Since the proposal of tetravalent and self-linking carbon in 1857-8, structural diagrams and physical models have played an increasingly important role in organic chemistry. The demand for models adequate for discussing the complex molecules of biology and their interactions has driven the development of computer graphics to display virtual models.

To understand, appreciate, and use molecular formulae and models, it is helpful to stand back and see just what they are, and how they do and don't reflect experimental reality. The historical perspective gained by looking at very early diagrams and models is particularly helpful in this process, because we can see so clearly when they were "right" and when they were "wrong", and how they have influenced what we write and build today.

Here are some questions to consider as you read:

How did Kekulé's tetravalent carbon models differ from those of Hofmann and Dewar, and how might this difference have influenced his student van't Hoff?

Science lectures for general audiences of all classes of society were a pervasive form of entertainment in Victorian England, where self-improvement was a popular avocation. One of the most influential forums was the Friday Evening Discourses at London's Royal Institution of Great Britain, which were organized from 1832 to 1862 by Michael Faraday (discoverer of benzene and so much else). The Discourses continue to the present day in Faraday's historic lecture theater. These events, which begin promptly at 9 pm and last exactly one hour, have for most of two centuries attracted the most influential leaders of British science, intellectual and commercial life, government, and society, including the royal family.

The Discourse on Friday, April 7, 1865, was presented by August Wilhelm Hofmann, who had been Liebig's student and assistant in Giessen. Twenty years earlier Liebig had recommended to Prince Albert, Queen Victoria's German consort, that London's new Royal College of Chemistry hire Hofmann as professor. Albert's hope was to bring British chemistry up to modern German standards.

This Friday evening was Hofmann's British swan song, since he was about to assume the chair of chemistry at the University in Berlin. His topic was "On the Combining Power of Atoms". For the occasion he had apparently lined the lecture theater with row upon row of formulae for organic molecules.

Hofmann's discourse was presided over by the 23-year-old Prince of Wales, later Edward VII, who was thus indulging one of his more admirable enthusiasms.

To set this particular Friday evening in historical context consider that at 10 pm Greenwich time, the very moment when Hofmann was concluding his lecture, Ulysses S. Grant was writing as follows to his former comrade-in-arms Robert E. Lee to propose a meeting that would take place on Sunday at Appomattox Court House:

5 p.m. April 7, 1865
The result of last week must convince you of the hopelessness of further resistance - I - regard it as my duty to shift from myself the responsibility of any further effusion of blood, by asking of you the surrender of that portion of the Confederate States Army known as the Army of Northern Virginia.

The following Friday evening, Good Friday, there would be no Discourse. Who can know how history might have changed if, on that day, Abraham Lincoln had been able to attend a Discourse at an American equivalent of the Royal Institution, rather than the British comedy"Our American Cousin" at Ford's Theater in Washington?

---

Thus we distinguish the chlorine atom as univalent, the atom of oxygen as bivalent, that of nitrogen as trivalent, and lastly the carbon atom as quadrivalent. - this I believe I can show you by a very simple contrivance. I will on this occasion, with your permission, select my illustration from that most delightful of games croquet.

Let the croquet balls represent our atoms, and let us distinguish the atoms of different elements by different colours. The white balls are hydrogen, the green ones chlorine atoms; the atoms of fiery oxygen are red, those of nitrogen, blue; the carbons atoms, lastly, are naturally represented by black balls. [Note that most model sets still use Hofmann's colors for the atoms!] But we have, in addition, to exhibit the different combining powers of these atoms. This we accomplish by screwing into the balls a number of metallic arms (tubes and pins), which correspond respectively to the combining powers of the atoms represented, and which, while constituting an additional feature of distinction, enable us at the same time to join the balls and to rear in this manner a kind of mechanical structures in imitation of the atomic edifices to be illustrated. [Like Kekulé, Hofmann had been a student of architecture, until Liebig's lectures lured him to chemistry. In fact Hofmann's architect father was responsible for expanding Liebig's laboratory.]

