The simplest kind of agreement of truth is a resemblance between the representation and its object. I call this verisimilitude, and the representation a copy.
Resemblance consists in a likeness, which is a sameness of predicates. Carried to the highest point, it would destroy itself by becoming identity. All real resemblance, therefore, has a limit.
A likeness represents its object by agreeing with it in some particular. [—] Scientifically speaking, a likeness is a representation grounded in some internal character - that is whose reference to a ground is prescindible.
… the relation of a repraesentamen to its object (correlate) may be a real relation and, then, either an agreement or a difference, or it may be an ideal r[elati]on or one from which the reference to a correspondent (subject of representation) cannot be prescinded by position. In the first case, that is where the repraesentamen has a real agreement with its object, the representation consists in a likeness; a simple quality of the object is shown but the object itself is not said to exist.
… I must call your attention to the differences there are in the manner in which different representations stand for their objects. In the first place there are likenesses or copies - such as statues, pictures, emblems, hieroglyphics, and the like. Such representations stand for their objects only so far as they have an actual resemblance to them - that is agree with them in some characters. The peculiarity of such representations is that they do not determine their objects - they stand for anything more or less; for they stand for whatever they resemble and they resemble everything more or less.
A reference to a ground may also be such that it cannot be prescinded from a reference to an interpretant. In this case it may be termed an imputed quality. If the reference of a relate to its ground can be prescinded from reference to an interpretant, its relation to its correlate is a mere concurrence or community in the possession of a quality, and therefore the reference to a correlate can be prescinded from reference to an interpretant. It follows that there are three kinds of representations.
1st. Those whose relation to their objects is a mere community in some quality, and these representations may be termed Likenesses.
2d. Those whose relation to their objects consists in a correspondence in fact, and these may be termed Indices or Signs.
3d. Those the ground of whose relation to their objects is an imputed character, which are the same as general signs, and these may be termed Symbols.
The likeness has no dynamical connection with the object it represents; it simply happens that its qualities resemble those of that object, and excite analogous sensations in the mind for which it is a likeness. But it really stands unconnected with them.
An icon is a sign which stands for its object because as a thing perceived it excites an idea naturally allied to the idea that object would excite. Most icons, if not all, are likenesses of their objects. [—] It may be questioned whether all icons are likenesses or not. For example, if a drunken man is exhibited in order to show, by contrast, the excellence of temperance, this is certainly an icon, but whether it is a likeness or not may be doubted. The question seems somewhat trivial.
Peirce often used the term "likeness" instead of "icon", especially in the 1890s. However, as Peirce refined his notion of the icon (and defined it in terms of pure Firstness or possibility) he increasingly preferred "icon", at times suggesting that "likeness" does not cover exactly the same semiotic terrain as "icon". This possible distinction between likeness and icon has often been overlooked, perhaps because of errors in the Collected Papers. Judging by the available evidence, the editors of the Collected Papers now and then substituted "icon" for Peirce's "likeness", (see e.g. CP 2.299 and EP 2:9). On the other hand, the distinction between "icon" and "likeness" is not strictly adhered to by Peirce himself; as he suggests, the whole question may be somewhat trivial. MB
… I had observed that the most frequently useful division of signs is by trichotomy into firstly Likenesses, or, as I prefer to say, Icons, which serve to represent their objects only in so far as they resemble them in themselves; secondly, Indices, which represent their objects independently of any resemblance to them, only by virtue of real connections with them, and thirdly Symbols, which represent their objects, independently alike of any resemblance or any real connection, because dispositions or factitious habits of their interpreters insure their being so understood.