The variations in luminosity in these stars increases abruptly following an eruption which effect all or part of the star’s atmosphere. This increase is accompanied by a rise in temperature and so by major variations in the spectra, when the maximum brightness has passed, the star dims and slowly cools.


Fast Nova

These stars are characterized by an extremely abrupt rise to maximum, some increasing more than ten magnitudes n one day. The very speedy rise means that it is not often observed and even then only in its upper part. The decline is also fairly rapid. An arbitrary but practical criterion to characterize them has been created: this is based on the time T(3) or the nova to lose three magnitudes with respect to the maximum. Oscillations have shown up in some nova during their decline, some as long as several months induration and over a magnitude in amplitude. There has been a large number of fast nova: out of 121 galactic nova whose type has been determined, 82 belong to type Na.

Slow Nova

The nova Del 1967 (HR Del) is a good example of this type. The star passed from magnitude 12 to one of six in one month and then continued to increase slowly with the maximum magnitude of 3.7 being reached five months later. The decline was even slower. These stars are less numerous than fast nova, only 32 being known out of the 121 known nova of all types. The difference between slow nova and ultra-slow nova is not at al clear-cut but ultra-slow nova are usually classed on their own as novoids.

Recurrent Nova

Recurrent nova are stars that show repeated nova-like variations. The brightest recurrent nova is the brightest with to maxima of magnitude 2. The most active has been T Pyx, whose five maxima were separated by 12, 18, 24, and 23 years.

It is important to point out that the maxima of a given star are the same as regards both the amplitude and the shape of the light curve. This is clearly shows, as we have seen, that the occurrence of the nova, in spite of its violence, does not appreciably alter the structure of the star.

Sometimes the spectrum of the star shows that it is other than it appears to be. WZ Sagitta is such a case. Long considered a recurrent nova, with three explosions 1913 (mag 8.5), 1946 (mag 8.7), 1978 (mag 8.7). However spectroscopic observations made during its last maximum showed that it is not to be classified with the recurrent nova but with the dwarf nova. In addition, it exhibits a spectroscopic oscillation common to dwarf nova. Finally, its absolute magnitude of approx. +10 is different from that of nova.

Spectra of Nova

The spectrum of a nova, before and during maximum brightness, is not unlike that of a supergiant star of Class B, A or F, with the absorption lines greatly shifted toward the violet. After maximum the spectrum changes rapidly with the development of wide, complex bright lines, un-displaced in the mean from their normal positions; complex structure, corresponding to several different velocity shifts, also appears in the absorption lines, which become progressively fainter as the bright-line spectrum increases in intensity. It has long been realized that the violet shift of the absorption lines is a result of a rise of the absorbing material, and this upheaval of the star's surface helps to account also for the brightening of the nova, since it is evident that the rapid increase in brilliance cannot be attributed solely to increase in surface brightness; the area of the radiating surface must be increasing as well.

It has been suggested that a nova outburst begins with the expansion of the star as a whole, and that near maximum brightness the star ejects a shell of nebulous matter, and then contracts and grows fainter. Though this model interprets many features of the behavior of novae, it fails in several important points. Another model supposes that the expansion is confined to the photospheric layers, which are blown away from the stellar surface at the time of the first outburst. The star then becomes a sort of planetary nebula, in structure if not in physical conditions. In the early stages the shell should still be dense enough to radiate like a stellar photosphere, receiving diluted high temperature radiation on its inner surface, and reradiating the transmitted radiation undiluted at a lower temperature. The observed progression of the nova spectrum from an absorption spectrum to a bright-line spectrum, which soon develops the nebular lines and ultimately shows Wolf-Rayet characteristics, is in general harmony with this model. More observations are needed to determine the correct model, whatever that turns out to be.

Secondary Phenomena

Amplitude and Absolute Magnitude

Nova have varied amplitudes that range from 7 to more than 19 magnitudes, but the value cannot always be determined since the star is often very faint at its minimum. Nevertheless, the amplitudes are known for 76 stars. There are two peaks in the frequencies with which they occur, one for an amplitude of nine magnitudes and the other one of 12. However, it is probable that large amplitudes are more common but are not known.

Recurrent nova have very small amplitudes ranging from eight to ten magnitudes. Slow nova can have large amplitudes, but nova with amplitudes greater than 13 magnitudes are mostly fast nova. Note that tis figure does not include all nova and omits the largest ever Nova Cyg 1975, which is greater than 19 magnitudes. This star is considered by many to be an exception – intermediate between a nova and a supernova.

Much work has been done towards the establishment of absolute magnitudes. Standard methods for determining distances cannot be used at the distances of nova. Other measurements, such as the intensity of interstellar lines, (intensity increasing as the distance of the object increases) and secondly by obtaining the apparent velocity of expansion of the nebulosity, (enabling the distance to be known if radial velocity of the gases has been determined).

Two groups can also be detected by considering the maxima, one with absolute magnitudes around -6 and the other around -9, and these correspond to the two groups in the distribution of amplitudes. This shows that there is a correlation between the absolute magnitude at the maximum, the speed of decline T(3), and the amplitude; the very fast nova and this of large amplitude are also those with the greatest luminosity at maximum.

All these results are corroborated by the observation of nova in the Andromeda galaxy and in the Megallanic Clouds. They also have two peaks in their frequencies of occurrence around -6 and -9, and there is also a correlation observed between T(3) and the absolute magnitude. There are therefore no different from the galactic nova.

Variations at the Minima

The pre-nova are generally not very well known, this is not surprising since it is not possible to predict which stars will become nova. However some nova had been known to be variables and so there is a pre-nova history on some stars.

These stars are obviously followed more closely in their post-nova phase. Some f them have fluctuations that are occasionally appreciable, with some sort of small secondary maxima of short duration but which may exceed one magnitude.

High precision photometry has revealed another type of variation that we shall find in many eruptive variables: this is “flickering,” which consists of small rapidly varying flares following each other without interruption.

RS OPh shows a semi-regular variation (P = 70 days) at an amplitude of 0.6 magnitudes. This confirms that there is an M giant in the system linked to a blue star.


In 1954 it was shown that DQ Her (nova 1934) is an eclipsing binary with a very short period of 4h 39m. Since then all the nova bright enough to be clearly observed have been shown to be double. In this case, therefor, doubling is a general feature.

These binaries are formed from a red star that is large but no very massive and a blue star of high density, which resembles a white dwarf. This dissimilar pair is generally closely bound and has a very short orbital period, usually a few hours.

There are several exceptions to this structure, some pairs have a red component that is a giant star and other cases it is a sub-giant; the pair containing a sub-giant having a much longer rotational period.

In some of these binaries small changes in period which arise from variations in the two stars has been detected. The most interesting case is that of V1500 Cyg. The period has changed from 0.1410 days at the beginning of September 975 (the time of the explosion) to 0.1399 days at the end of October and to 0.1384 days in May-June 1976. The period of the binary system may thus have been changed by the violence of the explosion.

The Cause of Nova

Much thought has been given over the last few decades to the cause of nova. The consensus of opinion is that the basic process of the explosion is the same as that for man-made nuclear weapons. To begin with, energy is produced at a comparatively gentle rate in material that is covered by a layer of relatively inactive other materials, tamped as one says. Because of the tamping materials the energy produces by nuclear reactions cannot escape and must therefore accumulate within the reacting material itself. Provided the tamping effect is strong enough, the rising temperature and pressure causes the nuclear reactions to become less and less gentle as the process proceeds. The same cycle of events is repeated and repeated again with the energy released and repeated again with the energy released from the nuclear reactions accelerating at an ever-increasing rate, until eventually the situation gets quite out of hand. Or at any rate until the tamping effect of overlaying material fails at last and the material is blown entirely out of the strong gravitational field of the white dwarf star, its chemical composition, and the amounts of the reactions and the tamping materials, the details of explosion can vary in ways that are subject to mathematical calculation, and which have been found to agree with many of the observed features of nova.

The mechanism causing a star to become a nova can be broadly describes as follows. Consider a pair of stars one of them a large reddish star of low density which has reached or exceeded its Roche Lob, and the other a very dense white dwarf. Some of the material of the red star escapes and is transferred to the white dwarf. The transfer takes place in two stages: the material first falls into a kind of disc or ring which surrounds the white dwarf known as an accretion disc. At a later time, the matter which is attracted by the strong gravitational force from the whole dwarf leaves the disk and falls on to the dwarf at high speeds and with great turbulence.

The arrival of the gas at the accretion disc, already hot and becoming hotter still as it falls on to the white dwarf, provokes the eruption: a powerful nuclear explosion is produced in the atmosphere of the star, and the white dwarf suddenly ejects the blanket of foreign matter which covers it: this is the beginning of the phenomenon known as a nova.

The mass of the ejected gas is relatively small: 0.0001 to 0.00001 of a solar mass. On the scale of the stars involved, this is extremely minute and the structure of the star is not affected by the explosion, yet it is as much as several times or several dozen times the Earth’s mass. This is the reason that recurrent nova are possible.

Galactic Distribution

It is interesting to see how nova are distributed through the Galaxy. The distribution of longitudes is remarkable: out of 161 certain galactic nova, 74 occur between longitudes 345 degrees and 15 degrees, in other words , at least 15 degrees from the galactic center. Nova are thus particularly numerous in the direction of the galactic halo. As regards latitude, nova situated in the direction of the center are never more very far from the galactic plane. In other directions, several nova are known that are quite a long way from the galactic plane, but very few are more than 1500 parsec from it.

All his leads to the conclusion that nova form an intermediate Population II. However, they do not seem to form a homogeneous population. Two nova are known in globular clusters that we know form a typical Population II. In addition they have been found in three elliptical galaxies, which also form a typical Population II.

Many nova have been observed in nearby galaxies: 200 have been found in M 31 alone, which is as many if not more than are own Galaxy (they are on average of magnitude 16-17). Four nova are known in the Small Megallanic Cloud, six in the Large Megallanic Cloud, and a significant number in other galaxies. In all over 500 galactic and extra-galactic nova are now known.

Finally it should be pointed out that the nova observed in other galaxies are roughly of the same magnitude (-6 to -9) and have the same type of light curves as those in our galaxy. These nova are sometimes used to determine the distance of the galaxies in which they are observed. The results are uncertain, however, since the relationship between the speed of decline and the absolute magnitude is not rigorous one.


The Galactic Supernova

Supernova Remnants(SNR)

Extragalactic Supernova

The Spectra

Frequency of Supernova

The Origin of Supernova

The Role of Supernova

Frequency of Super Explosions


Under the general title of novoids we group together stars with very different characteristics but with a single feature in common that they are eruptive variables. Some of them have been classified as one of three types.

Gamma Cassiopeia stars consist of B stars: rapidly rotating subgiants or dwarfs with emission spectra. The variations are often small in amplitude.

Z Andromeda stars are symbiotic star. These are binaries consisting of a red star similar to a long-period or irregular variable linked to a hot star.

S Doradus stars are supergiant with emission spectra showing irregular and often large variations.

Gamma Cassiopeia Stars

Gamma Cassiopeia was not considered a variable with a steady magnitude of 2.20. In April 1937, it passed through a maximum (visual mag 1.47) and then declined with small oscillations until May 1938 (mag 2.52). A short period of growth restored the magnitude to 2.2 and then the decline resumed (mag 3.0 in 1940). Since then it has oscillated irregularly between 2.2 and 2.7.

This Be type star has a B spectrum with intense emission lines (Balmer series) which vary continuously in strength as noted in 1933. Variations in the line profiles showed a periodicity of 0.70 day. Some very large fluctuations in the emission intensities were observed during the maximum of 1936-37, and at the same time absorption lines appeared.

Another well-known Be star is Pleione (BU Tau) in the Pleiades. Its variability was suspected in 1880 and confirmed in 1936. Its fluctuations, sometimes rapid, are accompanied by significant variations in color index. Its amplitude of 0.7 mag is smaller than that of Gamma Cas.

Although Be stars are plentiful, with 10-20% of B stars having an emission spectrum, not all of them are variable. The amplitude of these stars rarely exceeds 1 magnitude and is often only a few tenths of a magnitude. The spectra range from O8 to B8 with a maximum frequency of occurrence of around B2-B3, and it should be noted that the intensity of the emission increases as the star becomes hotter. These stars are subgiants or dwarfs with absolute magnitudes from -2 to -4.

They are young stars that are rotating very rapidly (more than 300km/s at the equator in the case of Gamma Cas. O. Struve reported in 1942 that the gravitational force is only just sufficient to keep the material rotating with the star. If this is true, only a slight internal pressure is needed for the equatorial zone of the star to eject material and form an envelope or shell.

Although this could produce some of the photometric variation, it is now believed that this is not the only cause. The variation in brightness remains a mystery that is partially unexplained even today.

It is interesting that these stars have effectively the same spectra and the same absolute magnitude as the pulsating variable stars of the Beta CMa type. Stars that show the two kinds of variation of the same time are now known, so that there is a connection between these two types of stellar groups. IT is, however, a connection that is still not very well understood, and a long series of photometric and spectroscopic observations are still needed to establish it conclusively.

Z Andromeda Stars

Z And normally varies between magnitudes 10 and 11, with the fluctuations having a semi-periodic form, while at very irregular intervals ir can increase to magnitude 8.5 It is the prototype of what is called a symbiotic star, for the emission spectrum of a hot star superimposed on the normal M2 III type spectrum. It is therefore a binary system, but the blue component can only be spectroscopically separated with difficulty because it is very faint compared with the red component.

About 30 stars of this type are known. In some cases the variations are irregular while others (AX Per and AG Dra) show a certain degree of periodicity. All these stars have a red giant spectrum sometimes K but mainly M) linked to a much weaker emission spectrum from a blue star. Bright lines belonging to highly ionized elements (He, O, Fe, etc.) can also be observed in the spectrum.

It is believed that these stars are similar to long-period variables, with the presence of a very hot and variable companion producing perturbations. In addition, some of them are surrounded by a gaseous nebula which is reminiscent of planetary nebula. The periodic or pseudo-periodic variations are due to the red component, while the blue star is responsible for the greatest maxima that occur at irregular intervals.

It has been shown that some of these stars (CH Cyg and EG And, have rapid variations similar to the flickering we have already mentioned, and attributable to the blue component.


A pseudo-nova or very slow nova is the name of a star with an eruptive variation of amplitude comparable with that of a nova but sometimes extending over several decades. This is by no means a homogeneous group, and so a few individual cases are described below.

At the beginning of 1909 RT Serpentis was fainter than magnitude 16, bit in July of the same year a magnitude 13.9 magnitude was observed. In 1920 it reached magnitude 11.5. After that, the increase in brightness became very slow and underwent small fluctuations, but it continued to increase until 1924, when it reached 10.5. The decline then started, with appreciable oscillations. In 1941, Rt Ser had a magnitude of 13.5 and it has since fallen below 15.

The spectrum first resembled that of a slow nova with an intense emission spectrum including hydrogen and some forbidden lines (O, Fe, etc.) showing a high degree of ionization. In 1931, the nebular spectrum characteristics of nova in decline made its appearance and this was observed until the star became too faint for its spectrum to be recorded.

The star RR Tel, was discovered by Flemming in 1908 and was thought for a long time taken to be semi-regular variable. Until 1930, it varied more or less regularly with a period of 383 days and an amplitude of 1.5 magnitudes. In 1930, the variation became less regular and amplitude increased. In 1944, the pseudo-periodic variation stopped, and the star which had until then oscillated between 13 and 15.5 magnitudes rose to 12 and then to 7 in 1948. It remained stable for several years but a decline started in 1958 and still persists. Examination of the Harvard photographic library showed that it had already produced a maximum in 1989, but one that was weaker (magnitude 9) and of shorter duration.

The spectroscopic study of RR Tel was made from 1949 onwards. At that time, its spectrum was that of an F supergiant with emission lines, similar to that of Eta Car. In 1915, the nebular spectrum appeared, but more recent spectra show a resemblance to the symbiotic stars.

RT Tel cannot be considered to be a nova since the phenomena is too slow to be called an explosion. It is more a continuous but moderate ejection of stellar matter, with the formation of an extensive gaseous envelope responsible for the observed nebular spectrum.


