Mike

I think your example assumes there is a difference in kernel size. I am saying there is NOT a difference in kernel size...just a difference in density in this case.

When you do a test weight, you don't weigh the sample THEN fill the cup, you fill the cup, level it...then WEIGH the sample.

My example has 21,000,000 kernels of the same size but the higher tw sample weighs more. Again, those 21,000,000 kernels are from 1 acre, irregardless of weight or size...there were 21,000,000 kernels there. Another corn across the road could also have exactly 21,000,000 kernels but that load could weigh differently.