GALAXIES, MOLECULAR GAS IN JUDITH S. YOUNG Much of the beauty in the Milky Way and other spiral galaxies depends on star formation, an ongoing process that began when galaxies began to form more than 10 billion years ago. Although most of the stars in the night sky are billions of years old, there are a small number of young, high-mass stars that formed recently (i.e., within the past 10 million years) from clouds of interstellar gas and dust. These young stars heat the gas clouds from which they formed, and produce an abundance of glowing nebulae that are commonly found in spiral galaxies. These stars finally explode as supernovae, the best recyclers of all, and return the processed elements from their interiors back into the galaxy to be incorporated into future generations of stars. Recent observations of star-forming regions made with radio telescopes have revealed that the birth sites for future generations of stars are giant clouds of molecular gas, typically more than 100 light years across and containing a mass of gas more than one million times the mass of the sun. These clouds of gas, which indicate the potential for future star formation in a galaxy, can be studied in detail in our own Milky Way galaxy. In other galaxies we can determine the large-scale distribution of molecular gas in relation to past and present star formation. Thus, studies of molecular gas in galaxies will help provide an understanding of the large scale processes that influence the evolution of galaxies. Because stars form in molecular clouds, most studies of molecular gas in galaxies have been confined to studies of galaxies in which young stars are present, that is, primarily spiral and irregular galaxies. These studies have been made with radio telescopes operating at millimeter wavelengths. The radiation from the molecules in the giant gas clouds arises from changes in the rotation of molecules as a whole. The most abundant molecule in the giant molecular clouds is molecular hydrogen (H*), with numerous trace molecules, the most important of which is carbon monoxide (CO). Determinations of the molecular gas content and distribution in galaxies are based on observations of the abundance of the CO molecule. It is only within the past 15 years that sensitive radio receivers have been built that can detect the emission from molecules at millimeter wavelengths. Prior to the discovery of molecules in the interstellar medium of our galaxy, our knowledge of the gas content of galaxies was based on observations of atomic hydrogen. Studies of these atomic gas clouds in galaxies were very important, but led to an incomplete view of the gas content and evolution of galaxies. MOLECULAR GAS DISTRIBUTIONS IN GALAXIES Ever since pioneering studies of the atomic gas content of galaxies in the 1950s and 1960s, it has been known that most spiral galaxies have similar distributions of atomic gas. Thus, with the discovery of the molecular component of the interstellar medium, it was initially expected that molecular gas distributions among spiral galaxies might also be similar. However, this expectation turned out to be incorrect: There are large differences from galaxy to galaxy in the distribution and total mass of molecular gas, and elucidating these differences has led to a new understanding of star formation in galaxies. The distribution of molecular gas in the Milky Way galaxy has been studied in detail since 1975. Molecular clouds were found to be extremely abundant in the central 500-1000 ly out from the galactic center, but their number fell off at larger radii. It was indeed surprising when a ring of molecular clouds was discovered in our galaxy, with a peak at about 15,000 ly out from the galactic center. Following the study of the molecular gas distribution in the Milky Way, a number of questions arose that could best be addressed by observations of other galaxies. For example, is the ring of molecular clouds in the Milky Way a common feature of other galaxies? Does the abundance and distribution of molecular gas in a galaxy depend on the form of a galaxy? Does the total luminosity of a galaxy depend on the quantity of molecular gas in a galaxy? Why are spiral arms apparent in spiral galaxies? Specifically, do more stars form in the arms because there is more gas, or do stars form more efficiently in spiral arm locations in galaxies? Studies of molecular clouds in the disks of luminous, nearby face-on spiral galaxies have revealed that the molecular gas distributions are unlike the distributions of neutral atomic hydrogen gas. That is, the molecular gas is concentrated in the centers of most galaxies, with distributions that decrease with increasing radii from the galactic center. Thus, most of the molecular gas in spiral galaxies is confined to the inner half of the optical disk of the galaxy. In contrast, the distributions