“Mendeleev Saw the Table in a Dream”: The Story Has a Single Source, and It Is Not Mendeleev

Every Russian schoolchild is told that Mendeleev saw the periodic table in a dream. The story is short and handsome: he struggled for three days, fell asleep from sheer exhaustion, saw the finished table, woke and wrote it down on a scrap of paper — and only one correction in one place was needed afterwards. On this site I go through the cases where a dream genuinely gave someone something, and I am obliged to go through the most famous of them honestly. Let me say it at the outset: it does not survive checking — and there is something useful in it, only not where people look for it.

A single source — and it is not Mendeleev

Not one of Mendeleev’s papers, not his diaries, not his letters to friends contain a word about this dream. The story has exactly one source: Alexander Alexandrovich Inostrantsev (1843–1919), a geologist, professor at St Petersburg University and a colleague of Mendeleev’s. He retold it in his recollections, printed after Mendeleev’s death — Mendeleev died in 1907 and could not object. Inostrantsev liked telling the anecdote to his students, and it was from them that it entered wide circulation.

And note a detail that usually passes unremarked. The words “I see in a dream a table where all the elements fall into place as required; I woke up and immediately wrote it down on a scrap of paper” travel through books and articles as a direct quotation — in quotation marks, in the first person. But in Inostrantsev this is a retelling of someone else’s account, not a document, and nowhere does it exist in Mendeleev’s own hand. The first person came later: somebody rewrote the retelling as direct speech because it is more convincing that way. After that it gets cited as eyewitness testimony.

What Mendeleev himself said

“I have thought about it for maybe twenty years, and you think: he sat down and suddenly… there it was” — this remark survives in memoir form as his irritated reply to a question about how the discovery was made. Formally this too is second-hand, and I am not going to pass it off as a document. But it is at least consistent with everything the papers show about how he worked. The dream story is consistent with nothing.

What is known about 17 February 1869

Here the firm part begins. The first version of the system — the manuscript titled An Attempt at a System of Elements Based on Their Atomic Weight and Chemical Affinity — is dated 17 February 1869 in the Julian calendar, that is 1 March in ours. The drafts from that day survive, and Bonifatiy Kedrov went through them sheet by sheet in his book The Day of One Great Discovery (1958): in effect a reconstruction of a single day from the archival papers.

What it shows is ordinary work done by hand. Mendeleev was due to leave that day for Tver province to inspect co-operative cheese dairies — he held such a commission — and he postponed the trip. The first sketches of that day were made on the back of a business letter about that very trip, from A. I. Khodnev, secretary of the Free Economic Society: the man used whatever was to hand. Then come comparisons of groups, rearrangements, crossings-out. And on 6 March the paper was presented to the Russian Chemical Society — not by its author: N. A. Menshutkin read it on his behalf.

“One correction in one place”

The most vulnerable detail of the legend is precisely the one that gives it its shine: a finished table, seen whole, and then a single fix. In reality Mendeleev kept reworking the system for two more years — moving elements, revising atomic-weight estimates — and in 1871 published a substantially revised version, the one his famous predictions of undiscovered elements grew out of. Two years of rework and “one correction” are not the same thing.

How this looks from the dream-research side

This has been examined outside Russian too. In 2001 the journal Dreaming carried a note by George Baylor with the question in its title: what do we really know about Mendeleev’s dream (Dreaming, 11(2), 89–92). The conclusion is the same: no dream report exists, everything rests on a second-hand retelling, and on the archival evidence the system had been arrived at before the alleged dream. Baylor does leave an honest possibility open: a dream showing a better arrangement of what had already been found may well have occurred — later. That is plausible and it changes nothing. Refining something in a dream and receiving the solution from a dream are different things.

Why this myth does harm

It looks like a harmless schoolroom anecdote. But look at its shape: a genius, three days, a dream, the finished answer. There is not a single step in it that anyone can repeat. It says that solutions come to the chosen and come free, and everyone else can wait for their night. I see the consequences constantly: someone tries for a week, nothing “comes”, and he concludes he is not one of those people. The truth is more useful: twenty years of work and one day when it fell into place. Work can be started today; a lottery cannot be started at all.

What really is there at the threshold of sleep

Now the honest line in the other direction. That Mendeleev did not get the table out of a dream does not mean there is nothing to do with dreams. Sleep does not hand out answers — but it hands out material: links, images, unexpected juxtapositions the waking head does not build because it holds on to familiar logic. This is not mysticism, and it is exactly what the people I have written about separately worked with: Dalí caught images right at the threshold of falling asleep, Fellini wrote down and drew his dreams for thirty years. Neither of them ordered a result. Both made sure there was something to catch it with.

What of this can be trained

You cannot order a solution, but you can load the problem before sleep, and that is a separate skill: dream incubation. The question is put briefly and concretely and held in mind while falling asleep — and you must not demand an answer of yourself; demanding only gets in the way. The second and more important half: what is not written down does not exist as far as memory is concerned. Everyone dreams, every night, and dreams vanish within minutes of waking, which is why a dream journal is not a ritual but the condition without which there is simply nothing to check. Mendeleev, incidentally, always had paper within reach: he made that first sketch on whatever was lying on the desk.

Where to start

Start with the journal: write on waking, without getting up and without opening your phone. In two or three weeks your recurring dream signs will start to show in the entries — the places, people and situations that come back specifically to you. Add reality checks during the day and MILD before sleep, so that awareness crosses the threshold with you and not only images. From there lucid dreaming opens up — the state in which a question can be asked from inside the dream. And there you need a separate skill, holding the state, or everything breaks off at the first second of excitement.

How to walk this path with guidance

Mendeleev took twenty years, and no dream at all. You have an advantage he did not: dream recall, observation and awareness inside the dream are today assembled into a clear sequence you can work through step by step — without myths about sudden illumination and without waiting for a special night. If you would like to walk it with guidance, read about my method and apply.