The star V Sge oscillates around magnitude s 12-13 but sometimes shows periods of high activity during which it increases to magnitude 9.5. In addition, it shows the “flickering” effect like the nova at their maximum, that is very rapid variations of small amplitude attributable to flares. It is an eclipsing binary with a period of 12 hours and amplitude of 0.6 magnitude.

The spectrum resembles that of nova at the end of their brightening phase; a nebular spectrum with intense emission lines (H, ionized He, and highly ionized C and N). The high radial velocities indicate ejection of matter. V Sge is considered to be an old nova, but it is not the only one. Several others have similar appreciable photometric variation, such as EM Cyg, which varies from 11.9 to 14.4 in B. Other stars are quieter and, apart from eclipses, show hardly anything but the flickering (VV Pup and UX Uma). One of the most interesting could be MV Lyr; it usually oscillates between 12.1 and 14.0 but sometimes rises to 10.5 However in August 1979 it was discovered to have fallen abruptly to magnitude 18, only to rise again to 14 several months later. This star is similar to AM Her but has a greater amplitude.

S Doradus Stars

S Doradus is a strange star. For more than 60 years it remained around magnitude 8.6 with small fluctuations, but occasionally it showed decreases in brightness of about 1 magnitude. The variations were not at all periodic: minimums were observed in 1891, 1900, 1930, 1940, and 1955. There was another in 1964, deeper than the rest with the star falling to magnitude 11. The most curious feature is that, this time, it did not rise as usually to its normal maximum but has oscillated ever since then around magnitude 10.

The spectrum of A Doradus is that of an A5 supergiant with emission lines similar to that of P Cyg. S Doradus is not a galactic variable but belongs to the Large Megallanic Cloud. It is extremely luminous: taking into account of the distance to the Large Megallanic Cloud, its apparent magnitude corresponds to an absolute magnitude of -10.2, making it a luminous as the most powerful nova at their maxima. At about 60 solar masses, it is a very massive star but, given the intensity of the radiation emitted it must be evolving rapidly. Wolf estimates that it has lost one solar mass in 1500 years. Note also that is has lost 1.5 magnitudes during the century since it was first observed.

P Cygni

Another interesting object classified as an S Doradus star is P Cygni. This was observed in 1600 and was then of the third magnitude. It declined and disappeared to the naked eye between 1620 and 1626, but reappeared in 1655 at a magnitude of 3.5. It oscillated between 4.5 and 5.5 for the entire 18th century and is now at magnitude 6.

P Cygni has a spectrum that is peculiar to this star and is known as the P Cygni spectral type. Strong and very broad emission lines are bounded on the violet side by absorption lines to the expanding gaseous envelope surrounding the star. The radial velocity, also variable, oscillates between 50 and 240 km/s.

P Cyngi is very luminous: it is 2100 parsecs away and this gives a maximum absolute magnitudes of -11.9 and a current one of -8.9.

In 1953, a variable was discovered in Messier 31 which oscillated slowly between the magnitude 15.1 and16.5. Since the distance of this star is known, an absolute magnitude of -9.5 could be determined. Four variables of the same type were then discovered in Messier 33 fluctuating between magnitudes 16 and 17, giving absolute luminosities of the same order. Eight more were discovered in 1972: three in Messier 31 and five in Messier 33, one of which has an absolute magnitude of -10.3 at maximum. The mean has been established as -8.8 for Messier 31 and -9.0 for Messier 33. One AE And in Messier 31, has a spectrum similar to Eta Carina.

Dwarf Nova

Dwarf nova share many things in common with nova. Some areas of difference are less spectacular variations in brightness as well as frequent explosions of limited amplitude found in the dwarf nova versus a single or very few, very infrequent explosions of with large variations in brightness. There are two categories distinguished according to the form of the photometric variations:

U Geminorium stars: the normal state of these stars is at their minima. At intervals of time varying widely from one star to another (from ten to several hundred days), there are abrupt maxima with amplitudes ranging from three to six magnitudes.

Z Camelopardalis stars: these stars hardly ever stay at their minima; intervals between maxima are usually from 10 to 25 days and the amplitude is from two to four magnitudes. In some cases, activity stops almost completely and the star undergoes what is called a “standstill.”

In spite of these observed photometric differences, the two categories seem to form a single physical group. U Gem, discovered in 1855, was the first known variable of this type. Normally faint (B magnitude = 15.0), it had maxima taking it to magnitude 9. The interval between two maxima can vary a lot (from 60 to more than 200 days), the mean being 103.4 days.

The variability of Z Cam was discovered on 1904, but it was only in 1923 that its peculiar photometric character was reported. Ot varies from 10.4 to 14.8 magnitudes in B, the mean time between maxima is 50.2 days. However , at irregular intervals it has “standstills” during which it undergoes very small oscillations around magnitude 11.5.

Var Assoc. w/ Galactic Nebula

Stellar Associations

Classification of Variable Stars Associated With Nebula

Given the variety of light curves among the stars, it is difficult to establish a simple photometric classification for them. For all the variables that are associated with diffuse nebula, the General Catalog has therefore introduced a classification that uses physical characteristics. Three main types exist.

In Stars

These include irregular variables that are visibly connected to a nebula or located in its immediate neighborhood. These stars are dwarfs or sub-giants. There are several sub-types.

Is Stars

These have the same photometric and spectral characteristics as those of In stars but are not apparently connected with the nebula. There are several sub-types.

UVn Stars

These include flare stars connected to the hydrogen clouds.