of atomic hydrogen gas are relatively flat as a function of radius in spiral galaxies, and these gas distributions extend well beyond the optical edges of galaxies. In the most luminous spiral galaxies, there is more molecular than atomic gas in the inner parts of the galaxy, and more atomic than molecular gas in the outer parts. Because atomic gas is probably the reservoir of material out of which molecular clouds form, these results indicate that the formation of molecular clouds is most efficient in the inner disks of spiral galaxies. Within spiral galaxies where the amount of molecular gas exceeds that of atomic gas, it has been discovered that there is a one-to-one proportionality between the distribution of molecular gas and the distribution of luminosity in young stars. These observations have been interpreted to indicate that the rate of star formation, deduced from the luminosity in young stars, depends simply on the amount of molecular gas present. This implies that more star formation occurs when more gas is present, and that it does not matter whether the cloud is located in the interior of a galaxy or in the outer parts. Recent observations of face-on spiral galaxies have revealed that the spiral arms have different properties than the disk. Specifically, molecular clouds are found everywhere in the inner disks of spiral galaxies, both on the arms and in the interarm regions, whereas young stars are found primarily on the spiral arms. This indicates that the rate of star formation per unit molecular gas mass, or the star formation efficiency, is higher on the spiral arms than in the interarm regions of a galaxy. Thus, the formation of high-mass stars is enhanced in the arm regions of a spiral galaxy, leading to the spiral appearance of the galaxy. EFFICIENCY OF STAR FORMATION IN GALAXIES Through a survey by the Infrared Astronomical Satellite (IRAS), the infrared emission has now been measured from galaxies over the entire sky. This emission is from dust that is heated by young, high-mass stars embedded in molecular clouds. The luminosity of a galaxy in the infrared, therefore, provides a measure of the rate of formation of high-mass stars. Figure 1 illustrates images of six galaxies with a wide range of absolute size and total star formation rate. The nearby, well-known, actively star-forming galaxies M82 and NGC 253 are quite small on an absolute scale. One of the contributions of the IRAS survey is that many distant galaxies with high rates of star formation were identified and subsequently found to have suffered from collisions with other galaxies. This result led to studies of the rate and efficiency of star formation in galaxies in different environments. Specifically, when the most isolated galaxies are compared with strongly interacting galaxies, it is found that both samples exhibit similar ranges of molecular gas mass, whereas the interacting galaxies display more stellar luminosity for a given supply of gas. Thus, the interacting galaxies have -5 times higher ratios of young stellar luminosity to molecular gas mass than the isolated galaxies. This indicates that the environment of a galaxy has a strong influence on the efficiency of formation of high-mass stars. Studies have also been conducted comparing the rates of high-mass star formation with the molecular gas contents in different types of spiral and irregular galaxies. Spiral galaxies exhibit a variety of forms, or morphologies, including the flat disk of the galaxy (with the spiral arms) and the spheroidal bulge centered on the galaxy. Galaxies of type Sa exhibit large bulges and tightly wound arms; galaxies of type Sc exhibit small bulges and loosely wound arms; galaxies of type Sb are intermediate in terms of bulge size and spiral arms. The bulges of spiral galaxies are generally composed of old stars, which have red colors, whereas the spiral arms have blue colors indicative of the presence of young stars. Because of the large bulges in Sa galaxies, these galaxies have red colors, and because of the small bulges and prominent spiral arms in Sc galaxies, these galaxies have blue colors. This color difference in galaxies as a function of type has led to the perception that there is less star formation in Sa galaxies than in Sc galaxies, when in fact the color difference can be attributed largely to the presence of the bulge and not to the absence of star formation. If there is star formation occurring in Sa galaxies, then these galaxies should contain sufficient quantities of the prerequisite molecular gas out of which stars form. This has been found to be the case: Sa galaxies have the same range of molecular gas masses as Sc galaxies. In fact, Sa galaxies also have the same range of star formation rates as Sc galaxies. Thus, when investigating the rate of star formation per unit molecular gas mass in the