We review some possible improvements of mean-field theory for application t
o nuclear-binding systematics. Up to now, microscopic theory has been less
successful than models starting from the liquid drop in accurately describi
ng the global binding systematics. We believe that there are good prospects
for developing a better global theory, using modern forms of energy-densit
y functionals and treating correlation energies systematically by the RPA.
(C) 2001 MAIK "Nauka/Interperiodica".