[rkhampton]

I have to respectively disagree with Paladin2. The factors which he lists in his explanation are referred to collectively as "rolling resistance." Rolling resistance increases somewhat linearly with speed; whereas, air resistance increases exponentially. At low speeds, rolling resistance is the major factor, but for any car, there is going to be a point at which air resistance takes over and does so very quickly, due to its exponential nature. For the average car, this seems to be around 55 mph. Now, the only question is, "Does a Chrysler Concorde vary that much from the average car out there?" I would tend to say no, but I don't own one.

[Paladin2]

There could also be some other variables. As the Federal emissions and fuel economy tests are performed at speeds below 65, it is possible to run slightly lean at higher speeds and gain a few percentage points of engine efficiency. Note that it is difficult to fill the gas tank to the same level every fillup. A half gallon fill difference (1/4 gallon both ways) can make a big difference in the calculated fuel economy. Running tests can result in lots of variability unless the measurement system is quite capable and lots of variables are well controlled. That's one reason why they say "your mileage may vary".