The formation and evolution of galaxies is rather poorly understood. Two very different theories are proposed as an attempt to explain observed properties of galaxies.

In the monolithic collapse scenario galaxies form in a single burst of star formation, and thereafter evolve only quiescently, as their stars age and slowly fade. This simple picture predicts that galaxies in cluster should all have very similar properties, as they are all born at the same time under similar conditions. These predictions find a great deal of support in observations of the colours of cluster galaxies, their luminosities and their mass-to-light ratios.

On the other hand, the hierarchical model of structure formation predicts that galaxies should be assembled over their liftime though mergers between smaller galaxies. The decrease in the number of massive clusters of galaxies with redshift, and the decrease in the number of luminous early-type galaxies with redshift support this prediction.

The epoch of Galaxy Assembly.

My work has concentrated on determining the luminosity functions of galaxies. Using the K band luminosities of galaxies to measure their mass, and determining a typical mass of a galaxy in a cluster by fitting the luminosity function of galaxies, it is possible to explore if any significant merging has occurred in the lifetimes of the galaxies with the cluster. The properties of the galaxy populations of three high redshift clusters are consistent with purely passive evolution, and require no merging to be postulated. Note however, that in very massive systems such as the ones studied, we are examining an extreme class of object. Such massive clusters are amongst the most overdense regions of the universe, and thus the object within their cores are likely to be very old within a hierarchical model, i.e. all their merging would be complete at relatively high redshifts.

The evolution of K* as a function of redshift, compared to passive evolution models. The data are consistent with purely passive evolution with a redshift of formation of z>2. The circles are from my thesis. The squares are taken from De Propris et al. 1999.

The Epoch of Star Formation.

It is important to distinuish between the epoch at which galaxies are assembled, through processes such as merging, and the epoch of star formation. In the monolithic collapse scenario, assembly and star formation occur simultaneously. In the merging model though, it is possible that bursts of star-formation may be induced by mergers. Such a process would severely disrupt the homogeneous qualities often observed in a cluster of galaxies. If, however, merging is a dissipationless process, then very little star formation would occur as a result of a merging event. This would mean that the stellar populations of galaxies may continue to appear old, even though the galaxy may have assembled relatively recently.

The colour-magnitude relation offers a useful insight into the star-formation history of cluster galaxies. The colour-magnitude relation has been shown to be primarily a relation between metallicity and mass. More massive (brighter) galaxies contain more metals and hence are redder. The overall colour of the relation is indicative of the ages of the galaxies, and the scatter reveals the presence of any merging.

The colour-magnitude relation of ClJ1226.9+3332. A massive cluster of galaxies at z=0.89 has been studied using VRIzJK data obtained at the UKIRT and SUBARU telescopes, and F606W and F814W data from the Hubble Space Telescope Advanced Camera for Surveys. Spectroscopy has been used to determine cluster membership for the brightest galaxies.

Three colour image of Cl J1226 from Subaru SuprimeCam VRI observations. Red circles denotte cluster members as confirmed by Keck LRIS/ Gemini GMOS spectroscopy. Green circles denote non-members. See an image of the cluster core.

The fraction of early-type galaxies in ClJ122.9+3332 is low, in contrast with nearby clusters. The following plot shows the evolution of early-type fraction with redshift, taken from Van Dokkum (2001) with the early-type fraction of ClJ1226.9+3332.

The early-type fraction of clusters as a function of redshift. The data are a compilation from Van Dokkum (2001) with the early-type fraction of ClJ1226.9+3332 shown as the large filled circle.

This implies that there has been morphological evolution of galaxies since z=0.9, perhaps as a result of infalling late-type galaxies.

In contrast, when early-type galaxies alone are considered, very little evolution is observed since z=0.9. The following plot of the R-K colour-magnitude diagram shows very small scatter implying the early-type galaxies are a homogeneous population, and are already very old at z=0.9.

The colour-magnitude relation of ClJ1226.9+3332. Red points are confirmed early-type members, blue points are confirmed late-type members and green points are confirmed members of unusual morphology.

The K band luminosity function and the colour-magnitude relation show very little evolution of galaxies since z=1. On the otherhand the early-type fractions of clusters show evidence for morphological evolution of galaxies. If late-type galaxies are being transformed into early-type galaxies, e.g. through infall, then they must evolve to be remarkably similar to the existing early-type galaxies in the cluster. Furthermore the morphological evolution must take place after the cessation of star-formation which is responsible for the change in colour of the galaxies, to preserve the tight colour-magnitude relation from z=1 to z=0. Evidence for this is seen in ClJ1226.9+3332; when the morphologies are examined from the HST data it can be seen that there are several late-type galaxies on the CMR, and we postulate that these are galaxies in the process of transformation.

The morphological composition of the colour-magnitude relation of ClJ1226.9+3332. Red points are confirmed early-type members, blue points are confirmed late-type members and green points are confirmed members of unusual morphology.