The facts observed are (I) a very rapid brightening (frequently many thousandfold in two or three days) followed by a slower fluctuating fall in brightness; (2) a steady sequence of changes in the star's spectrum, which in the beginning consists of lines, due to ionized atoms of such elements as calcium, iron and titanium, and finishes with those due to ionized atoms of oxygen, nitrogen and helium; (3) the rapid expansion of a disc round a central bright nucleus, with evidence of complex internal motions.
The first nova which counts in modern astronomy was that of Tycho Brahe, N Cassiopeiae, 1572, the first nova definitely allot ted not to the solar system but to the stellar universe by measure of its distance. This nova was brighter than Jupiter or Venus, was easily visible at midday, and remained a naked-eye object for 16 months. N Coronae Borealis, 1866, was the first nova to be ex amined through the spectroscope, and its spectrum was found by Sir William Huggins to consist of bright hydrogen lines on a con tinuous spectrum which also showed many absorption lines. Ten years later Cornu noticed that the bright line spectrum of N Cygni, 1876, was very like that of the chromosphere, the layer of the sun's atmosphere which flashes into view just as the disc of the sun passes behind the moon at a total eclipse. This was amply confirmed in the case of later novae, the next point to be noticed being that this spectrum is in general, at discovery, a dark line spectrum; the dark lines are always displaced markedly to the violet, suggesting that the gases which give rise to them are mov ing rapidly towards the observer.
The majority of novae appear either in the Milky Way, i.e., with galactic latitudes less than 2o°, or else in spiral nebulae. In the Milky Way they are concentrated in the direction of the star clouds of Sagittarius, in galactic longitude, 32o° to and judging from analogy with the 86 novae discovered mainly by Hubble in the Andromeda nebula, this means that the conditions are most favourable for the production of novae in the central regions of the Milky Way and of the spiral nebulae. Many novae
settle down again as faint variable stars of about their initial mag nitude, showing very little sign if any of the upheaval through which they have passed. The range of magnitude involved in their outbursts may be from 9 to 14 magnitudes, the rate of rise just before maximum varying from 7.6 to 0.m2 per diem, the rate of fall, per diem, being from io to 5o times as slow. The brightness at maximum is probably about —5.o absolute magni tude, though values have been found ranging from o•o to absolute magnitude. In general it may be said that the more violent the outburst, the greater the contrast between the rates of rise and fall, and the more rapidly the star runs through its spectral changes.
To explain the startlingly high velocities of the out-rushing gases, reaching at times the value of 3,000km. sec. or the velocity of light, E. A. Milne has advanced the theory that a sud den brightening of a stellar nucleus, with the accompanying in creased pressure of radiation, involves the blowing out of the gases of the outer atmosphere of the star with accelerations which might easily lead to limiting velocities of the values found in novae. Atoms just balanced under gravity and radiation pressure are driven outwards by the increase in the radiation pressure ; they then absorb light of a shorter wave-length coming from the star and are no longer screened by the intervening layer of atoms which absorb light of the wave-length which has been mainly in volved in supporting them before the outburst. A stream or shell of atoms thus driven out gives the dark lines displaced to the vio let which are seen by the distant observer. It is possible to ex plain in terms of Milne's theory many of the miscellaneous com plications observed in the multiple dark lines with their varying shifts, but, as special assumptions have to be made for each com plication, it is wiser at this stage of our knowledge not to strain the one hypothesis but to wait the ripe time for further simplifi cations.