Collisions of deformed nuclei such as U may provide 40% more hard processes
and about 30% larger energy densities, compared to central PbPb collisions
. They also produce excited systems which are strongly deformed in the tran
sverse plane, which are much larger than possible in peripheral PbPb collis
ions. We discuss how, even without a polarized target, one can study these
phenomena by selecting particular events. Collisions are studied by a simpl
e Monte Carlo model, and it is shown what can be achieved by making cuts in
two control parameters-the number of participants and ellipticity. We also
discuss how UU collisions may resolve a number of outstanding issues, from
corrections to hard processes to elliptic flow (the existence of a quark-g
luon plasma), to the mechanism of J/psi suppression.