During Steffen’s time at Columbia in February this year, we were talking a lot about erosion of globular clusters by the gravitational field of their host galaxy. Steffen came up with the great idea that this erosion process may be responsible for the observed variations of the number of globular clusters around different kinds of galaxies. In general, you see that galaxies with a small total mass will have a small number of globular clusters. But if you divide the number of a galaxy’s globular clusters by the mass of the galaxy, you will not get a flat distribution as naively expected. Instead, you get a U-shaped distribution, telling you that very low-mass galaxies and very high-mass galaxies have more globular clusters than average, whereas intermediate-mass galaxies have suspiciously few clusters. Why should those galaxies be any special??
Our idea was that this variation is due to the concentration of the galaxies: intermediate-mass galaxies tend to be more highly concentrated, meaning that the gravitational tidal field is stronger inside the galaxy than for a more extended galaxy. Since this tidal field is the main driver of cluster erosion, you expect these galaxies to have eaten more of their globular clusters by now. We used a catalog of galaxies for which the mass is accurately known, and for which the total number of globular clusters had been counted. The ratio of the two (S_N) is shown for all these galaxies by the black dots in the above figure. M_V on the x-axis corresponds more or less to the mass of the galaxy, where low-mass galaxies with 300 million times the mass of the sun are on the right and high-mass galaxies with mass of up to 3,000 billion solar masses are on the left. This distribution has a similar U-shape as the expected number of surviving clusters for these galaxies (blue and green points), hinting at a possible connection between these two quantities. The paper was written and published in a few weeks. Definitely one of the fastest publications I’ve ever had!