Correct thought would prevent the extension of the value of the symbol beyond the original element of similarity. More than this, it would recognize the fact that similarity does not suppose identity, but the reverse, to wit, defect of likeness; and this dissimilitude must be the greater, as the original and symbol are naturally discrepant. The supernatual,[TN-11] however, whether by this term we mean the unknown or the universal--still more if we mean the incomprehensible--is utterly discrepant with the known, except by an indefinitely faint analogy. In the higher thought, therefore, the symbol loses all trace of identity and becomes merely emblematic. The ancients defended symbolic teaching on this very ground, that the symbol left so much unexplained, that it stimulated the intellect and trained it to profounder thinking;[205-1] practically it had the reverse effect, the symbol being accepted as the thing itself. Passing from these general rules of the selection of symbols, to the history of the symbol when chosen, this presents itself to us in a reciprocal form, first as the myth led to the adoption and changes in the symbol, and as the latter in turn altered and reformed the myth. The tropes and figures of rhetoric by which the conceptions of the supernatural were first expressed, give the clue to primitive symbolism. A very few examples will be sufficient. No one can doubt that the figure of the serpent was sometimes used in pictorial art to represent the lightning, when he reads that the Algonkins _straightly_ called the latter a snake; when he sees the same adjective, spiral or winding, (ἑλικοιεδης) applied by the Greeks to the lightning and a snake; when the Quiché call the electric flash a strong serpent; and many other such examples. The Pueblo Indians represent lightning in their pictographs by a zigzag line. A zigzag fence is called in the Middle States a worm or “snake” fence. Besides this, adjectives which describe the line traced by the serpent in motion are applied to many twisting or winding objects, as a river, a curl or lock of hair, the tendrils of a vine, the intestines, a trailing plant, the mazes of a dance, a bracelet, a broken ray of light, a sickle, a crooked limb, an anfractuous path, the phallus, etc. Hence the figure of a serpent may, and in fact has been, used with direct reference to every one of these, as could easily be shown. How short-sighted then the expounder of symbolism who would explain the frequent recurrence of the symbol or the myth of the serpent wherever he finds it by any one of these! This narrowness of exposition becomes doubly evident when we give consideration to two other elements in primitive symbolism--the multivocal nature of early designs, and the misapprehensions due to contiguous association.