This next section of the class deals with understanding the universe --- its age, its size, its history, its future. To do this, we first have to understand our local piece of the universe, starting with our own city of stars we call the Milky Way Galaxy.

We have already encountered two constituents of the Milky Way --- stars and the interstellar medium. We have already learned quite a bit about stars: we will now talk a little bit about the interstellar material, which is continually being formed into, and ejected out of stars.

Remember that 9 out of every 10 atoms in the interstellar medium is hydrogen, and most of the remainder is helium. Thus, most of the interstellar medium in the form of gas. We perceive this gas in several ways.

Most of gas in the interstellar medium is atomic hydrogen, with a single electron circling its proton in its lowest energy state. Actually, the ground state of hydrogen consists of two different levels, with a very, very small energy difference between them. Neutral hydrogen atoms in space will occasionally get jostled and the electron will be collided up to its slightly higher energy state. When it decays back down, it will emit light (just like any decay). The energy difference between the two levels is so small, that it corresponds to a photon at radio wavelengths. Specifically, these photons will have a wavelength of 21 centimeters. Most radio telescopes you see around are probably tuned to this frequency. Atomic hydrogen in the interstellar medium is called H I.

Recall that dust is much more efficient at scattering short wavelength light than long wavelength light. (That why the sky is blue and the sun is red at sunset.) Radio light is the longest wavelength light there is, and is not affected by dust at all. Consequently, any H I emission, anywhere in the Galaxy, can be seen with a radio telescope. Of course, because of dust obscuration, we may not be able to see what is doing the emitting, but we can certainly record the direction an H I photon is coming from. From what we see, the galaxy is a very flat place -- all the H I is in a plane across the sky.

In dense regions of the interstellar medium, atoms will collide up against one another and make molecules. All sorts of molecules will can be created. The hydrogen molecule, H_2, is most abundant, but there is also water, alcohol, methane, ammonia, and even some very complex molecules. (Note that complex molecules that have carbon are called organic molecules.) For various technical reasons, the easiest molecule for astronomers to detect is carbon-monoxide, or CO. Like atomic hydrogen, CO has a strong emission line in the radio portion of the electromagnetic spectrum. It can therefore be seen anywhere in the galaxy. These dense regions of interstellar gas are called molecular clouds. When we look at the galaxy at the wavelengths emitted by CO, again we see a very flat place.

Interstellar gas that happens to lie next to a hot star will be ionized by the star's high-energy photons. The subsequent recombination of electrons and their decay into lower energy states produce emission lines, which can be seen in the optical. These regions of ionized hydrogen are called H II regions. H II regions are extremely easy to recognize. On the main sequence, only O and B main sequence stars are hot enough to ionize this hydrogen. But these stars are bright, so the H II regions they produce are very bright. When we look at the galaxy in the light of a hydrogen emission line (produced in H II regions), once again the galactic plane stands out. Finally, the most obvious type of interstellar matter is dust. This sooty material, which is made in the atmospheres of red giant stars, can be seen in silhouette, since it obscures the light of background stars. This obscuration is a problem -- it makes it very difficult for astronomers to study the structure of our Galaxy. Plus, in the optical, we can only see dust that is projected in front of stars. Fortunately, there is another way to view this dust. The dust that is lying far from stars is cold, but it's not absolute zero. It therefore emits via blackbody emission. At the temperature of this dust (perhaps 20 or 30 degrees above absolute zero), dust emits in the far infrared part of the spectrum. When we look at these wavelengths, we see, as above, a flat galaxy.

The last tools we need for understanding the structure and age of our galaxy is the star cluster. First, consider the following three assumptions that are implicit when trying to understand star clusters.

1) We assume that when stars are born, they are always born in groups, and that stars of all masses are made. This appears to be true; although usually there are more small stars made than large stars, the star formation process does seem to produce stars of all masses.

2) When we observe a star cluster, we assume that each star is exactly the same distance from us. Of course, this is not precisely true, but it is equivalent to saying that all the people in Honolulu are the same distance from State College. Given the distances involved, that's a pretty good assumption.

3) We assume that all the stars of a star cluster are born at the same time. Again, this is not precisely true, but given the astronomical timescales of everything else in the universe, this is pretty much true.

Now consider a bunch of main sequence stars on the HR diagram. The O main sequence stars are 50 times more massive than the Sun, but they burn 10,000 times more luminous. The only way this can happen is that they are burning their fuel faster than the Sun does. Conversely, M main sequence stars are 0.1 times the mass of the Sun, but they emit only 1/10,000 as much energy. They are burning their fuel very slowly. In other words, the lifetime of a star on the main sequence depends on its mass: the higher the mass (and more luminous the star), the shorter the lifetime. For instance a star clusters with no O main sequence stars, but with B main sequence stars must be a only few of million years old. Clusters with A main sequence stars, but no O and B main sequence stars must be a couple of billion years old. By observing where the main sequence turns off , i.e., where upper part of the main sequence ends, we can measure the age of a star cluster.

The key to understanding the age, size, and structure of the Milky Way is the Milky Way's star clusters. Generally, they fall into two categories.

The first type of star cluster is the open cluster. Open clusters have hundreds or even thousands of stars, and, as evidenced by their main sequence turnoff magnitude, they are all fairly young. This makes sense. When stars are born, they each some have random motion of their own, and, over time, they will wander away from the stellar nursery. Open clusters cannot be more than a couple of billion years old; since the number of stars in the clusters is relatively small, there is insufficient gravity to hold the cluster together.

The other type of galactic star cluster is the globular cluster. Globular clusters contain hundreds of thousands, or even millions of stars. There are so many stars in a globular cluster that the random motions of the stars within them is insufficient to overcome their mutual gravitational attraction. This type of cluster will not disintegrate over time, but will stay together virtually forever. Observations of the main sequence turnoff of globular clusters shows that these clusters are all extremely old -- as old as 13 billion years.

To understand the structure of our Galaxy (and others), it is best to introduce the concept of stellar populations. Let us follow an idea of Walter Baade in 1955, and define Population I as anything that is associated with young stars, i.e., star formation regions, young star clusters, etc. Similarly, let's define Population II as anything that is old and has been around a long time. So, for example, open clusters, clouds of molecular gas, and bright O stars would all be Population I. Conversely, globular clusters, metal-poor stars, and RR Lyrae stars would all be Population II.

If you find a solitary O main sequence star, it must be young, since such stars do not live very long. But what about a solitary G main sequence star. It might be young, or it might be 10 billion years ago. Is there any way we can tell the two apart?

There is one way. Recall that each generation of stars is made out of the material lost by previous generations of stars. Recall also that, with each generation, some supernovae go off, ejecting newly formed heavy elements into the interstellar medium. Over time, therefore, the relative abundance of all elements heavier than helium has increased. Consequently, newly born stars must have more heavy elements than older stars. (By the way, as a short hand, astronomers use the word metal to describe any element that is not hydrogen or helium.)

If the above scenario is correct, then old stars should have very few metals, and the oldest stars should have zero (or, at least only a trace amount of) metals. And, sure enough, spectra indicate that some stars, like those in globular cluster stars are made up almost entirely of hydrogen and helium. There is almost no metal in them. Conversely, normal stars made later in the history of the galaxy are (by mass) almost 2% metals. It is possible to separate a Pop II main sequence star from a Pop I star (or the same main sequence spectral type) by its metallicity.