Dear all; I have the following SAS code: proc mixed data=mydata covtest asycov update; class rep cross; model ht = rep; random col1-col5 / type=toep(1); random cross/type=vc; ods output covparms=_varcomp asycov=_cov; run; where col1-col5 are columns of 0 and 1 (Z matrix of dummy variables) what I'm looking for is some hint of how to replicate this analysis using lme in R, especifically how to translate the variance structures used for the random effects: the commands type=toep(1) (banded Toeplitz in SAS terminology) and type=vc (variance components in SAS terminology). please let me know if you have any suggestion thanks in advance PM