Tabster, we have completely reviewed your past year and in order to be absolutely just we have put it into the context of the previous ten years. Regression analysis, histograms, scatter analysis, relative standard of deviation, co-efficient of variation, and other tools were used to build matrices of analyte content which were as free of bias as possible (a confidence level of plus-or-minus 1%).

We can say with confidence that our methods produced the most complete and accurate such set of matrices ever achieved. We will, in fact, be submitting our methodology to The Journal of Statistical Behaviour Predicative Studies.

There can be no doubt about the conclusions reached. Absolutely none.