# The Gaps He Left Empty > Dmitri Mendeleev sorted the known elements by weight, found that their properties repeat, and then did the dangerous thing: where the pattern demanded an element nobody had found, he left the square empty and published its predicted weight and density. Sort the cards, open the gaps, and check his 1871 numbers against the germanium found in 1886. Episode 06 of the Great Scientists series. Interactive film: https://www.planetpast.com/stories/scientist-mendeleev/ Publisher: PlanetPast (https://www.planetpast.com/) · Published 2026-10-13 · Chemistry, History ## Prologue ### The gaps he left empty *Sixty-three cards and three holes* Anyone can spot a pattern. He bet on where it was missing. By 1869 chemists knew about sixty-three elements, and several people had noticed that if you line them up by weight, their properties repeat. Mendeleev did something more dangerous. Where his pattern demanded an element that nobody had ever seen, he refused to close the gap. He left the hole open, wrote down what the missing element would weigh and how dense it would be, and published it. Within seventeen years, three of those holes were filled exactly as specified. ![One card per element: name, weight, how it behaves. Then he sorted them.](https://www.planetpast.com/stories/scientist-mendeleev/img/c00.jpg) - **~63**: elements known in 1869 - **3**: holes he left on purpose - **17**: years until all three were filled > I saw in a dream a table where all the elements fell into place as required. (Mendeleev, recalling the night of 1869) Prologue · Why he matters #### The gaps he left empty Others saw the repetition. Mendeleev **left holes** where the pattern needed elements nobody had found, and published their predicted weights and densities. **Gallium, scandium and germanium** filled them. - **Born**: Siberia, 1834 8 February, the youngest of a very large family - **Died**: St Petersburg, 1907 2 February - **The table**: 1869 written while making a textbook How to use: pick a chapter below or press ← → to change step · ▶ plays the step · drag to rotate · three steps have buttons you can press. Sources: [Wikipedia: Dmitri Mendeleev](https://en.wikipedia.org/wiki/Dmitri_Mendeleev); [Wikipedia: history of the periodic table](https://en.wikipedia.org/wiki/History_of_the_periodic_table) ## Chapter I · Sixty-odd elements ### A heap of facts *Chapter I · Before* Everybody had the ingredients. Nobody had the nerve. Chemistry in the 1860s was a enormous catalogue with no index. Sixty-three elements, each with its own behaviour, and nothing saying why. The pattern was half-visible. Döbereiner had found triads of similar elements. In 1865 John Newlands pointed out that properties repeat roughly every eighth element and called it the law of octaves. The Chemical Society refused to publish it, and a fellow of the society asked him publicly whether he had tried arranging the elements alphabetically. Lothar Meyer in Germany had a table very like Mendeleev’s at almost the same time. The difference was not the table. It was what Mendeleev was willing to say about things that did not exist yet. ![Sixty-three elements, each with its own list of reactions, and no reason why any of them should be related.](https://www.planetpast.com/stories/scientist-mendeleev/img/c01.jpg) - Döbereiner: elements come in triads - Newlands: properties repeat every eighth - He was asked if he had tried alphabetical order - Meyer had a similar table, published later - **1865**: Newlands’ law of octaves, laughed at Chapter I · Step 1 #### A heap of facts Döbereiner’s triads and **Newlands’ law of octaves** (1865) were close — and Newlands was mocked for them. **Lothar Meyer** had a near-identical table at nearly the same time. The difference was the predictions. Sources: [Wikipedia: John Newlands and the law of octaves](https://en.wikipedia.org/wiki/John_Newlands_(chemist)); [Wikipedia: history of the periodic table](https://en.wikipedia.org/wiki/History_of_the_periodic_table) ### Out of Siberia *Chapter I · The man* His mother spent her last years getting him into a university. Mendeleev was born in 1834 in a Siberian village, the youngest of a very large family. His father went blind and lost his job; his mother reopened a family glassworks to keep them fed, and then it burned down. In 1849 she took her youngest child and travelled across the country, first to Moscow, which would not admit a Siberian, and then to St Petersburg, which did. She died a few weeks later. He dedicated his first major work to