InfluentialPoints.com Biology, images, analysis, design... 

"It has long been an axiom of mine that the little things are infinitely the most important" 

Just a note
Notice that, although a 5 point running mean (n=5) has 4 fewer means than there are observations (loosing two at each end of the graph), a 9 point running mean (n=9) looses 8 points (4 at each end of the graph).
An alternative approach is to calculate 'consecutive' running means  in other words, running means of running means. The last graph of this set shows the result of calculating three consecutive 3point running means  in other words, a 3 point running mean, of a 3 point running mean, of a 3 point running mean. Aside from the quality of smoothing, notice that consecutive running means tend to loose fewer points at each end (in this case a total of 6, compared to 8 for the 9 point running mean). In this case, either the 5point running mean (n=5) or the three consecutive 3point running means provide a reasonable smoothing of the data  without loosing too much information on trends. R calculated and plotted a running mean from a set of data (such as the butterfly data) using these instructions:
