fitting smaple within plot data

We're looking at some leaf growth data with multiple measurements (nominally 10 PLANTS) per PLOT. The plots have a spatial structure (ROW x COL) in the field, but the plants are random within plots and so do not relate to each other in any structure. I'm using ASREML-R.

While I have simply calculated sample means per plot and then done used a standard ar1 x ar1 R structure, I'm wondering if there is a way to do this using the original data, and making use of any sample variance information (or at least saving me the trouble of calculating the means)?

Consider the data to have the following FACTORS (num levels) with GENOTYPEs as random and leaflength as the variate....

PLOT (400) GENOTYPE (200) REPLICATE (2) ROW (20) COL (20) PLANT (10) leaflength

thanks,
scott c.
Received on Thu Oct 14 2007 - 20:28:54 EST