In old-school D&D, a magic-user’s ability to gain new spells differs based not only on which edition you’re playing, but on the DM’s whim. For instance, Holmes Basic allows you to learn new spells from scrolls and other magic-users, but it gives the DM carte blanche to decide what spells you’ll run across. One DM might provide stacks of scrolls and armies of friendly mentors, while another’s game may have no scrolls or mentors at all! Moldvay Basic is even stricter; not only does it restrict magic-user spell acquisition to leveling up and spell research, but even then it explicitly allows the DM to decide what spells you start with. He may let you choose a powerhouse spell like Sleep as your starting spell, or he may stick you with Hold Portal or Shield.
Spell research is available as an option by which you can individualize your magic-user’s spell selection, but it’s also very difficult to pull off. Not only is it expensive and risky, but it takes the magic-user out of play for weeks at a time, a severe penalty for a class that’s already slow to level up.
To encourage spell research rather than discourage it, I’m experimenting with some changes to the Moldvay rules:
1) Since we’re using Carousing rules (in which PCs gain experience points from frivolous expenditures of gold), I’ve ruled that gold spent on research counts as carousing, and thus is a good way to cash in your gold for XP, especially if it bypasses the usual caps on how much gold you can spend in one go.
2) I’m greatly reducing the time required to research new spells. Two weeks for a first level spell will take a magic-user PC out of play for several sessions, and it only gets worse from there! Halving that is a good start, and I’ll halve it again when researching a spell that’s similar to one that the magic-user already knows; this should encourage themed spell research akin to the suite of Bigby’s Hand spells.
3) Attempting to learn a spell from a teacher, captured spellbook or scroll will have a chance of failure, and failing to learn the spell precludes another attempt until the magic-user gains a level. Researching a variant of the spell in question should prove a viable alternative to waiting to level up.
How these rules will work in practice is anybody’s guess. If the party’s magic-users continue to die off at their customarily swift rate, it may never come up…