I always like simple amusing examples that illustrate an important statistical point. An email by JW on EDSTAT-L offer a couple of examples.
The first is the statistician who reviewed fire department records and concluded the more fire engines you send to the scene of a fire, the more damage they cause. This is an example illustrating that it is not necessarily A causing B, but rather B causing A. I have used this example in my classes often.
A second example is good, and it is one that I had not heard before, though it is apparently used quite commonly. See for example, the Wikipedia entry on correlation and causation:
This example shows a strong correlation between ice cream sales and crime rates. The “obvious” interpretation is that ice cream causes crime. A more logical interpretation is that both ice cream sales and crime rate are correlated with outdoor temperature.
You can find an earlier version of this page on my old website.