Okay, I think you've both convinced me that each spell should be its own separate opportunity for a discovery. So the researcher chooses between an artificially inflated spell (which, as Jonathan points out, maybe is a reasonable tactic for discovering something new) or multple, independent spells, each of which has its own chance for a discovery. Two little points remain, both having to do with the latter option:
- Should we overrule the rule on ArM5 page 102, about one Extraordinary Results roll applying to all spells, in the case of original research? If each one has a chance to yield a discovery, seems like we should be rolling separately for each.
- We might as well decide, just in case two spells yield discoveries in the same season: one season per spell to stabilize? or one season overall - a more exact recreation of the experimentation? Again there are opposing incentives - speed vs. Warping/Twilight; I'm just trying to pin down what's possible.
And one other thing (sorry, not trying to be nitpicky here!): Jonathan, I made one last comment on the simple-die-during-research question. Just wondering if that affected your judgment on the issue.