disks of Sa versus Sc galaxies, it appears that the two types of galaxies are indistinguishable. That is, Sa and Sc galaxies have similar mean values of the yield of young stars per unit mass of molecular gas. This indicates that the process of star formation is generally not affected by global properties, such as the galaxy morphology, but is a local process primarily sensitive to the amount of molecular gas present. The result that star formation is a local process and depends primarily on the amount of molecular gas present was previously discussed for the yield of young stars per unit mass of molecular gas within the disks of individual galaxies. The exceptions to this rule are found in spiral arms and in interacting galaxies, where the yield in young stars per unit mass of molecular gas is enhanced. One explanation for the difference in the efficiency of formation of high-mass stars in interacting versus isolated galaxies and in spiral arms versus the disks of spiral galaxies, is that the formation of high-mass stars is enhanced when more molecular gas is accumulated in a small region, be it on a spiral arm or in the center of an interacting galaxy. The formation of these high-mass stars could result directly from collisions between giant molecular clouds or from the overcoming of the magnetic or turbulent forces that generally support the clouds against collapse. RATIO OF MOLECULAR TO ATOMIC GAS MASS IN GALAXIES Ever since the discovery of atomic gas in the universe and studies of the gas content of spiral galaxies, it has been recognized that the mass of atomic gas per unit stellar luminosity in Sa galaxies was about 5 times lower than the atomic gas mass to luminosity ratio for Sc galaxies. This led to the perception that Sa galaxies had less gas than Sc galaxies, when in fact they simply have larger stellar luminosities due to the presence of the central bulge. In order to determine whether or not Sa galaxies truly have less gas than Sc galaxies, it is necessary to measure the total gas content of spiral galaxies, that is, molecular plus atomic gas. For a sample of -200 galaxies, it has recently been found that although the atomic gas content per unit luminosity increases from Sa through Sc galaxies, the molecular gas content of these same galaxies per unit luminosity remains constant. Thus, there is a change in the ratio of molecular to atomic gas in spiral galaxies as a function of morphological type: Sa galaxies have -4 times more molecular than atomic gas, whereas Sc and Sd galaxies have -5 times less molecular than atomic gas as shown in Figure 2. Overall, the total gas content of galaxies is fairly constant, with Sa galaxies having only a factor of 2 less total gas mass than Sc galaxies. The principal result as a function of morphological type is that the phase of the gas changes with galaxy form. Because the morphology of a galaxy is a reflection of the mass distribution, and if atomic gas is the reservoir out of which molecular clouds form, this result indicates that the conversion of atomic to molecular gas is most efficient (i.e., leads to the highest molecular to atomic gas ratios) in galaxies with centrally concentrated mass distributions, where more gas is likely to be collected in the gravitational potential. SUMMARY Molecular gas is found to be an important constituent of galaxies because it is the reservoir of gas out of which stars form. In the disks of spiral galaxies and among isolated spiral galaxies, stars form in proportion to the amount of molecular gas present, independent of the morphology of the galaxy. Within these galaxies, the property that does depend on galaxy form is the ratio of molecular to atomic gas. Because the formation of a molecular cloud from an atomic gas cloud is a large scale process, it is reasonable that gravity on a large scale is important. This is consistent with the observation that the molecular to atomic gas ratio in spiral galaxies changes with galaxy morphology, because morphology is a reflection of the large scale mass distribution. Star formation, on the other hand, is a small scale process relative to the size of a giant molecular cloud, and in the disks of isolated galaxies star formation appears to be a local process that is insensitive to the large scale gravitational field. Only in interacting galaxies or in spiral arms does the enhanced conversion of gas into stars appear to operate. Additional Reading Hubble, E.(1936). The Realm of the Nebulae. Dover, New York. Schweizer, F.(1986). Colliding and merging galaxies. Science 231 227. Scoville, N.(1989). Molecular gas in spiral galaxies. In Evolutionary Phenomena in Galaxies, J.E. Beckman and B.E.J. Pagel, eds. Cambridge University Press, Cambridge, p. 63. Scoville, N. and Young, J.S.(1984). Molecular clouds, star formation and galactic structure. Scientific American 250 (No. 4) 42.