Here is Hofmann's set of models:

Here is how he demonstrated the formation of water:

and of Marsh Gas:

and the Oxides of HCl:

and the Chlorides of Marsh Gas:

His pičce de résistance at the close of the lecture was to use his croquet-ball models in a risky way, for a purpose for which they had not been designed, namely, to explain the reactivity of the olefiant gas (ethylene, C₂H₄):

In building up the molecule of olefiant gas by the insertion into the marsh-gas molecule of one atom of carbon only, we obtain what hitherto we would have called an unfinished molecule, i.e. a molecule in which two of the attraction units of the second carbon atom are unsatisfied. Indeed, a glance at our model shows us that two carbon arms project uncovered. We are thus led to inquire whether unfinished molecules, i.e. molecules in which a certain number of attraction units remain unbalanced, are capable of a separate existence. This question is accessible to experiment. Olefiant gas, indeed, possesses all the characters which, granting for argument's sake the possibility of its existence, we are inclined to attribute to an unfinished molecule. In the cases hitherto considered, we saw the chlorine atom, when admitted into a molecular structure, always entering with substitution, hydrogen separating from the chlorinetted molecule in the form of hydrochloric acid. - Submitting on the contrary, olefiant gas to the action of chorine, we find that the chlorine is fixed directly, without substitution, the chlorine atoms meet, so to speak, with vacant spaces existing in the olefiant gas molecule; in order to get in, they need not expel a corresponding number of hydrogen atoms to make room for them. The compound generated is the so-called Dutch liquid, an oily substance first produced by an association of Dutch chemists at the close of the last century. It was the production of this oily liquid that gave rise to the name of olefiant gas. -

We have thus been led, step by step, to a distinction of a novel kind, that of finished and unfinished molecules ; or, to use the more frequently employed expression, that of saturated and non-saturated compounds. I need not tell you that this distinction carries us to the threshold of a new field of research, hitherto crossed only by a small band of fearless pioneers, who are encountering difficulties on all sides. Admitting, as we are compelled to do, the existence of what we have called unfinished molecules, we inquire under what special conditions, at what special stages the growth of a molecule may be arrested? How is it that as yet the marsh-gas molecule is known only in the finished state, CH₄, that none of the fragmentary marsh-gases, CH₃, CH₂, and CH₁, which might exist, have ever been obtained? Again, how is it that the molecule of hydride of ethyl exists, so to speak, finished and unfinished ; and, lastly, that of the several fragmentary states in which this molecule might be met with, two only, namely the two states, C₂H₄ (olefiant gas), and C₂H₂ (acetylene), have ever been observed?

The intricate formulae that hang upon these walls, and the boundless variety of phenomena they illustrate, are beginning to be for us as a labyrinth once impassable, but to which we have at length discovered the clue. A sense of mastery and power succeeds in our minds to the sort of weary despair with which we at first contemplated their formidable array. For now, by the aid of a few general principles, we find ourselves able to unravel the complexities of these formulae, to marshal the compounds which they represent in orderly series ; nay, even to multiply their numbers at our will, and in a great measure to forecast their nature ere we have called them into existence. It is the great movement of modern chemistry that we have thus, for an hour, seen passing before us. It is a movement as of light spreading itself over a waste of obscurity, as of law diffusing order throughout a wilderness of confusion, and there is surely in its contemplation something of the pleasure which attends the spectacle of a beautiful daybreak, something of the grandeur belonging to the conception of a world created out of chaos.

