[R-sig-ME] Split plot (with repeated measure?)

[Mike Palmer]

Hello,

I would like advice on if and how I might viably analyse this dataset: 

The experiment has 3 blocks with a three-level main plot treatment, split to a two-level treatment. In this case most parameters are also measured repeatedly at 1-3 monthly intervals. Ideally I would like to test the full response~treatment_1*treatment_2*sample_time model, but don't know what the structure should be, or even whether this is a viable analysis. I should probably add that neither the main or sub plots are randomised.

Thanks

Mike Palmer