I am interested in the evolution of protein folds. Current protein structures that we have seen only occupy a small proportion of the possible configuration space. If we cluster protein structures into non-redundant folds there are many orders of magnitude fewer unique folds than there are unique sequences. What can we tell about the way evolution affects protein structure by looking at properties of these folds? In particular, if we trace folds' occurrences across genomes in the tree of life we can establish an estimate for the relative age of each structural fold. Are there any domain properties that correlate with age and how can we use them to explore protein evolution?