her, and wrote that she had taught him to seek divine and scientific truth through patience and work. ![He cut his hair and beard once a year, in spring, and refused to do it for the Tsar.](https://www.planetpast.com/stories/scientist-mendeleev/img/s2-mendeleev.jpg) ![His mother took him thousands of kilometres to find a university that would take him.](https://www.planetpast.com/stories/scientist-mendeleev/img/c02.jpg) - 1834: Born in Siberia, the youngest of many - 1847: His father dies; the family glassworks burns down - 1849–50: His mother takes him to Moscow, then St Petersburg - 1850: Admitted to the Main Pedagogical Institute Chapter I · Step 2 #### Out of Siberia Born in a Siberian village in **1834**, the youngest of a very large family. After his father went blind and the family glassworks burned, his mother took him **across Russia** to find a university that would admit him. She died weeks after he was accepted. Sources: [Wikipedia: Dmitri Mendeleev](https://en.wikipedia.org/wiki/Dmitri_Mendeleev) ## Chapter II · The cards ### A textbook with a problem *Chapter II · 1869* He needed an order to teach them in. The table was not born in a flash of theory. It was born out of a teaching problem. Mendeleev was writing a two-volume textbook and had reached the point where he had to decide which element to describe after which. There was no principled order. So he made a card for each element with its weight and its usual compounds, and spent days laying them out on a table and moving them around. Press the buttons to do the same thing: sort them by weight in one long row, then wrap the row. ![He was writing Principles of Chemistry and needed an order to teach the elements in.](https://www.planetpast.com/stories/scientist-mendeleev/img/c03.jpg) - Write one card per element - Put its weight and its typical compounds on it - Lay them out and move them around - Look for an order that repeats - **1868–70**: writing Principles of Chemistry - **1869**: the table is drafted Chapter II · Step 3 · **Interactive** #### A textbook with a problem The periodic table came out of writing **Principles of Chemistry**: Mendeleev needed a principled order in which to teach the elements, so he made a card for each and sorted them. Press **Sort by weight** for one long line, then **Wrap the row**. Similar elements fall into columns. The cards here are the first thirty-odd elements with their 1869 weights. Sources: [Wikipedia: Dmitri Mendeleev](https://en.wikipedia.org/wiki/Dmitri_Mendeleev); [Wikipedia: history of the periodic table](https://en.wikipedia.org/wiki/History_of_the_periodic_table) ### Families fall into columns *Chapter II · The pattern* Sort by weight and the chemistry lines up sideways. Lay the cards out in order of increasing weight and something odd happens. Every so often you reach an element that behaves like one you have already passed. Lithium, sodium and potassium are all soft metals that attack water. Fluorine, chlorine and bromine are all savagely reactive. Wrap the row at the right length and those families stack into columns. That is the periodic law: the properties of the elements are a periodic function of their atomic weight. But only if you are willing to leave the row ragged where it needs to be. - Lithium, sodium, potassium: soft, violent in water - Fluorine, chlorine, bromine: biting, reactive gases and liquids - Each family lands in one column - That is what periodic means - **8**: elements between repeats, in the light part of the table ![When a card did not fit, he preferred to move everything else rather than force it.](https://www.planetpast.com/stories/scientist-mendeleev/img/c04.jpg) Chapter II · Step 4 #### Families fall into columns Ordered by weight, chemical families recur: **Li, Na, K** in one column, **F, Cl, Br** in another. That is the **periodic law** — properties are a periodic function of atomic weight. Sources: [Wikipedia: the periodic law](https://en.wikipedia.org/wiki/Periodic_law) ## Chapter III · The gaps ### Leaving the hole open *Chapter III · 1871* The brave move is refusing to fill it. Here is where he differs from everyone else who saw the pattern. Between zinc and arsenic the columns demand two elements that did not exist in any laboratory. The easy fix is to shuffle the known elements along and close the gap. Mendeleev refused. He left the squares empty, gave the missing elements provisional names — eka-aluminium, eka-boron, eka-silicon — and then went further. He said what each one would weigh, how dense it would be, what oxide it