"Model: Anything of a particular form, shape, size, quality, construction, or the like, intended for imitation". [Webster's Twentieth-Century Dictionary of the English Language, The World Syndicate Publishing Company, Cleveland and New York, 1936.]

Chemistry has been called "the molecular science". Yet none of us has ever seen a molecule. Our knowledge of their properties, shapes, and behaviors is almost entirely inferential. Although devices such as the scanning tunneling electron microscope or the atomic force microscope allow us to create "pictures" of molecules, these fuzzy collections of lumps carry little information for us beyond what we already know - from models.

Nobel Laureate Roald Hoffmann has written: "What we represent (when we draw a molecule) is what we want to represent: we abstract a piece of reality to show it to another person." [R. Hoffmann and V. Torrence, Chemistry Imagined: Reflections on Science, Smithsonian Institution Press, Washington, 1993.]

In other words, we build a model that carries the particular information about the molecule that we wish to convey. In fact, sometimes we build an actual physical model, of plastic or metal or wood. But more frequently, we build the model on paper. Consider the representations below. Each is a model of menthol, the familiar odiferous terpene:

The structure on the left is drawn as molecular structures were presented in textbooks when I was learning organic chemistry. It is intended to tell us which atoms are connected to which, and, if we cared to do a little addition, the numbers of atoms of each kind present in menthol. Indeed, it is a form of Lewis structure, in which carbon-carbon bonds are represented as lines, and carbon-hydrogen bonds are implicitly understood to be present but unrepresented.

The next picture is more familiar; any organic chemist could easily interpret it and extract the same information conveyed by the first structure. The third model attempts to show the viewer something of the three-dimensional structure of the molecule, using dashes and wedges to indicate relative placement in space. At a higher level, we provide this information in a perspective drawing, as in the fourth structure, showing also that the 6-ring is not flat.

Here is yet another model:

Even a layperson would suggest that this is a molecule! Toy stores sell kits that allow children to assemble wooden versions of this model. It provides yet more information, displaying the "correct" proportions of one part of the molecule relative to another. But the sizes of the balls representing the atoms and the sticks representing "bonds" are arbitrary. So we take one step further toward "reality":

Here we try to represent the volume of space actually occupied by the atoms, the van der Waals surface. Now, however, the positions of the atomic nuclei are somewhat obscured; we have sacrificed some information to gain a different sort.

But of course, not even this model represents "reality". Molecules are not static. The atoms vibrate about the mean positions indicated in these last two models. Furthermore, the volume occupied by the atoms is also arbitrary. It is chiefly the volume occupied by the electrons, which cannot be specified exactly, but rather must be described in terms of probabilities. [And the colors, also, are arbitrary. Is oxygen really red?]

Which of these models is menthol? All of them - or none of them. None is the real thing, but each has useful aspects. We select a model that contains the information we need to talk about a particular set of molecular properties, and draw it on paper, or project it on a screen. For that moment, in that place, the model is the molecule.

We must be very careful, however, not to lose sight of the arbitrariness of our model-building. Lewis structures, hybrid orbitals, resonance - all provide adequate descriptions of some properties of organic molecules. But in a sense, each is simply a serviceable lie. Each is merely a model that can be improved upon by molecular orbital theory (which may someday in its turn be improved upon by some newer model.)

At this highest current level of model building, hybridization, resonance, electron pair bonds, all vanish. Molecules are held together by electrons occupying molecular orbitals, of which only the very rare example is associated with two and only two atoms.

Many years ago, the British theorist C. A. Coulson wrote: "Sometimes it seems to me that a bond between two atoms has become so real, so tangible, so friendly that I can almost see it. And then I awake with a little shock: for a chemical bond is not a real thing: it does not exist: no one has ever seen it, no one ever can. It is a figment of our own imagination.... Here is a strange situation. The tangible, the real, the solid, is explained by the intangible, the unreal, the purely mental." [C. A. Coulson, J. Chem Soc., 1955, 2069.]

Here is the molecular orbital occupied by the highest energy pair of electrons in menthol: