GALAXIES, IRREGULAR DEIDRE A. HUNTER Even within the great family of normal spiral galaxies there is a large variety of structures. As one looks along the Hubble sequence of spirals towards later and later type, one sees the nucleus of the galaxy become less pronounced, and from Sc-to Sm-type galaxies the spiral arms become more ratty. Eventually one arrives at the class of Magellanic-type irregular galaxies, which Hubble described as "lacking both dominating nuclei and rotational symmetry." The Magellanic irregular galaxies, designated as Im galaxies, are the end of the line, at least in terms of the Hubble sequence. However, morphologically, the transition from spiral to Im galaxies is a fairly smooth one. The Magellanic irregular galaxies, as their name implies, are chaotic in appearance, lacking the symmetrical spiral patterns (see Fig. 1). A galaxy can look chaotic because it has been disrupted by a collision with another galaxy. However, there exists a class of normal, noninteracting, intrinsically irregular systems, which comprise **-** of all galaxies. There is another class of irregular galaxies with an appearance that is very different from that of the Im galaxies: the amorphous irregulars. These galaxies are very smooth in appearance and are not resolved into the many luminous star clusters that give the Im galaxies their distinctive jumbled appearance (see Fig. 2). In fact, they more closely resemble elliptical galaxies, except that, unlike elliptical galaxies, they are blue, contain much neutral hydrogen gas, and are actively forming massive stars. The amorphous irregulars are rare compared to Im galaxies. THE MAGELLANIC IRREGULARS GLOBAL PROPERTIES Compared to spiral galaxies, the Im galaxies are smaller, less-massive systems. Consequently, they are also less luminous and have lower rotational velocities. The optical light that they emit is generally bluer in color, which indicates the relative importance of a hot, young stellar component. A larger fraction of the mass of Im galaxies is in the form of gas, the fuel for star formation, and this gas contains relatively less of the heavy elements (that is, heavier than He), the products of stellar evolution. These latter two characteristics indicate that the irregulars are less evolved than spirals in the sense that less of their mass has been locked up in or processed through stars. Because they lack the spiral density waves that are the cause of the spiral arms, the irregulars are also somewhat simpler systems. A typical Im galaxy has an absolute blue magnitude M* of -17 (range-13 to -20) and a mass of 10* M* (range -10*-10** M*, where M* is the solar mass). Average surface brightnesses within an isophotal diameter of 25.0 mag arcsec-* are 22-24 mag arcsec-*. Abundances of oxygen generally range from somewhat less than to * of the solar value. A typical late-type spiral galaxy has a maximum rotation velocity of -175 km s**, and most irregulars have values less than half this. Diameters to an isophote of 26.6 mag arcsec-* are 5-50 kpc. Star formation rates per unit area are 10** M* pc-* and the timescales to exhaust the current gas supply at the current rate of turning the gas into stars is 10*-10** yr. STELLAR CONTENT As a group the irregulars are generally the bluest of the normal galaxies, having average color indices of U-B=-0.3 and B-V=0.4, although there is a continuum of normal irregulars from a U-B=-6 to U-B=0. The blue color indices of irregulars correlate well with other indicators of young stellar populations and are therefore primarily indicative of the presence of massive stars. The global B-V color indices cover a smaller range than U-B, as is expected if the B and V light are primarily contributed by older, intermediate-age stars. Near-infrared colors of irregulars are even more homogeneous, with mean values of J-H =0.6 and H-K =0.2. These color indices are nearly the same as the averages of all types of spirals and ellipticals, which probably reflects the basic similarities in old and intermediate-age red stars. A few irregular galaxies are near enough so that the most luminous stars can be individually resolved. Color-magnitude diagrams for these systems are similar and show pronounced blue supergiant branches and more sparsely populated red supergiant branches. Thus, evolved massive stars are an important component of the stellar population in irregular galaxies. The normal and comparatively uniform properties of young, massive stars in irregulars is demonstrated by other information as well: (1) the optical stellar luminosity functions for irregulars are similar to one another, (2) Wolf-Rayet stars, the descendents of very massive stars, are found in many irregulars, and (3) ultraviolet spectroscopy of luminous star clusters reveal normal mixes of hot massive stars. The luminous young stars in irregulars are often seen to be superimposed on a spatially more extended, partially resolved sheet of red stars. The spatial smoothness of this sheet is consistent with an older stellar population in which the effects of individual star-forming events have been averaged out. Precise ages are difficult to determine, but the smooth sheet of red stars must originate from stellar populations having ages of *1 Gyr (1 Gyr =10* yr). THE INTERSTELLAR MEDIUM The Im galaxies are very rich in pristine atomic hydrogen and He gas; typically 20-50% of the mass of an irregular is in this form. Within the galaxies the gas is lumpy, which is expected because gas density enhancements are required to form stars. What is