would form, how its chloride would behave. Press a gap and see the prediction. ![Between zinc and arsenic the pattern needs two elements that nobody had ever isolated.](https://www.planetpast.com/stories/scientist-mendeleev/img/c05.jpg) - Zinc, then a hole, then another hole, then arsenic - He named them eka-aluminium and eka-silicon - Eka is Sanskrit for one — “one past aluminium” - And he said what they would weigh - **3**: gaps left open in 1871 Chapter III · Step 5 · **Interactive** #### Leaving the hole open Where the columns demanded elements nobody had found, Mendeleev left the squares **empty** and named the absentees **eka-aluminium, eka-boron and eka-silicon** — then specified their weights and densities. Press **Show the gaps**, then pick one to see what he said would be found there. Sources: [Wikipedia: Mendeleev’s predicted elements](https://en.wikipedia.org/wiki/Mendeleev%27s_predicted_elements) ### What a prediction looks like *Chapter III · The numbers* Not “something will be found”. Four decimal numbers. This is what makes it a scientific prediction rather than a hunch. He did not say an element would turn up. He said this element will have these values. For eka-silicon in 1871: atomic mass about seventy-two point six. Density five point five. A refractory dioxide of density four point seven. A chloride boiling below one hundred degrees. Fifteen years later Clemens Winkler isolated germanium and measured seventy-two point six three, five point three five, four point seven, and eighty-six degrees. Press reveal and compare them yourself. - **72.64**: predicted atomic mass - **72.63**: germanium, measured - Density predicted 5.5, measured 5.35 - Oxide density predicted 4.7, measured 4.7 - Chloride boiling point predicted under 100 °C, measured 86 °C - Chloride density predicted 1.9, measured 1.9 - **15**: years between the prediction and the element Chapter III · Step 6 · **Interactive** #### What a prediction looks like Eka-silicon, 1871: mass **72.64**, density **5.5**, oxide density **4.7**, chloride boiling **under 100 °C**. Germanium, 1886: **72.63**, **5.35**, **4.7**, **86 °C**. Press **Predict** then **Reveal** to put the two columns side by side. Sources: [Wikipedia: germanium](https://en.wikipedia.org/wiki/Germanium); [Wikipedia: Mendeleev’s predicted elements](https://en.wikipedia.org/wiki/Mendeleev%27s_predicted_elements) ## Chapter IV · The proof arrives ### The table corrects the chemist *Chapter IV · 1875* Mendeleev told the discoverer his measurement was wrong. It was. In 1875 Paul-Émile Lecoq de Boisbaudran found a new element by its spectrum and named it gallium. Mendeleev read the paper and wrote to say that the density reported must be wrong: his table said eka-aluminium should be about five point nine, not four point seven. Lecoq, who had the only sample in the world, repurified it and measured again. Five point nine. That is the moment the periodic table stopped being a classification scheme and became a theory that could overrule a measurement. ![A metal that melts in your hand, found by its spectral lines in 1875.](https://www.planetpast.com/stories/scientist-mendeleev/img/s7-gallium.jpg) - 1875: Lecoq de Boisbaudran finds gallium - 1879: Nilson finds scandium - 1886: Winkler finds germanium - Lecoq measured gallium’s density as 4.7 - Mendeleev wrote from Russia: it should be 5.9 - Lecoq purified his sample and remeasured: 5.9 - The table corrected the experiment Chapter IV · Step 7 #### The table corrects the chemist Mendeleev told gallium’s discoverer that his measured density was wrong. **Lecoq repurified his sample and the table was right.** Scandium followed in 1879, germanium in 1886. Sources: [Wikipedia: gallium](https://en.wikipedia.org/wiki/Gallium); [Wikipedia: scandium](https://en.wikipedia.org/wiki/Scandium); [Wikipedia: Mendeleev’s predicted elements](https://en.wikipedia.org/wiki/Mendeleev%27s_predicted_elements) ### Three for three *Chapter IV · 1886* Different chemists, different countries, the same empty squares. Clemens Winkler pulled a new element out of a silver ore in 1886 and called it germanium. Its properties matched eka-silicon almost line for line. Three predictions, three hits, by three chemists in three countries, none of whom set out to prove a Russian right. After that the table was no longer one chemist’s filing system. It was the shape of matter. Germanium itself had a second life: the first working transistor, in 1947, was made of it. ![A brittle grey semiconductor. Sixty years later it became the first