Mimics know that their chances of success increase if they mimic objects that adventurers aren't expecting to be mimics. There is, of course, the eternally unstable metagame around mimicking chests. It's all too easy to say "surely nobody opens a chest in a dungeon without checking first," but have you ever known adventurers to observe any safety rule with consistency? Nevermind that mimics have in many cases adapted their tactics to discourage proper checking. Nobody wants to ask the party's warrior to give that delicate-looking gold reliquary a good solid chop with an axe. So adventurers often fall back on ineffectual tests; little taps with a staff, for example, that a mimic knows to ignore. Or they may rely on outdated and superstitious ways to "tell", such as looking at the position of chains or clasps. And so, adventurers continue to regularly be eaten by mimics.
Still, mimicking a chest is a conservative strategy that succeeds more often if a dungeon has at most one chest mimic; perhaps two. Inasmuch as mimics are capable of rudimentary cooperation, this leaves juvenile mimics to imitate things other than chests. The most popular choice are mobile forms of treasure such as magic weapons or jewellery. Mimic larvae often imitate coins; though they are too small to reveal themselves, this disguise allows them to eventually be transported to a different dungeon through the adventurer to village to dragon's hoard cycle. This mechanism is the most common way mimics spread; it is estimated that about 70% of dungeons with mimic infestations were originally infected through this method.
However, this behavior poses a particular risk to the mimic. Mimics are obligate cohabitants of dungeons with other creatures. They are neither capable of nor inclined to clear out a lair for their exclusive habitation. As their diet consists exclusively of adventurers, mimics must also live in dungeons that are desirable to delve; this of course requires the presence of some dungeon inhabitant that either poses a threat to local surface communities, or can be extractively looted for resources such as a dragon's treasure or a lich's tomes.
Such inhabitants will of course make the dungeon their home, which often involves rearranging the furniture; unbeknownst to them, the furniture are often mimics. Some mimics will be dormant in chest form. Some will be in the form of objects that your typical garden-variety kobold is compelled to store in chests. You see the problem, right?
Mimics do not take well to being placed inside other mimics. Whether the phenomenon is metaphysical or purely physiological, mimics in such circumstances can become entangled. The mobile and mutable fibers that make up their bodies weaving together to form knots. Adventurers typically report such cases in the wild: Large 'chests' (adult mimics) that look bloated, inflamed, or misshapen. When they attempt to unfold and attack they reveal the dangling joined-up bodies of juvenile mimics clinging to their innards. Such 'mimic kings' are not really capable of serious combat; though they are often older specimens, their inability to fluidly move renders them completely vulnerable.
This phenomenon may explain the apparent limits to mimic populations. Naive calculation would suggest that, even given the mimic's slow reproductive cycle, many long-infested dungeons ought to contain enormous numbers of mimics; perhaps one mimic to every 'real' weapon, small mobile treasure, or chest. In reality, such dungeons typically contain either a mimic king, or used to.
