I agree with Alex’s statement that the original graph was to show the strong correlation, but adding another line showing the same actual data points on the same graph isn’t the solution. It would be really distracting because people would assume that a third line meant a third set of data and not just the same data plotted differently. I prefer the revised plot shown above. The correlation is still discernable shown like this.

]]>So why not add a third line? Something like this:

Show the actual NPP and CO2 numbers with differently colored solid lines, as shown in the 2nd version. Add a third dotted line in the same color as the CO2, labeled CO2 (Inverted to demonstrate absolute correlation).

If you show the actual numbers clearly, then equally clearly distinguish the inversion, you can make both points without misleading, or allowing your graph to be misused.

]]>The **real** reason I left the circles and squares as-is is that I was editing a postscript file by hand, not starting from scratch with raw data, and I didn’t even think to make the change.