unusual is that in Im galaxies the gas often reaches far beyond the extent of the stars in the galaxy. In one galaxy, NGC 4449, the Hi extends to 10 times the optical radius of the galaxy. In spiral galaxies, on the other hand, generally less than 20% of the Hi lies beyond the optical galaxy. Because the density of the gas declines radially, the fact that star formation has not occurred beyond some radius in the gas indicates that there is a threshold gas density below which clouds do not condense and form stars. Another component of the interstellar medium of a galaxy is molecular gas. It is from clouds composed mostly of H* that stars appear to form. The symmetric H* molecule cannot be directly observed in quiescent clouds. However, it has been found that the easily observed CO molecule can be used as a tracer of H*. A proportionality constant is applied to the observed CO flux to infer the quantity of H* that is present. Observations of CO emission in Im galaxies, however, show that the CO fluxes are much lower relative to the star formation activity compared to spirals. The most likely explanation at present is that the proportionality constant used to deduce the H* content from the CO observation is inappropriate to Im galaxies, where the CO flux is low but the H* content is not. This conclusion is based on CO observations of individual molecular clouds in the nearest Im galaxy, the Large Magellanic Cloud. The velocity dispersion in a cloud is also a measure of the mass of the cloud, and a plot of the velocity dispersion against the CO flux for clouds in the Large Magellanic Cloud shows a linear relationship that parallels but is offset from that for Milky Way clouds. This is interpreted as evidence for a different proportionality factor necessary to convert CO fluxes to H* masses, and is consistent with theoretical expectations based on the overall lower abundance of heavy elements in Im galaxies. Dust is another component of the interstellar medium. In optical pictures of spiral galaxies one can see dark patches and lanes that are the result of obscuration by concentrations of dust. Such dark patches, however, are less prevalent in optical images of irregular galaxies. Generally the dust reradiates in the far-infrared wavelength region the starlight that heats it. From infrared fluxes one finds that the global dust-to-gas ratios are lower in Im galaxies than in spirals. Furthermore, the characteristic temperature of the dust that radiates at 60-100 ** is warmer and that which radiates at 12-25 ** is cooler in Im galaxies than in spirals. This is at least partially the result of the higher ultraviolet surface brightnesses and lower opacities of the irregulars. The grain characteristics, however, can also affect the temperature of the dust. All else being the same, graphite will be warmer than silicate materials and small grains will be warmer than large grains. If the average composition of the grains depends on the metallicity of the galaxy as a whole, this could be an additional factor in producing the observed properties of the more metal-poor irregulars. Global galactic kinematics also influence the state of the galactic interstellar medium. Most spiral galaxies rotate differentially, shearing the interstellar medium. Irregular galaxies rotate very slowly, often as a solid body, like a record on a turntable. Thus, there is little, if any, shear. In addition, noncircular and random motions seem to be significant in some Im galaxies. Furthermore, bar structures in the distributions of the older stars, seen in visible light, are frequently present and do not necessarily lie at the center of rotation of the galaxy. This adds additional perturbations to the velocity fields, and it has been suggested that gas flows at the ends of the bars may enhance the formation of large star forming complexes there. STAR FORMATION In the preceding section we saw that there are differences between irregulars and spirals in terms of various aspects of the interstellar medium. The heavy-element abundance is lower, the dust-to-gas ratios are lower, the CO content is lower, and the shear due to rotation is lower in irregulars. Because stars are forming from the interstellar medium, we might expect these environmental differences to have observable consequences on the star formation process. For example, theoretical models suggest that the temperatures and sizes of the natal clouds may influence the masses of stars that are formed; lower abundances of heavy elements may lower the efficiency with which a cloud of gas turns into stars, and a lower quantity of dust relative to gas and smaller dust grains may be necessary for the formation of very massive stars. However, an observational comparison of the optical and ultraviolet properties, sizes, luminosities, and morphologies of giant HII regions in Im galaxies with those in spirals shows that they are very similar. Thus, the differences in interstellar-medium characteristics between spirals and Ims have not led to obvious differences in the nature of individual star-forming complexes. Most irregular galaxies are too distant for us to readily distinguish any but the very brightest individual stars. Therefore, we must find a means other than counting stars to learn about how fast a galaxy is forming them. The most massive stars (**10 M*) are the shortest lived, so they act as tracers of the most recent star formation activity. The prodigious fluxes of ultraviolet photons from these stars ionize the natal gas around the stars. By relating the