---

---

Crum Brown's Structural Diagrams (1861)

Alexander Crum Brown was a medical student at the University of Edinburgh (and won the gold medal in chemistry) in 1858, when Couper, one year his senior, published his "New Chemical Theory" in a leading British journal. Late in that year Couper, having been dismissed from Wurtz's Paris lab for intemperate complaints about Wurtz's delay in having the French version published, returned to Edinburgh as assistant to Playfair, the chemistry professor. Couper became permanently incapacitated within a few months, but still it seems incredible, though apparently true, that Crum Brown did not of know of him and his amazing paper. In his 1861 M.D. thesis Crum Brown proposed a notation for valence that refined Couper's line representation, which he may not have known. Crum Brown's notation has survived with minor changes to the present. The following examples are taken from "On the Classification of Chemical Substances, by means of Generic Radicals", which was read to the Royal Society of Edinburgh February 5, 1866. (Transactions of the Royal Society of Edinburgh, 24, 331-339, 1867)

He showed several reactions in the paper including addition of HCN to a ketone (note that he makes the ketone generic by denoting the carbon substituents by R, which stands for "radical", a symbol first used by Berzelius in 1838.):

and dehydration of "true" and "pseudo" alcohols to the same alkene (now they are called primary and secondary):

The year after he wrote the footnote promoting his "sausage" notation, over Crum Brown's formulae, Kekulé succombed to using lines to denote bonds. He had, after all, been using ball-and-stick valence models in his lectures. He used the following model of benzene in 1866. Note that the sticks could be linked by short pieces of rubber tubing.

This type of model must surely have influenced Kekulé's alumni, including Körner (who almost certainly showed such a model to Paternň in Palermo), and van't Hoff.

Dewar's Brass Strip Models (1866)

---

James Dewar was 25 years old when he delivered this paper to the Royal Society of Edinburgh. The son of the keeper of a Scottish pub, he was just finishing his chemistry Ph.D. in Edinburgh. He sent one of his models to Kekulé in Ghent with a request to work in his laboratory. He was accepted and became great friends with Körner, another assistant, as well as an admirer of Kekulé. He would go on to professorships at Cambridge and the Royal Institution and to introduce the Dewar flask (or "thermos") for his studies of liquified gases. The structure in the bottom right of his figure is now called "Dewar benzene".

---

On the Oxidation of Phenyl Alcohol, and a Mechanical Arrangement adapted to illustrate Structure in the Non-saturated Hydrocarbons.

By J. Dewar, Esq. (from Proceedings of the Royal Society of Edinburgh, VI, Session 1866-67, pp. 82-86)

- Kekulé's original and elegant speculation on the structure of benzol and its derivatives induced me to try the effect of oxidising agents on benzol, with the view of eliciting whether the carbon atoms would separate in the way theory pointed out. The carbon atoms in benzol may be suposed to be arranged in a closed chain, where the carbon affinities are bound two and one alternately. -

In connection with this subject, I bring before the Society a simple mechanical arrangement adapted to illustrate structure in the non-saturated hydrocarbons. This little device is the mechanical representative of Dr. C. Brown's well-known graphic notation. A series of narrow thin bars of brass of equal length are taken, and every two of the bars clamped in the centre by a nut, so as to admit of free motion the one on the other. Such a combination represents a single carbon atom with its four places of attachment. In order to make the combination look like an atom, a thin round disc of blackened brass can be placed under the central nut. At the ends of the arms are holes to connect one carbon atom with another by means of a nut. The filling up of the places of attachment may be effected by slipping on the arms round discs of brass having a groove attached, and placing the symbol of the chemical element on the round projection. A carbon atom would then look like the following diagram.

As it is only intended to express the number of places of attachment along with the arrangement, when a given number of carbon atoms are combined in different ways, it is better to dispense with the symbols, rememberring that every free arm represents a place of attachment. When a number of carbon atoms are joined together, all the joints and arms being moveable, it is easy to show saturation in a closed or open chain, and the many arrangements of the atoms corresponding to the same formula. Although the bars are of equal length they are not intended to represent equal forces. We have no unit for comparing the values of the unit affinities in different atoms, and they may be incommensurable. In filling up the places of attachment we saturate the fixing powers of the individual atoms, taking no account of the influence the different elements and radicals have in affecting the energy of the whole compound.