I've observed the same behavior that Maggie and others have reported,
and have harbored what I'm sure will be an unpopular theory to explain
it: rolling hills can be good for gas mileage.  Although it may first be
counterintuitive, I think it's quite possible that at some speeds, and
for some grades of hills (depending on the engine and gear ratios) going
up and then down hills can consume less gas than driving on a flat
portion of equal length.

Before someone says that this is obviously impossible, consider this:
If my goal was to maximize gas mileage over a 100 mile stretch (i.e.,
minimize gas usage to cover the 100 miles), I would not be at all
surprised if my best bet was to climb a very steep hill for the first
10-20 miles, then shut off the engine (or let it idle) and coast down a
shallower 80 mile hill, leaving me at my original altitude, probably at
well over the normal 20 mpg.

It seems like something similar could be going on when we drive over
more normal hills: the van sometimes gets better gas mileage by
concentrating its work on the climb, then taking a break on the coast.

-Wes
'89 1.9l.  missing the hills.

Maggie Dew wrote:
> Well, it's not just that the "grass is greener" - I have recorded every tank
> of gas and every mile i've driven since I got the bus.  I spent all summer
> above 7,000 ft. for 2 years, and all winter in the Oklahoma flatlands for 2
> years, and the mpg is consistently higher in the mountains, although she
> seems to be at lower power climbing the hills
>
> "From:    Paul Connelly <vanagonhummingbird@GMAIL.COM>
> Subject: Re: MPG
>
> I think its just a "grass is greener on the other side thing"! I live at
> 9,100" and most of my driving is done at over that, and there is always lot=
> s
> of climbing, so when I get down to the plains, even over 5,000', my van
> seems to have wings with all that "extra power".
>
>