intensity of the optical atomic line cooling radiation produced by the ionized gas to the number of stellar ultraviolet photons needed to maintain the ionized gas, one can infer the number of massive stars that have formed recently (in the past 10 million years). However, stars of masses down to about 0.1 M* also form, and the lower-mass stars are more numerous, but they are cooler and do not ionize their natal gas. To find the total number and mass of stars recently formed, one must combine the number of massive stars observed with a function that describes the number of lower-mass stars expected for a given number of massive stars. This latter, called the initial mass function, is an empirical relationship that is determined from studies of the Milky Way and several nearby galaxies. If one uses the above method to measure the rate at which galaxies are converting gas into stars, one finds that Im galaxies have star formation rates that are comparable to those of spiral galaxies relative to their size. This implies that spiral density waves, which spirals have and Im galaxies do not, are not necessary for vigorous production of stars. In the past, the spiral density waves were believed to play the key role in triggering galactic star formation. As the wave passes through the disk, it compresses the interstellar gas into dense clouds from which stars would precipitate. The Im galaxies, however, show that star formation can take place without spiral density waves. Our empirical picture of star formation in Im galaxies is like that of a pot of oatmeal bubbling on the stove: Star formation moves around the galaxy with time. One region will form stars, deplete the gas locally, and die down; then later another region will become active. However, the bubbling is not entirely random. The star formation rate is not only constant over the lifetime of the galaxy in all but a few peculiar systems; but, on average over long periods of time, it is constant as a function of radius within the galaxy. When density waves were deposed as the sole mechanism for initiating star formation, there remained the question of what does trigger a region to form stars in Im galaxies. One model that addresses that question, the stochastic self-propagating star formation model (SSPSF), depends on the newly formed stars themselves. Massive stars affect the interstellar medium that surrounds them. They ionize the gas, emit tremendous ionic "winds," and eventually explode as supernovae. The energy dumped back into the interstellar medium must have some effect on the galaxy as either a positive or negative feedback. In the Large Magellanic Cloud and the nearby spiral M31, one can see instances where clusters of massive stars have blown kiloparsec-sized holes in the disk gas. In these cases the gas around the hole has been compressed and appears to have been induced to make stars. The SSPSF model argues that this mechanism allows star formation to propagate around the galaxy as one region after another ignites its neighbor. Thus, there is a causal relationship between a young star forming region and its older neighbors. However, although star-induced star formation has been seen to occur in a few cases, whether it occurs on a scale and at a frequency necessary to be the dominant mechanism in a galaxy's star formation activity is not now known. AMORPHOUS IRREGULARS Although the amorphous irregulars differ morphologically from the Im galaxies, in terms of most other global properties the two classes of irregulars are quite similar. Furthermore, because the amorphous irregulars are blue and contain ionized gas, we know that they are forming stars at the present time. Many amorphous irregulars, however, differ from the Im irregulars in yet another way: the spatial distribution of the star forming regions. In Im galaxies the star forming regions are generally distributed over the disk of the galaxy. In many amorphous irregulars, on the other hand, the star formation is concentrated to a single supergiant region located at the center of the galaxy. In NGC 1140, for example, the massive stars in the central star forming complex have produced a region of ionized gas that has a luminosity 100 times that of the supergiant HII region 30 Doradus in the Large Magellanic Cloud and contains thousands of O stars. Why are the amorphous irregulars both so similar and so different compared to the Im galaxies? One model has been proposed that explains the amorphous irregulars as a result of a gentle interaction between two galaxies. Several billion years after the interaction the gas will have piled up at the center, giving rise to a central concentration of the star formation, while the stars are relatively unaffected by the interaction. In this model amorphous galaxies originally must have been Im galaxies, because the global properties are so similar. A crucial test of this model, mapping the HI distribution in these galaxies, has not been completed. This entry was drawn in part from Hunter, D.A. and Gallagher, J.S. (1989). Science 243 1557 (copyright 1989 by the AAAS). Additional Reading Gallagher, J.S. and Hunter, D.A.(1984). Structure and evolution of irregular galaxies. Ann. Rev. Astron. Ap. 22 37. Hunter, D.A. and Gallagher, J.S.(1986). Stellar populations and star formation in irregular galaxies. Publ. Astron. Soc. Pac. 98 5. Hunter, D.A. and Gallagher, J.S.(1989). Star formation in irregular galaxies. Science 243 1557. Sandage, A.(1961). The Hubble Atlas of Galaxies. The Carnegie Institution of Washington, Washington, DC.