Frequency and Amplitude of Flares

The variation in brightness of these stars is often irregular and very rapid. An analysis of the fluctuations of some T Tauri stars has shown that they have flares that succeed each other very quickly and become superposed. However, other stars have different variations: in some cases such as T Cha, RU Lup, the fluctuations have a quasi-periodic form. It can be seen that belonging to types In or Is, that is, to be connected or not to a nebula, introduces no systematic differences in the form of the light curves.

The amplitudes of the variations frequently exceed 2, 3 and even 4 magnitudes but it is generally smaller for stars with earlier types of spectra (Ina or Isa types) than for the others.

The Orion Complex

The magnificent Great Nebula in Orion and very familiar to amateurs, is highly complex. The Great Nebula is actually several nebula forming a physical grouping. The oldest known of these, which is also the brightest and largest, is Messier 42 (NGC 1975) discovered in 1610 by a student of Galileo. It is surrounded by several others, such as Messier 43 (1984), NGC 1977 and NGC 1999. The whole array covers a vast region of several square degrees corresponding to more than 30 parsecs. The nebulae include an O association called Orion O1 and a T association Orion T2.

In 1656, Huygens noted a group of four quite bright stars in Messier 42 forming a multiple physical system, Theta (2) Orionis, called the trapezium because of its shape. The four main stars in this group are: A is n eclipsing binary (V1016 Ori) of long period (197.5 days); B is also an eclipsing binary (BM Ori) with a period of 6.470 days, while C is a spectroscopic binary.

There are numerous stars near or within the nebula. A catalog in 1863 contained 1101 stars. Other catalogues have been published since then, the most important being that of 1954 which includes 2982 stars in a region of 9 square degrees. It should be added, however, that all are not linked physically to the nebula; some are nearer or more distant stars which lie in front of or behind the nebular formation. The firs variable star in the region was T Ori. The search for variables became systematic after 1890. One of the most important of the extensive number of surveys done was done from 1950 onwards in Mexico and a second was carried out over the same timeframe in Italy. These two series were continued into the mid 1980’s.

The research was fruitful: it totaled nearly 780 variables in this region, the most of small magnitude (15018) and often difficult to detect amongst the bright nebulosity. The best known of these variables is T Ori. The other variables are similar. It was pointed out that the light curves differ from each other in these variables; some stars remain consistently near the maximum with a minimum from time to time; with others, the converse occurs; for others again, the fluctuations are completely unpredictable, with no preference for increases or decreases.

The Orion region contains a large number of UVn-type flare stars, with about 130 of them being known out of a total of 780 variables. These stars have late spectra from dK5 to dM and are fairly faint, most having minimum photographic magnitudes of 16, 17 or 18 and hence absolute magnitudes of +8 to +10.

The amplitude of the flares is fairly large, sometimes 3-4 magnitudes. However, they differ from the flares of typical red dwarfs in that they are less rapid. They often last an hour or two, whereas those of typical red dwarfs of the UV Ceti type sometimes end in a few minutes. For some stars the flares are even slow.

The flares in the Orion complex are not distributed like In variables. They are distinctly less concentrated in the central part of the aggregate; their distribution is more uniform than that of In stars. This difference probably reflects a different origin: the stars in the two groups are not “born” under the same conditions.

Some Other Nebular Regions

The table below gives details of 12 of the largest and best observed nebular regions similar to the Orion complex. The number of variables reflects those in 1987 and has certainly increased since then. As noted previously T-associations are of various sizes, some having 20, 30, 40, parsecs or larger, and sometimes containing more the 100 variables. On the other hand, there are small associations such as Cephi T1 and others still smaller. In some cases, nebulae called cometoids have been detected with the shape of a comet’s tail: NGC 2261, containing only 3 variables (of which on is R Mon) is like this, and so is NGC 6729, a nebula with 7 variables, one of which is R CrA.

Among the important stellar associations is Monoceros T1, which includes NGC 2264: it is a vast (40 parsecs across) and contains more than 200 variables. Then there is Sagittarius T2, the well-known Messier 8 – the Lagoon Nebula. Lastly, the stellar association Cygni T1 is fairly large and contains many flare stars; it includes a beautiful collection of nebulae, of which the best known is NGC 7000 (the North America Nebula).

It has been found that there are groups of associations. For example, the association Orionis T3 (which is included in the Horsehead Nebula [IC 434]) and its surroundings are situated at the same distance as T1and T4 in the north of Orion, for Tuari T2 and T3, and for others also. This can be explained by the fact that all the stellar associations occur in the arms of the Galaxy where hydrogen clouds also occur; it is conceivable that one cloud of enormous size could give birth to two stellar associations separated by several tens of parsecs.

FU Orionis Objects

An object discovered in the north of Orion in 1939 that was first considered to be a slow nova. In 60 days, this object (since designated FU Ori) passed from magnitude 16.5 to 9.7. It remained at this brightness for about 600 days, and then over the next 50 days declined to magnitude 10.3, at which it stabilized. It must be emphasized, and this is an important point, tht FU Ori is connected with a nebula, B35, and that it belongs to a stellar association. Orionis T4. This shows that it is not a nova, and in any case its spectrum is very different from that of a nova: it is currently that of a type F supergiant (F2p I-II) with the H emission line of hydrogen and metallic lines, some of which are unusual, such as those of lithium.

For a long time, FU Ori was regarded as a unique object. However, in 1971 a similar example was found, that of V1057 Cygni. It is interesting that this star was already classified as a T Tuari type variable” for more than 50 years, it oscillated irregularly between magnitude 15 and 16. Then in 1969, its brightness increased and the star reached magnitude 10 in 430 days. After that, there was a slight decline during which it oscillated between 10 and 11, and in 1971 was of magnitude 10.2.