transistor.](https://www.planetpast.com/stories/scientist-mendeleev/img/c07.jpg) - **1886**: Winkler isolates germanium - **1947**: the first transistor is made of it - Three predictions, three hits - Each by a different chemist in a different country - None of them were looking for Mendeleev’s element Chapter IV · Step 8 #### Three for three Germanium, isolated by **Winkler in 1886**, matched eka-silicon almost line for line. Three independent discoveries filled three deliberately empty squares — and germanium later became the **first transistor**. Sources: [Wikipedia: germanium](https://en.wikipedia.org/wiki/Germanium); [Wikipedia: Mendeleev’s predicted elements](https://en.wikipedia.org/wiki/Mendeleev%27s_predicted_elements) ## Chapter V · Where it broke ### Right answer, wrong reason *Chapter V · The anomaly* He swapped two elements and blamed the scales. The law had a defect he could not explain. Tellurium weighs more than iodine, but tellurium belongs in the sulfur column and iodine in the chlorine column. Ordering strictly by weight puts them both in the wrong family. Mendeleev put them in the chemically right order anyway, and announced that the measured atomic weights must be in error. They were not. The weights were fine. In 1913 Henry Moseley showed that the real ordering principle is the positive charge of the nucleus — the atomic number. By that measure iodine does come after tellurium, and the anomaly disappears. Mendeleev was right to swap them, and wrong about why. - Tellurium weighs more than iodine - But tellurium behaves like sulfur, and iodine like chlorine - So he put them in the “wrong” order on purpose - And said the measured weights must be wrong - **1913**: Moseley: order by atomic number, not weight ![Three pairs in the table sit in the wrong weight order. All three are right by atomic number.](https://www.planetpast.com/stories/scientist-mendeleev/img/c08.jpg) Chapter V · Step 9 #### Right answer, wrong reason Tellurium is heavier than iodine but belongs before it chemically. Mendeleev swapped them and blamed the measurements. **Moseley (1913)** showed the true order is by **atomic number**, which puts them in exactly the order Mendeleev had insisted on. Sources: [Wikipedia: Henry Moseley and atomic number](https://en.wikipedia.org/wiki/Henry_Moseley); [Wikipedia: the periodic law](https://en.wikipedia.org/wiki/Periodic_law) ### A whole missing column *Chapter V · 1894* Six elements that react with nothing at all. Then came a harder test. In 1894 Rayleigh and Ramsay isolated argon, a gas that will not combine with anything. A table organised by how elements combine has no slot for an element that does not. Mendeleev’s first reaction was to doubt that argon was an element at all. Within four years there were five more of them. Rather than break the table, they fitted into a new column at its edge, between the violently reactive halogens and the violently reactive alkali metals. A classification that can absorb six unexpected elements without being rebuilt is telling you about reality, not about filing. - 1894: Argon isolated — it fits nowhere - 1895–98: Helium, neon, krypton, xenon follow - 1900: A new group is added to the table - They react with nothing - So they had no place in a table built on combining power - Mendeleev first doubted argon was an element - The table absorbed them by growing a column Chapter V · Step 10 #### A whole missing column **Argon (1894)** reacts with nothing and had no place in a table built on combining power. Five more noble gases followed — and the table accommodated them by **growing a new group** rather than breaking. Sources: [Wikipedia: the noble gases](https://en.wikipedia.org/wiki/Noble_gas); [Wikipedia: history of the periodic table](https://en.wikipedia.org/wiki/History_of_the_periodic_table) ## Chapter VI · The rest of him ### Balloons, oil and one lost vote *Chapter VI · The rest* He spent more of his life on Russian industry than on the table. The table took a few years of a long and busy life. He spent far more of it on Russian industry: oil refining, coal, agriculture, tariffs, and running the national bureau of weights and measures. In 1887 he went up alone in a hydrogen balloon to observe a solar eclipse, having pushed the pilot out to save weight. In 1906 he was nominated for the Nobel Prize in Chemistry and lost by a single vote. He died the following year. And the story that he defined vodka at forty per cent is false. That standard was set by the Russian government in 1843, when he was nine years