A scientist grows calcite crystals at three different temperatures and in three different concentrations of solution. The crystals grow in a pyramid shape on the flat substrate. The base of the pyramids is a parallelogram. Two adjacent sides of the pyramid incorporate magnesium (mg) at a higher rate than the other two sides.

The scientist measures the mg content of a 60 micron (10-6 meters) square sample at 3 micron intervals. The measurement tool gives the percent of mg and a standard deviation for each measurement. The tool is being pushed to the limit of it’s capabilities, so the measurements with high standard deviations are considered to be bad.

The scientist would like to throw out the high standard deviation measurements at some threshhold, divide the remaining measurements of a sample into the high mg and low mg sides of the pyramids, and get an average of the high mg measurements and another average of the low mg measurements. When this has been done for each sample, then the resulting high and low mg averages of all the samples should be analyzed to show interesting features of growth in the three teemperatures and three solution strengths.

In the Excel file, the X and Y columns are the coordinates that the measurement tool uses in collecting the data from the samples. Since each sample is mounted in the tool in a slightly different position, the coordinates do not match from sample to sample. The mg column is parts per million of mg in the sample at those coordinates. There are also columns for the standard deviation of the measurement, the solution concentration, the temperature, and the identifying number of the sample.