The star is associated with the North American Nebula (NGC 7000), to which it is linked by a fine filament. Its spectrum was initially of late K type, with the H emission line. It has since changed considerably: at its maximum, it was A1 IV, with an abnormal abundance of certain metals, particularly lithium. Since then it has changed further and currently it is of type F, resembling that of FU Orionis. Racial velocities are of the order of 30-40 km/s.

Another case is PV Cephei. Discovered in 1975, this star is associated with the diffuse nebula NGC 7023 and passed in about 18 months from magnitude 18.5 to 15, and then stabilized. Some weak nebular filaments which surrounded the star have disappeared! It’s very faint spectrum is not very well established but it has nevertheless also been observed to resemble that of FU Orionis.

PU Vul was discovered in 1979, on old plates it appears at magnitude 16, increased to magnitude 15 in 1977 and to 8.6 in April 1978. It remained more or less stable for nearly 2 years and then fell to a photographic magnitude of 14 towards the end of 1980. A second maximum occurred, bringing the star at magnitude of 8.5 in V in October 1981. By chance, the spectrum had been obtained before its increase in brightness: it was then of type M4. Then at the beginning of 1979, just after its maximum, it had become a type A4. One year later it was found to be F5, with hydrogen and sodium emission lines and many metallic absorption lines. The radial velocities are fairly small (30-40 km/s).

This is a very special class of star: in a few months or a year, it gains 6 or 7 magnitudes, which corresponds to a ratio of 250 between the maximum and the minimum brightness. It then loses between 0.5 and 1 magnitude. At the same time, its spectrum passes from a late K or M type to a type A at maximum and type F slightly later, implying a large increase in the effective temperature. Finally, the small measured radial velocities show that there is no question of an explosion. What may appear most surprising are the size and great speed of the change in the spectra.

Herbig-Haro Objects

The first of the Herbig-Haro objects were discovered in 1954 in the region of the Great Nebula in Orion. Considerable changes on the photographs taken between 1947 and 1954: certain small nebulosities visible in 1947 had become fragmented seven years later and had assumed a quasi-stellar appearance.

Since that time, a large number of Herbig-Haro objects have been discovered in various regions. In all cases, the changes have been rapid. Observations made at various wavelengths, but particularly in the infrared, makes us think that what is being observed is a gaseous envelope surrounding a protostar that is still very young and for the moment invisible.

Star Factories

Evaluation of the age of clusters and associations has been made possible through an evolutionary theory by Hayashi in 1961. It shows that the positions of clusters and association stars in the H-R diagram are a function of their age. These stars therefor only have to be located in the H-R diagram and this is possible if good measurements of their magnitudes are available in several colors to ensure precise color indices. The age of the formations can then be estimated.

It has been shown in this way that O and T associations are young or very young objects. The oldest have not been in existence for more than 10 million years and some associations are even younger, for instance Cygni T1 (1 million years) or Mononceros T1, associated with the nebula NGC 2264 (400,000 years). Some stars in the associations are even younger: 10,000 years or less.

The presence of an element such as lithium in the spectrum is very important. The atomic nucleus of lithium is very fragile and is destroyed by nuclear reactions taking place in stellar interiors. The presence of large quantities of lithium in the atmosphere shows that a star has not yet reached the stage of nuclear reactions. It is therefore not a star in the astrophysical sense of the word, but a protostar; an object still in the contracting stage. The contraction produces heat which, according to Hayashi, is removed by convection in the interior of the protostar, whose radiated energy is therefore of gravitational and not nuclear origin.

It is inferred from this that T Tauri objects are not yet stars but protostars. Observations imply very small masses of about 0.2 – 0.3 solar mass, but a large diameter of 1 to 5 solar masses, since they are not completely contracted. A protostar is often surrounded by a ring of nebular material and also by a cloud of silicate “granules” which are stable at those temperatures (1000K). Observations of the infrared region carried out over several tears have revealed the complex nature of the phenomena occurring around the stars.

In many regions of the sky, small dark globules can be observed that are easy to detect since they form dark spots against the background of bright nebula. They were discovered in 1947 by B.J. Bok and E. Reilly in Messier 8, and since then a large number have been observed in nebular formations. We now begin to understand the structure of these objects whose dimensions lie generally between 0.3 and 1 parsec. At the center there is a protostar or even several; on the outside, a sort of “cocoon” masks the protostar, absorbs its radiation and re-emits it in the form of infrared radiation. The globule thus constitutes a primitive pre-stellar stage.

Infrared observations have revealed the existence of objects much more compact then the globules. These have been known for some 20 years, but observations made using satellites, and particularly IRAS, have enabled a large number of them to be detected, especially in the region of Orion, in the clouds surrounding Rho Ophiuchus and in a large star cloud, Chameleontis I, which contains no less than 70 infrared sources. These objects are interpreted as protostars already at a more advanced stage than those inside the globules.

Where can the FU Orionis type objects be placed? Since the presence of lithium excludes nuclear reactions , at an advanced stage of contraction, the temperature of the protostar must be high enough to dissociate the molecules in the cocoon that surrounds it. This cocoon collapses in a few weeks or months, and when it has disappeared it is the much more luminous protostar that we observe. The increase in brightness in thus interpreted as due to the breaking up of the cocoon. After that the protostar shows only small variations.

Nebular regions are the seat of extremely complex phenomena, one of which is of particular interest within this context. One phenomenon is the interaction between the gas and the stars, which is not yet very well understood. It is nevertheless responsible at least in part for the variations of brightness in stars. In some cases, using radio waves, “jets” of material have been detected that are responsible for appreciable radio-wave emission.