old. - Director of the Bureau of Weights and Measures - Worked on Russian oil, coal and agriculture - Flew in a balloon alone to watch an eclipse - Lost the 1906 Nobel Prize by one vote - **40%**: vodka: not his doing - **1907**: he dies of influenza, aged 72 - **101**: the element later named after him Chapter VI · Step 11 #### Balloons, oil and one lost vote Mendeleev spent most of his career on Russian industry and metrology, flew a balloon alone to see an eclipse in 1887, and **lost the 1906 Nobel Prize by one vote**. Element **101, mendelevium**, is named for him. The famous claim that he set vodka’s strength at 40% is **false**: the standard dates from 1843, when he was nine. Sources: [Wikipedia: Dmitri Mendeleev](https://en.wikipedia.org/wiki/Dmitri_Mendeleev) ## Epilogue ### The pattern before the reason *Epilogue* Quantum mechanics explained the table forty years later. Mendeleev had no idea why his pattern existed. Atoms were still controversial; electrons were thirty years away. The reason is that chemical behaviour is set by the outermost electrons, and the number of those cycles as you add electrons shell by shell. The columns are electron configurations. That explanation arrived in the 1920s. The table had already been working for fifty years. It is a useful thing to remember about science: a reliable pattern is worth having even when nobody can yet say why it holds. - Columns repeat because outer electrons repeat - Which Mendeleev could not have known - The pattern was visible before its cause - That is how most science actually goes - **118**: elements in the table today - **94**: of them occur naturally ![Every semiconductor, catalyst and battery chemistry is chosen by position in this grid.](https://www.planetpast.com/stories/scientist-mendeleev/img/c09.jpg) Epilogue · Step 12 #### The pattern before the reason The columns repeat because **outer-electron configurations** repeat — an explanation that arrived in the **1920s**, fifty years after the table was already working. Sources: [Wikipedia: the periodic law](https://en.wikipedia.org/wiki/Periodic_law); [Wikipedia: history of the periodic table](https://en.wikipedia.org/wiki/History_of_the_periodic_table) ### What the empty squares were for *Epilogue* The gaps are the reason we remember him and not the others. Several people arranged the elements in 1869. We remember Mendeleev because he arranged them in a way that could be wrong. The empty squares were not an admission of ignorance. They were a set of experiments he was asking other people to run, with the answers written down in advance. Three of them came back right, from three different laboratories, within seventeen years. The chapter stills in this film are AI-generated. The photograph of Mendeleev and the sample photographs are real. Sources are listed below. - 1869: the cards are sorted - 1871: three gaps, with numbers attached - 1875 · 1879 · 1886: gallium, scandium, germanium - 1913: atomic number replaces weight - 1920s: electrons explain the columns - A classification you cannot test is just filing - He made his testable by leaving it incomplete - The empty squares were the experiment Epilogue · Step 13 #### What the empty squares were for Several chemists arranged the elements in 1869. Mendeleev is remembered because he arranged them in a way that **could have been proved wrong** — and then wasn’t. - 1869 the cards sorted into a periodic table - 1871 three gaps with predicted properties - 1875–1886 gallium, scandium, germanium - 1913 atomic number replaces atomic weight Chapter stills in this film are AI-generated illustrations. The photograph of Mendeleev and the photograph of gallium are real, credited where they appear. Sources: [Wikipedia: Dmitri Mendeleev](https://en.wikipedia.org/wiki/Dmitri_Mendeleev); [Wikipedia: history of the periodic table](https://en.wikipedia.org/wiki/History_of_the_periodic_table); [Wikipedia: Mendeleev’s predicted elements](https://en.wikipedia.org/wiki/Mendeleev%27s_predicted_elements); [Wikipedia: germanium](https://en.wikipedia.org/wiki/Germanium); [Wikipedia: gallium](https://en.wikipedia.org/wiki/Gallium); [Wikipedia: scandium](https://en.wikipedia.org/wiki/Scandium); [Wikipedia: Henry Moseley and atomic number](https://en.wikipedia.org/wiki/Henry_Moseley); [Wikipedia: the noble gases](https://en.wikipedia.org/wiki/Noble_gas); [Wikipedia: the periodic law](https://en.wikipedia.org/wiki/Periodic_law); [Wikipedia: John Newlands and the law of octaves](https://en.wikipedia.org/wiki/John_Newlands_(chemist))