Hi Courtney, if you are comparing four groups (i.e., Lima, Great Northern, Navy bean, and control), then SPSS won't do anything special with the control group - all four group-names are treated as text-labels, without any processing of the meanings of the names. 

The Tukey test can tell you whether the control group is different from the other groups. The negative sign may not indicate a significant difference. It might be useful to look at the variance or standard deviation for each group. Or you could ask SPSS to make a plot (within the General Linear Model procedure), and to put confidence limits onto the plot. The confidence limits would be proportional to the variance (unless your sample sizes are very different, as Jon suggested).