There is also the fact that these nebulosities provide us with valuable information about the formation of stars and the evolution of very young stars. Lastly, there is another interesting feature: radio-astronomical observations made with millimeter waves have shown that clouds also “manufacture” organic molecules. More than 50 of these have now been identified, and among some of the very complex ones are those necessary for life.

Some Nebular Regions
Stellar   R D Number of
Association Nebula (parsecs) (parsecs) Variables
Cep T1 NGC 7023 280 6 34
Cep T3 IC 1396 770 30 32
Cyg T1 NGC 7000 IC 5070 600 12 102
Mon T1 NGC 2264 790 40 205
Ori T1 SG 139 410 28 55
Ori T2 M42, M43, ... 400 34 778
Ori T3 IC 434, NGC 2024 400 28 130
Ori T4 S 280 410 25 35
Sgr T2 M8 (NGC 6524) 1300 23 69
Sgr T1 M16 (NGC 6611) 2300 10 24
Tau T2   170 18 20
Tau T3   170 15 81

Variable Red Dwarfs

Discovery of the UV Ceti Stars

Red stars with low intrinsic luminosity (dwarfs with K or M spectra) frequently show variations in brightness. The variations may be of two types: rapid and irregular flares, capable of reaching several magnitudes but of short duration, the typical star of this group being UV Ceti; or periodic variations of small amplitude due to the presence of a small bright region or spot on the star, the typical star being BY Draconis. Some of the flare stars are associated with clusters and their characteristics are slightly different from those of typical UV Ceti stars. They form a third group.

These are among the faintest known stars and only those very near to us can be detected. Almost all those now known are in fact at least 25 parsecs from the Sun. The first star of those type was discovered in 1940, a faint companion of the star BD + 44 2051 (getting the designation WX Uma); the second in 1943, Ross 882, or YZ CMi. IT is curious that these discoveries remained almost unnoticed for several years. It wasn’t until 1948 when a flare was observed in the B star of a small binary L 726-8, since called UV Cet, for the phenomena of flares to gain interest. The binary L 726-8 is formed from two red dwarf stars whose spectra dM5.5e and dM6e show strong emission. Both stars are variable, the B star appearing to be the most active, and any observations made generally refer to the pair without it being possible to say with certainty which has produced a flare.

Discoveries of UV Ceti stars became more numerous in the 1950’s and at least 120 are now known within a radius of 25 parsecs around the Sun. The absolute magnitude of the UV Ceti stars is magnitude +8, but some have absolute magnitudes as low as +15 to +18. Put another way they radiate 10,000, 100,000, or even 200,000 times less radiation at than the Sun. Many of these variables belong to a binary system and in several cases both components are variable. If the pair is some distance from each other so that they can be individually observed they receive two designations, if on the other hand, the components are closely bound and difficult to observe separately, they receive only a single designation.

Close binaries are of great importance: a determination of their orbits enables the masses of the components to be obtained. The UV Ceti stars, very faint red dwarfs, are the least massive stars known, some of them with masses less than a tenth of a solar mass.

Observations of Flares

From 1950 onwards, several observers undertook a systematic surveillance of UV Ceti and a few other stars of the same type. Photographic observations proved to be very difficult as the flares have short durations and a very fast rise to maximum. A visual watch was therefore kept, which was tedious since the unpredictability of the variations mean that the have to be observed continuously. The observations did prove successful as several dozen flares were seen on UV Ceti, one of which was more that fie magnitudes.

It is only since 1966 that various observatories have undertaken systematic observation using photoelectric photometers, wither with one color or several colors simultaneously (usually U, B, and V). In addition, programs have been combined with radio wave observations or with some in the far-ultraviolet and x-ray regions using satellites. It has thus been possible to obtain an idea of the process responsible for the variations. The number of flares observed from UV Ceti exceeds 1000 and has reached several hundred for other variables, notably AD Leo, YZ CMi and E Lac.

Flares occurring in the same star can take different forms; in UV Ceti for example, the most common form is a single fast eruption with a sharp peak and a more gradual decrease in brightness. Some are preceded or followed by a secondary flare. Slow flares also exist. A similar variety is found in other stars.

With rapid flares the rise to maximum takes place in a few seconds. So a flare of 3.5 magnitudes has been observed on UV Ceti in which the rise time was only 6 – 7 seconds. The decline is significantly slower, but in most cases does not last longer that ten minutes.

The amplitude varies greatly from one flare to another in the same star, and it also depends a great deal on the color. So a flare observed in V is smaller than in B, which is smaller than in U: small variations that are difficult to detect in V are easier to observe at shorter wavelengths.

In some cases, the existence of what may be called anti-flares has been noted. The variation is as abrupt as that in flares, but is a decrease in brightness instead of an increase.

For a given star, the frequency of flares varies with time. Sometimes they succeeded each other in rapid succession and at others are separated by long periods. It also varies from one star to another; there is a correlation between frequency of flares and the luminosity of the star. The weakest stars are the most active; while in relatively bright stars such as CR DRs flares are very infrequent.

The Spectra and Mechanism of Flares

During the times when no flares occur, UV Ceti has spectral type M, characterized generally by hydrogen lines and those of ionized calcium in emission (dMe). It has been noticed that there is a distinct relationship between the intensity of the hydrogen lines and the eruptive activity of the star; variables with strong emission lines have many flares. It should be pointed out, however, that some stars such as FI Vir have spectra without emission.

