How does a lower center of gravity affect fuel economy? It's been a couple years since I've taken a physics class, and I have been wondering about this basically since I bought my car:
I drive a 2010 Honda Fit, which has a street weight of approximately 2489 pounds (1129 kg).  One feature of the Fit is that the back seats can sit upright like a normal car, or they can completely fold down (this image from this page shows a folded seat next to an upright seat).
Since I rarely have more than one passenger in my car, I can leave the rear seats folded down almost all the time.  Having the seats folded down will result in a lower center of gravity which should increase my fuel efficiency (I think!).
I haven't been able to find an estimate on the weight of the rear seats, other than "pretty substantial" according to this forum post.  But I'm not sure how much of a difference it will make overall, anyway.
The first part of my question is, assuming that raising or lowering the center of gravity affects acceleration and deceleration, how does one quantify that?
Then, given that I may be able to move, say, 100 pounds of seat from the "dead center" of my vehicle to, say, 1-2 feet closer to the wheels, how can I estimate how that will affect my fuel economy?  Perhaps a more specific question would be, how can I estimate how that will affect my acceleration? (which is of course when I get the worst fuel economy, besides being stopped)
 A: The center of gravity should have no effect on economy, besides possibly some indirect effect via changing the load on front/rear wheel bearings or something like that. I would expect secondary effects like that to be insignificant.
Changing the CG won't have an effect on acceleration either, unless you are accelerating at the limit of your tire's grip. A car shifts more of it's weight to the rear tires during acceleration, thus causing the front wheels (in a front wheel drive car) to slip at a lower acceleration than they otherwise would. Lowering the CG of your car will reduce this weight transfer to some degree. So, if you were in a road race it may be a good idea, but for normal driving I doubt that it matters.
NB: The same weight transfer during acceleration occurs in a rear wheel drive car, but because the wheels with more weight on them are now the driven wheels this actually improves acceleration. This is one reason why most sports cars are rear wheel drive -- they can accelerate much harder when tire friction is the limiting factor. ie., when accelerating from a standstill. 
A: One of the sources of losses are the vibrations, i.e. the internal motions of the car, as it responds to the noneven road surface. Changing the CG would have an affect on the vehicles normal modes, and so probably has some effect. If the car had completely round tires, and the roadway was perfectly smooth, they would never be excited. It might not necessarily help however, the relative frequecies and modeshapes would be affected, but the devil is in the datails. In any case, some energy is lost to the excitation of these internal modes, and eventually to their damping via frictional forces. Car manufacturers pay a considerable amount of attention to these things, although they are more interested in the experience of the occupants, noise and harshness, but I doubt they optimize for rolling resitance.
A: If you lower the center-of-mass of your car, it becomes more stable on the road. Consequently, you may afford to buy a new motor and bring your car closer to a formula one. Formulas have inverse wings to keep them stuck on the street at high speeds which is even better than a lowered center-of-mass.
If you succeed to rebuild your car into a formula one, the impact on the fuel economy will be that the fuel consumption rate jumps by one order of magnitude. Instead of 7.5 liters per 100 kilometers that a normal car has, you will switch to 75 liters per 100 kilometers which is the average for formula one races.
But if you don't try to abuse your increased stability by replacing the motor or changing the speed, there will be no measurable effect on the fuel economy. In particular, if the speed of the car is constant, the friction forces etc. that slow the car down are independent of the height of the center of mass, so the fuel consumption is unchanged, too. Only when you accelerate or decelerate, there could be some secondary effects from misbalanced front and rear wheels but I don't expect them to be measurable.
A: Because I always take every corner at the limit, not. People take turns at speeds that are comfortable in terms of lateral g which is well below any slip angles of a normal vehicle. No scrubbing of tires or extra loss of momentum compared to a lower cg for normal drivers. 
A lower CG will allow for increased lateral loads but that is a performance as opposed to increased gas mileage thing. Even lowering a car doesn't change it's frontal area so no drag gains there either beyond the minimal reduction in exposed tire area. If the underside is designed for downforce, increasing it effect would actually hurt gas mileage. When it comes to the underside n a normal vehicle, it's more about turbulence and better (less) can vary either up or down. Lowering the CG or dropping a vehicle's height is all about dynamics and has no consistent effect on fuel efficiency.
A: Lowering car's CG would affect a lot in fuel consumption. Especially if you frequently enter / exit sharp corner. Cars with low CG will be able to maintain its current speed compare to cars with high CG. Therefore cars with high CG will use brake and accelerate more often than cars with low CG. 
