For me, matching the "hatch" means figuring out what the fish are eating and simulating it in a way that the fish will take using fly hooks and tying materials (which in themselves are constantly evolving). Tradition and sentimentality play a role in my preferences, but my overriding goal is to fool fish into taking my offerings. I believe this is the essence of artificial lure fishing - fly, spin, or baitcasting. Having spent the entire first decade of the new millenium plus a couple of years fishing the Ozarks tailwaters and spring fed fisheries for trout and smallmouth bass with a fly rod, I fished more scuds, sowbugs, and aquatic worm flies than I did dry flies during that time. Trout in the Ozarks also take small egg patterns very well, and most anglers think this because of spawn or the conditioning to pellet feed in hatcheries. But there is another "angle" on the small balls appearing as trout food that will also explain why totally wild pan fish and the rare smallie will take an egg pattern too. Aquatic worms and earthworms washed into streams often roll up into balls when washed free into moving water. It's a defense mechanism. So hatchery-reared trout in the Ozarks could see the round fly as one of several potential food sources, but even totally wild fish see them as one or two likely sources of sustenance as well...even when there is no spawn going on.

Neil, your scud and sowbug fly recommendations are as good as any and simpler than many. For my $$$, that makes them better choices for the fly angler who ties his/her own flies. In moving water, dead drifting them is almost always sufficient. Just bounce them along the bottom. If people are not hanging up every 3-4 cast (at least for a second or so), then you're not getting down far enough. In still water, you have to move the scuds and sowbugs just as you described.

In my opinion, anyone who goes to fish nutrient-rich water...spring-fed stuff in particular...without a fly box full of a good variety of this stuff is unarmed.