A certain number of spectra have been obtained during a flare, and they are very different from normal spectra. There is a reinforcement of the continuous spectrum, particularly in the ultraviolet region, an increase in the intensity of the hydrogen lines with respect to those of calcium and an appearance of emission lines from neutral or ionized helium. The color temperature also changes, which explains why the amplitude appears greater in B than in V and greater in U than in V.

A flare is an explosion produces in the atmosphere of a star, analogous to the eruptions that can be observed daily in the chromosphere of the Sun. It has a relatively moderate energy output, on the order of 10^25-26 Joules. Nevertheless an eruption of sufficient energy to produce an increase in brightness of the Sun by 0.1%, would if transposed to a red dwarf 10,000 times weaker, produce a spectacular flare of several magnitudes.

The very short duration of the explosions indicates that the phenomenon is localized in a very small part of the atmosphere; it has been estimated that only 1-3% of the atmosphere is affected by it.

BY Draconis Stars

This group of stars also consists of red dwarfs, but it differs from UV Ceti group in the process responsible for the variations. Here there is a region on each star that is brighter than the rest of it, in other words, a luminous spot. As the star rotates, this spot passes in front of the observer and produces an apparent increase in brightness.

The effect repeats itself with a period equal to the star’s rotation. The amplitude is always small, no more than a few hundredths of a magnitude, and the variations can only be detected by photometric observations.

A few dozen of these types of stars are known. The spectroscopic period and the photometric period for each star are often different. Two of the spectroscopic binaries, YY Gem and CM Dra, show eclipses. These are two of the least massive eclipsing binaries known; the mass of the YY Gem pair is 1.30 solar masses and that of CM Dra is about 0.45 solar masses. BY Draconis stars are often more luminous than UV Ceti stars, most of them having absolute magnitudes between +7 and +8. The spectral type often lies between dK5 and dM0 but, as with UV Ceti stars, emission lines from hydrogen and ionized calcium are often observed.

It is important to note that most BY Draconis stars also have flares so that they behave at the same time like UV Ceti stars. On the other hand some stars classed as UV Ceti stars also have a periodic variation of the BY Draconis type. This is the case with YZ CMi, whose photometric period is 2.78 days. It is, therefore to perfectly separate the two types.

Flare Stars in Galactic Clusters

Many flare stars occur in regions that are rich in hydrogen clouds, particularly the Great Nebula in Orion. An appreciable number have also been detected in young galactic clusters. The most typical; case is that of the Pleiades cluster. A systematic search for flare stars in this region in 1964 and is still ongoing.

The search has been extremely fruitful as regards the Pleiades region with the detection of several hundred very faint variables in a few square degrees. Many other regions have also been studied, especially the Preasepe cluster in Cancer where about fifty have been discovered. The are also about three dozen in the cluster Melb. 111 in Coma Berenices, where about three dozen are also known. The older clusters are les rich: only eight flare stars are known in the Hyades.

All of these stars are generally very faint: so, the Pleiades variables have an absolute magnitude of +11, on average. The spectral types range from dK5 for the brightest to between dM4 and dM5 for the faintest.

Given their low intrinsic luminosity, these stars are detectable in only relatively nearby galactic clusters. At 100 parsecs a star of absolute visual magnitude of +11, has an apparent magnitude of 16 visually and 17.5 photometrically. At 400 parsecs, the figures would be 19 and 20.5.

The flares are slightly different from those of the UV Ceti stars, and resemble those of the variables in the nebular regions. The amplitude frequency exceeds 3 – 4 magnitudes in B and 6 – 7 magnitudes in U. The rise is but the total duration of the phenomena is often more than one hour. Lastly, the frequency of the flares is also lower than for the UV Ceti stars.

Some Comments on the Phenomenon of Flares

There is a correlation between the absolute magnitude and the frequency of the flares: the least luminous stars are the most active. There is a second relationship between the absolute magnitude and the duration of the flares: stars of very low luminosity have flares that last for shorter periods than those of more luminous stars. There is, on the other hand, no clear correlation between the amplitude and duration of the flares; weak flares lasting a long time are sometimes observed, as are large amplitude flares of short duration.

These relationships explain why the variables associated with open clusters or with nebula have slower and less frequent flares than those of the UV Ceti stars. It is because the former are, on average, brighter stars.

Being a member of a binary system seems to be an important factor; clearly, some single stars show flares but it is noticeable that the spectroscopic or visual binaries are often more active. It seems likely that the presence of a nearby star encourages the formation of flares. For the pair V 1054 Oph, the explosions are more frequent when the components are at their minimum separation.

Population Type and Galactic Distribution

All the stars encountered in the category of variable red dwarfs have one common characteristic: the existence of flares –in other words, explosions of moderate power and short duration located a small region of the atmosphere, but recurring frequently. They share this characteristic with the UVn stars, the flare stars associated with nebula, and also the T Tauri stars whose spectra are of an earlier type but in which the mechanism is known to consist of many superposed explosions.

All these variables form what it is convenient to call the “evolving sequence”; these are young stars (or even protostars like the T Tauri) in rapid evolution. They very often have emission spectra and are frequently associated with great expanses of gas. The closely follow the spiral arms of the Galaxy and hence belong to a young Population I.

It is noticeable that the oldest galactic clusters do not contain flare stars that those of medium age have a few of them and that only the young clusters include a large number. There is thus a correlation between the age of the cluster and the number of flare stars it contains.

Some UV Ceti variables are older than others. This is so for those whose spectra do not show emission lines, such as FI Virginis. They belong to an intermediate population, but they do not seem to be very numerous.