Hi, In lmFit(object, correlation), "correlation" is "the inter-duplicate _or_ inter-technical replicate correlation". Correlation is calculated using duplicateCorrelation(object, ndups, block), where "ndups" is the number of spot replicates, and "block" is used to describe technical replicates. The help entry for duplicateCorrelation() says that "At this time it is not possible to estimate correlations between duplicate spots and between technical replicates simultaneously. If block is not null, then the function will set ndups=1." So what would be the best way to deal with both duplicate spots _and_ technical replicates of arrays? Calculate both, and use the most conservative (which is what duplicateCorrelation() implicitly does by assuming that between-array correlations are lower than within-array correlations I guess)? Calculate both, and use some kind of weighed average? Does someone have experience with this issue? thanks, Dries Knapen ******************************************** Dr. Dries Knapen University of Antwerp Department of Biology Ecophysiology, Biochemistry and Toxicology Groenenborgerlaan 171 - U711, B-2020 Antwerp Belgium tel ++32 3 265 33 49 fax ++32 3 265 34 97