The central limit theorem for random fields

A new variant of CLT is established for random elds de ned on Rd which are strictly stationary, with nite second moment and weakly dependent (comprising cases of positive or negative association). The summation domains grow in the van Hove sense. Simultaneously the indices of observations form more and more dense grids in these domains. Thus the e ect of combining two scaling procedures is studied. A statistical version of this CLT is also proved. Some stochastic models in Radiobiology based on dependent functional subunits are discussed as well.