Its outsized importance in my memories probably comes from entering my life at a very sensitive time. I'd started to feel stirrings of dissatisfaction with AD&D, but I had not yet been exposed to any alternatives. So a supplement that promised to give me alternate magic systems was something I hungered for. Even if I didn't quite understand the Skills & Powers character creation process (funnily enough, this book and Combat & Tactics were the only two Player's Option books I had while I was still actively playing AD&D - Skills & Powers and DM's Option: High Level Campaigns I picked up at used book stores many years after the fact, mostly because I remembered them being cited in these books I really loved).
My teenage campaign settings were heavily inspired by this book. I'd quite consistently chuck PHB standard mages for Artificers or Channelers or Dimensionalists. I was searching for . . . something, though I had no way of knowing that the Spells & Magic alternate magic systems were even worse than the AD&D default.
Oh, they're tricky, because there are some interesting ideas here. If you play a witch, then your character is on a campaign-long arc, where the more they use their magic, the greater the risk of drawing the attention of their dark patrons. Over time, they gain mutations that simultaneously grant them useful abilities and alienate them from human society.
This is accomplished with a percentage roll that you make every time you cast a spell. It starts off pretty modest, but by the time you're in the mid level range, it starts to get to 5-10% every time you cast one of your highest level spells. Since you can only fail 4 times before your character becomes an unplayable NPC, the degeneration is likely both inevitable and rapid. So it's an annoying mechanic that adds an extra step to every spell and eventually trashes your character. But if you could choose the pace and make the degeneration subordinate to your roleplaying it might actually be pretty cool.
Although the biggest problem with the Witch class is actually also the biggest problem with all the new spellcasting models - the new spell points system is a hot mess.
Brief tutorial on D&D's magic system. Spells are divided into 9 levels, ranging from basic level 1 magic like Light all the way up to game-breaking show-stoppers like Time Stop. As your character advances, they get spell slots. Each slot is associated with a particular spell level. So a 3rd level wizard would have two 1st level spells and one 2nd level spell. They fill those spell slots by "memorizing" spells, picking particular spells from their list of spells known to be able to cast. You can memorize spells at most once per day, but if you don't use your spells, they stay memorized. If you want to be able to use the same spell more than once per day, you have to memorize it twice or more.
It's exactly as exhausting as it sounds. It's not so much the resource management that's the problem as it is the constant necessity to second-guess the next day's adventure. In practice, you mostly wind up loading up on your most generically useful spells and hoping for the best.
So, if I told you that Spells & Magic introduced a spell point system to replace the standard way of memorizing spells, you might have certain expectations about that. The obvious idea - spells of different levels have a varying point cost and you spend points from your pool to cast them. The pool refreshes at a set rate, probably around once per day. Simple, easy, elegant. You know, not the Player's Option way at all.
The way Spells & Magic's spell point system works is that you spend points to memorize spells. This gives you more versatility, in that you can basically exchange slots at undesirable spell levels for more slots at your optimum spell levels. But that was never really the problem with the spell memorization system. You've still got to basically try and predict the course of your whole day.
As an option, you can devote some of your spell points to "free" slots, at double the usual point cost, allowing you to cast any spell you know instead of memorizing it. This does address the main problem with D&D's magic system, but sacrificing half your spell inventory is a bitter pill to swallow.
Even so, that would merely be an unnecessarily convoluted refinement to an AD&D system that somehow, inexplicably, winds up retaining the original rules' worst flaws - typical Player's Option stuff - were it not for the fact that the new special spellcasting paradigms then proceed to mutilate those rules almost beyond comprehension.
Take the Channeler, the magic class that draws upon the caster's health and stamina to power spells.
