I recently looked over the polymorph spell chain for Dungeons & Dragons 3.5e (Alter Self -> Polymorph -> Shapechange) while reading suggestions on forums for what to turn in to with the spell, and I found something that confused me.

Alter Self contains the line: "The new form must be within one size category of your normal size." Polymorph says it functions like Alter Self, except where otherwise noted. The only thing Polymorph says about Size is the line: "You can’t cause a subject to assume a form smaller than Fine", which doesn't seem like it would invalidate the Alter Self clause. Shapechange, on the other hand, specifically clarifies that whatever you turn into can be anywhere from Fine to Colossal in size, which seems like it would overrule the Alter Self clause.

This is confusing to me because when reading over forums and suggestions, people are suggesting that you use Polymorph to turn into Huge-sized creatures, such as Hydras. Is there an errata or clarification that I was unable to find that would allow this?

Can a (non-Large) character use the Polymorph spell to become a Huge creature?

Rules as written, it seems as though this is explicitly not allowed until you acquire Shapechange (unless you're a naturally Large-sized creature, since size-changing spells do not stack together as per the spell Enlarge Person).

I'm willing to accept that this is just an example of the polymorph chain of spells being poorly written, as the Alter Self clause: "A body with extra limbs does not allow you to make more attacks" had to be clarified by Skip Williams to only mean WEAPON attacks in the "Rules of the Game: Polymorphing (Part Two)" article on Wizards of the Coast's website.


1 Answer 1


which doesn't seem like it would invalidate the Alter Self clause.

Yes, it does.

Would it be better as a positive statement, saying polymorph can assume sizes Fine and larger? Yes, that would be more clear. But we have official rules text confirming a wide variety of differently-sized creatures as being possible for polymorph, and we have Skip Williams (one of the developers of 3.5e) saying polymorph can handle anything from Fine to Colossal.

First, there is this example

Same Effect with Differing Results: The same spell can sometimes produce varying effects if applied to the same recipient more than once. For example, a series of polymorph spells might turn a creature into a mouse, a lion, and then a snail. In this case, the last spell in the series trumps the others. None of the previous spells are actually removed or dispelled, but their effects become irrelevant while the final spell in the series lasts.

(Player’s Handbook pg. 172)

A mouse and a lion are Fine¹ and Large, respectively—there is no size category that is within ±1 size category of both, so it is impossible for two polymorph spells applied to the same creature to have both of those results if polymorph is restricted to ±1 size category from the target.

Moreover, Skip Williams directly addresses this question, saying,

  • The assumed form’s size can be anything from Fine to Colossal.

The kind of creature you choose for the assumed form determines the size. The subject becomes the same size as an average member of its kind. For example, if you turn the subject into a troll, the subject becomes size Large, which is the standard size for a troll.

(“Rules of the Game: Polymorphing (Part Three)”)

There are a number of problems with the Rules of the Game series (similar to, but perhaps less frequent than, the problems with the FAQ), and ultimately the “Same Effect with Differing Results” section is a secondary source, but considering we have an ambiguous statement in polymorph that seems (to many) to overrule the statement in alter self, looking to these give strong and clear indications of the authors’ intent: they intended that the line about sizes in polymorph would replace the size limitations on alter self.

  1. The mouse is statted in the Dungeon Master’s Guide, page 203, not in the Monster Manual. The specific stats are for a mouse familiar, but becoming a familiar doesn’t change a creature’s size, so the Fine listed for the mouse familiar would also apply to the non-familiar mouse form for polymorph. As far as I know, there are no stats for a “snail” creature, but it is presumably also Fine—even Fine is kind of large for your typical garden snail, though some snails do get that large or larger.
  • 1
    \$\begingroup\$ "For example, a series of polymorph other spells might turn a creature into a mouse, a lion, and then a snail" (Player's Handbook (2000) 154). The spell polymorph other, in part, says, "The new form can range from Diminutive to one size larger than the subject's normal form" (236)… among a host of other differences. My feeling, given the history, is that other was deleted to match 3.5 naming conventions rather than the author carefully considering the myriad of rules changes that polymorph effects had undergone in the ensuing years then deleting the word other. \$\endgroup\$ Mar 12, 2021 at 19:11
  • \$\begingroup\$ @HeyICanChan As the mouse is Fine (and the snail probably is), it wouldn’t be an eligible form for polymorph other, so that’s a whole ’nother kettle of fish. Unless the size of mice and snails changed, too, which I suppose is possible. \$\endgroup\$
    – KRyan
    Mar 12, 2021 at 19:14
  • \$\begingroup\$ The polymorph other spell from AD&D, Second Edition — which is likely the blueprint the designers were working from — suggested turning victims into goldfish, so that example's not too far away from the sacred cow they were so hesitant to slaughter, despite the new version maybe yielding improbably enormous mice and snails? The fact remains that it was a bad example 20 years ago, and the 3.5 revision made it worse. \$\endgroup\$ Mar 12, 2021 at 19:20
  • 1
    \$\begingroup\$ It's unlike your typical answer to put such faith in the designers—I mean, this is the same edition where power word stun is still a conjuration spell in the latest DMG (2012) (299)—, but I respect the stance despite me personally finding the evidence weak. \$\endgroup\$ Mar 12, 2021 at 19:49
  • 1
    \$\begingroup\$ However, in this answer, the idea that polymorph should remove size restrictions hinges on a lightly revised originally-bad-but-now-conveniently-almost-makes-sense example and a Web article in a series that distills much of its content out of thin air. I'm glad the answer has helped others, and I've not downvoted it, but I do think the answer demonstrates a lot of trust that the designers just never earned. I really think this issue is cloudy enough to require the DM making a ruling instead of this being regarded as standard, but I'll let it go. You can have the last word if you want. \$\endgroup\$ Mar 12, 2021 at 20:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .