I have to stick to my guns on this one. Admittedly there's far more brain cells floating around here than I have, but I'm invoking a wise old concept that "if it looks right..." The AV-36/Pelican configuration looks right, and it is known to fly right. An aluminum Pelican was, and is, on my short list of airplanes to create. So all of the ideas and concepts in this discussion have already been banging around and weighed against each other in my aching head. Using a tractor configuration will just make the airplane larger and more complex than it needs to be in this case. There are plenty of times where the pusher configuration creates more problems than the tractor, but for this exercise I think not: 1) A tractor engine, mounted low enough to not ruin half of your visibility, will require a longer landing gear. 2) A full VW will use a little larger propeller , raising the height of the landing gear even more. 3) A pusher propeller can be mounted slightly higher, in this case pilot's eye level, because the forward visibility is not affected. The thrust line has to be angled a few degrees (through the CG), but this is a very very small loss of thrust at these angles. So this makes it easier to use a mono-wheel and simple outriggers or "training wheels" under the Fauvel fins to allow takeoff without a sailplane "wing runner". 4) A pusher engine can have the exhaust and other components mounted in the wing or in the fairing above the trailing edge. 5) A tractor engine will move the pilot rearward, in order to balance the aircraft, and also because the engine is taking up the space where the pilot's feet would have been. Even if for balance alone, this makes the aircraft larger (longer) in order to balance itself out. 6) Moving the pilot rearward results in either the spar not being straight anymore, or the pilot sitting on top of the spar. If you crank the spar and sweep the wing forward (as with many flying wings) then the whole mess becomes more complex to engineer and build. If the pilot sits on top, this increases the aircraft's fuselage volume (drag, intersection drag, etc.). It also results in a low wing airplane, which results in an even longer, heavier, draggier landing gear. 7) The Pelican configuration allows the pilot seat to be hung right from the main spar. No long load paths, no cantilevered beam to mount the seat on, etc. As for the discussion of the U-2, it is also a neat airplane. But it is not fair to constrain the Pelican to a one piece wing and the U-2 to a three piece wing. Either airplane could be built either way. You could build a Pelican with tip panels the same as a U-2. You could also build a Pelican where the center section span was just under the maximum trailer width... 7 feet or whatever. You could leave the center section at 8 feet and just bank the airplane 30 or 40 degrees on the roll axis, keeping it within the trailer width limits. There was a famous photo of a Formula One Cassutt racer that went to the race on a trailer like that. (Raceair, do you remember this? I saw the photo years ago in one of the old race annual books.) In my version of this, the original Pelican design would be skewed just a little to address the big difference between Debreyer's original mission (sustained flight on minimum power) and my mission (easy to build, fuel efficient, inexpensive, but still able to operate like a "normal" airplane in real-world conditions). This is easy enough, because Debreyer started out with a very thick wing section to save weight and make lift. For my mission, the airplane doesn't need to be as light, and it doesn't need as much lift. So I can move away from the 17% thick section and go back to a slightly thinner one. Maybe 13-14%, I would have to ask one of our HBA high brain function types for the right number. It is also possible that one of Dr. Hepperle's airfoil sections would offer some advantages over the original Fauvel/Debreyer sections. Again, that is something I would have to research (translate: ask someone more educated than me). (NO, there is no need for a truly high performance airfoil on this aircraft, no Wortmann helicopter blade airfoil, it will not be built in a way that justifies that. Pop rivets, oil-canning wing skins, and probably overlap seams.) Using a slightly thinner airfoil will yield some modest improvement in cruise speed, but the flight characteristics and pitch behavior are far more important. If this is a 70 MPH airplane on 35HP, that is plenty good enough. Anyway, rant switch off, those are the reasons that I would favor the Pelican layout over the others for this particular mission.