While the character may have some spell points "allocated" or "tied up" in various fixed and free magicks, this actually makes no difference for a channeler. The initial selection of spells is simply used to create a slate of spell powers that the character can access and to define the cost in spell points for making use of these powers. The character may cast any spell that he has available through either a fixed or free magick, except that the the magick does not vanish from his memory once he's cast the spell. Instead, the character deducts the number of spell points required to energize the spell from his spell point total. For example, if a mage with 40 spell points has a magic missile memorized, he can cast that magic missile four times if he wants to! (Editor's Note - Magic Missile costs 4 spell points, making that example completely arbitrary and only slightly illuminating).The system kind of works, once you've wrapped your head around it. The Channeler almost matches the naive assumptions about how a spell point system should go, but then it also adds the overly-complex spell memorization system on top of that. And it very confusingly uses the same terminology for both. I guess the Channeler is "balanced" by not having access to their whole spell repertoire at any given time . . . except that they are given the option to devote some of their limited slots to free magic, and while the double point cost still stings, it's a bit less of a burden when you regain spell points by the hour. Ultimately, since spell point costs scale geometrically, only your highest level spell really needs to be put in a fixed slot.
Or, you know, you could just pick a carefully curated list of powerful and thematic spells and "memorize" those to give your character a fixed magical persona, which probably should have just been baked-in to the class from the start, instead of requiring you to do fucking algebra to figure out your character abilities . . . but what do I know?
Anyway, you can see the prototype of 3rd edition's sorcerer in these rules, and that's pretty neat, though Dungeons and Dragons is probably cursed to never have a truly good spellcasting system.
The only other part of this book I ever used was the new spells at the back. They're fine. Nothing stands out as "must have." And very little even seems worth toting around an extra book for, but there are a few gems, like Heart of Stone, which allows a wizard to ritually remove their own heart, thereby making them nearly invincible.
Most of the rest of the book is stuff that's . . . not useless, really, but mostly incredibly dry or needlessly complex. There's a system where you can use your proficiency slots to buy "signature spells," which you never don't want to do, because they give you extra castings as well as making those spells more effective. And there are rules for spell critical hits, which like all rpg critical hit tables, make the game dangerously unpredictable while reveling in the grotesque.
The best parts of the rest of the book are the DM advice, which comes this close to getting it. When it talks about creating new schools of magic for wizards, it almost comes to the realization that the core 8 are kind of awful, and that thematic schools like elementalism, dimension magic, and shadow magic, are much better, but it never quite delivers a value judgement. Instead it just goes for a kitchen sink approach, and suggests that whatever you want to do, it's cool. Not a bad position for a book meant to be universal, but frustrating nonetheless.
Spells & Magic did not turn out to be a very good book. It's a devastating revelation so far as my adolescent nostalgia is concerned, but ultimately kind of a relief. It means that once I finish up with these AD&D 2nd edition core books, I can be done with them forever.
Ukss Contribution: This one is going to be semi-cheating. I am picking something from this book, but the reason this particular entry is interesting to me is because of context from earlier books in the AD&D 2nd edition line.
Spells & Magic introduces a new spell called Leomund's Hidden Lodge. What it does is create for you a sort of temporary house that is disguised as part of the local terrain - a giant boulder, a copse of trees, etc.
What I find so amusing about this spell is how completely unnecessary it is. It's useful, but it's part of a chain that starts with Leomund's Tiny Hut at 3rd level, goes to Leomund's Secure Shelter at 4th, and ends with Mordenkainen's Magnificent Mansion at 7th level (I guess ol' Leomund crapped out before he could get to the deluxe level). Each iteration adds functionality, sure, but given the way D&D magic works, who is going to bother learning them all?
Well, Leomund, obviously. I'm not going to transfer the character directly, though, hilariously, when I looked him up on wikipedia, I discovered that he canonically co-authored a book called Architecture with his friend Mordenkainen, which sounds exactly like a joke I would have made. However, he also did a bunch of stuff that is of no interest to me whatsoever, so I'm just going to steal the concept of house-omancy.
I'm always on the lookout for concepts for new magic wands, and I think a Wand of Shelters might be pretty cool.