Am using the maanova package 1.14.0 Model includes two breeds (fixed breed factor) and 5 time points (random time factor) The three commands are > BDeset_gcrma_NoCon25LRmadata.raw.WithAffyID081109 <- read.madata("ABeset_gcrmaNoCon_25PsLRs081107.txt", designfile="DesignV4LR081109.txt", header=TRUE, spotflag=FALSE, intensity=2, probeid=1) > BGanova.Br_T.mix081109 <- fitmaanova(BDeset_gcrma_NoCon25LRmadata.raw.WithAffyID081109, formula=~Breed+Time, random= ~Time, covariate = ~1, verbose=TRUE) > ftest.Breed.mix_1000perm081109 <- matest (BDeset_gcrma_NoCon25LRmadata.raw.WithAffyID081109, BGanova.Br_T.mix081109, term = "Breed", n.perm = 1000, critical = .9, test.type = c("ftest"), shuffle.method = c("sample"), MME.method = c("REML"), test.method = c(1,1), pval.pool = TRUE, verbose = TRUE) With two breeds and five time points there are 9 possible contrasts. Does the Ftest in maanova study all 9 breed contrasts or just the 5 breed contrasts at the specific time points? Thank you -- Loren Engrav, MD Univ Washington Seattle