Welcome aboard LDR! And thanks for the interesting article. I knew Mark when I lived in Oregon - good scientist, and I suspect that he is right, for the following reasons:

Basically what happens when you put fish in a hatchery environment is that all the fish with "wild" characteristics, ie: shy of disturbance, terratorial, aggressive towards each other, bothered by overcrowding, etc., wind up dying out, while all those with domestic characteristics, those that follow the fish feeder around and don't mind being crowded and not having any place to hide, survive. Of course, when released, 90+% of them die instantly, either they starve or they get eaten the first time they run up to a heron on the shore to be fed! But those few that do survive pass these domestic characteristics on to their offspring, which, if they are unlucky enough to be born in the wild, are not real good characteristics for survival. Wild fish, on the other hand, since 90+% of them die in the wild before reaching smolt size, are custom made for survival in the wild, being survivors and the best of the best themselves. They will pass this on if given the chance.

I worked with fish in a severely manipulated environment for years (ever hear of Oreaqua?). We could completely change the characteristics of fish populations in one generation by selecting for the characteristis we wanted, like early return and ocean bright condition at return. Salmonids are severely mutable. This is why you can destroy all the wild characteristics in one generation by raising them in a hatchery, and why Mark sees these results.

There are a couple of ways to deal with this. One is to put hatcheries on streams with no habitat left, like those with impassable dams, or those that are gutshot by development, and leave the streams that support wild fish alone. The other way, which is what is done for the most part in Washington, is to have a strain of hatchery fish that spawns too early for any offspring to survive in the wild, and too early for them to outcross with wild fish. This is not a perfect solution - early wild runs of steelhead go extinct for example. Early winter hatchery fish also get mixed up with summer fish, which pretty much spawn at the same time, and produce morphodite runs with horridly compressed timing. Again, the solution is to have summer hatcheries and winter hatcheries but not on the same river. Finally, we need aggressive removal of hatchery fish and complete release of wild fish - lets do what we can to even the odds. And lets give up on sacrificing valuable wild fish in misguided hatchery broodstock programs.

Hey, everybody on this board doesn't go off like this, so don't regret signing up. And go catch some fish - except a trip or two to the OP in my buddy's drift boat I'm into blackmouth for the next few weeks, and then I descend upon the Columbia to do my damndest to remove those 460,000 inferior hatchery spring chinook from the gene pool singlehanded