It is pseudo-replicated. But it's okay; you can still get meaningful data. True replicates will need to be entire blocks located in sufficiently distant areas :)

Factor A is the main (whole) plot and Factor B (time) is the sub plot and replications are blocks. So the code using the MIXED procedure is as below. You can also use the GLM but it is not as accurate and the MIXED. The code for MIXED is also simpler and more straight forward as it does not need error term construction .

proc mixed; class A B Block; model Y = A B A*B; random Block A*Block; run;

The GLM code is like this:

proc glm ; class A B Block; model Y = A B A*B Block A*Block; test h=A e=A*Block; (Note that the "test" statement for introducing the roor terms of the main plot and the Block) run;

Good luck

It is pseudo-replicated. But it's okay; you can still get meaningful data. True replicates will need to be entire blocks located in sufficiently distant areas :)

Thanks all