Suppose I were setting prices at airline U. Clearly what I want to do is maximize net revenues. Now suppose a person wants to fly from Portland OR (PDX) to Washington Dulles (IAD). Airline U has a non-stop flight and several connecting flights to Dulles. On a non-stop flight, a person does not change planes, has much less chance of the luggage not getting there with him, and has a shorter flight time than on making a connection. That seems like a benefit to the passenger It sure is to me. I know there is something lively about O'Hare that I miss, something Carl Sandburg would have understood, but I can forego that, knowing that you can't avoid O'Hare forever. Is the passenger willing to pay extra for this benefit? Well, let's charge a premium for the non-stop flight, but make sure we can still fill the plane. If the passengers are willing to pay a premium for being on a non-stop flight from Portland to Dulles, then airline U, my imaginary employer, makes two gains, one from having fewer people go through a hub and the second from the premium charged to avoid a hub. In real life, I am willing to pay a premium, say $50, to fly non-stop from Portland to Dulles. After all, some people will pay a premium to fly into Washington National (DCA) instead of Dulles. Why does airline U insist on having so little point to point service and so much traffic through its hubs? Hey, I only set fares. I don't run the airline. Notice that if airline U had several non-stops from Portland to Dulles, and less service through hubs, then it would be the more expensive (to the airline) trip through O'Hare that would cost more in my scheme of things. There's been a lot written about this and some I've just skimmed, so if I have repeated someone's ideas, I apologize to the list; if I repeated your ideas, then remember, great minds think alike. :-). john Fan of Non-Stops John Kurtzke, C.S.C. Dept of Math Univ of Portland Portland OR 97203 503-943-7377