On this page
Most students meet the periodic table as a finished chart to memorise. Mendeleev's version was different. It had empty boxes in it, and he was confident enough to say what would go there. That confidence is the most useful thing to learn from him, because it turns a table into a tool for reasoning. This post covers who he was, what he predicted, how the predictions held up and how you can make one yourself.
Who was Dmitri Mendeleev?
The 1911 Encyclopaedia Britannica entry on Mendeleev, written soon after his death, records that he lived from 1834 to 1907 and was born at Tobolsk in Siberia, the youngest of a family of seventeen. He studied in St Petersburg, worked for a time in Heidelberg, and became professor of chemistry at St Petersburg University. His textbook The Principles of Chemistry was written between 1868 and 1870. Later he directed the Bureau of Weights and Measures.
Other chemists had noticed patterns in atomic weights before him. The same entry explains what set Mendeleev apart: he treated periodicity "not merely as a system of classifying the elements" but as a "law of nature" that could predict new facts and expose errors in old ones.
Why did Mendeleev leave gaps in his table?
If elements with similar chemistry must fall in the same column, then sometimes the next known element by atomic weight does not fit. Mendeleev's answer was to skip a box rather than force the element into the wrong family.
The Britannica entry says that in 1871 he asserted the existence of three unknown elements and assigned them definite properties. He named each one after the known element above its gap: eka-boron, eka-aluminium and eka-silicon. The RSC explains eka-boron as meaning "like boron". The Royal Society of Chemistry's page on scandium notes that he predicted ten unknown elements in all, four of them in detail.
He also questioned some accepted atomic weights because they did not fit the pattern, and later measurements supported him.
How close were his predictions?
The Royal Society of Chemistry's element pages record what happened.
Eka-aluminium became gallium. Paul-Émile Lecoq de Boisbaudran found it in 1875 from a new violet spectral line in zinc ore. Mendeleev had forecast an atomic weight of about 68 and a density of 5.9 g/cm³. The RSC gallium page lists 69.723 and 5.91.
Eka-boron became scandium. Mendeleev saw a gap between calcium (40) and titanium (48) and predicted an oxide of formula X₂O₃. Lars Fredrik Nilson found the element in 1879, and its oxide is Sc₂O₃.
Eka-silicon became germanium. Clemens Winkler found it in 1886 in a new mineral from a silver mine near Freiberg. Mendeleev had predicted an atomic weight of about 71 and a density of about 5.5 g/cm³. The RSC germanium page lists 72.630 and 5.3234.
Worked example: predict germanium the way a table invites you to
Using modern values from the RSC, germanium's neighbours are silicon above (28.085), tin below (118.710), gallium to the left (69.723) and arsenic to the right (74.922). Average the four:
(28.085 + 118.710 + 69.723 + 74.922) / 4 = 291.440 / 4 ≈ 72.86
The real value is 72.630, so the estimate is about 0.3% high.
Now try density, with silicon 2.3296, tin 7.287, gallium 5.91 and arsenic 5.75 g/cm³:
(2.3296 + 7.287 + 5.91 + 5.75) / 4 = 21.2766 / 4 ≈ 5.32 g/cm³
The real value is 5.3234 g/cm³. This is a classroom exercise with modern data, not a copy of Mendeleev's own working, but it shows why his kind of reasoning worked.
Common mistakes to avoid:
Averaging only up and down. Silicon and tin alone give a density of about 4.8, well off. Using all four neighbours does much better here.
Expecting this to work everywhere. Trends are much less smooth in some parts of the table. Our guide to periodic trends and their exceptions covers the ones that appear in exams.
Forgetting the modern ordering. Today's table is ordered by atomic number, not atomic weight, so check which quantity a question is asking about.
If a trend makes no sense to you, revise the idea behind it, such as effective nuclear charge, before memorising more exceptions. The Fermi Concept Library covers every concept in chemistry, each with its own lesson, so you can fill that one gap and return to the table.
How was Mendeleev recognised?
According to the 1911 Britannica, the Royal Society awarded him the Davy Medal in 1882, jointly with Lothar Meyer, and the Copley Medal in 1905. Element 101 carries his name. The RSC mendelevium page records that seventeen atoms of it were first made at Berkeley in 1955.
What can a student learn from how Mendeleev worked?
Trust a pattern enough to test it. A rule you believe should let you predict something you have not looked up.
Leave a gap rather than force a fit. In your own notes, mark what you do not know instead of covering it.
Question data that breaks a strong pattern. Sometimes the measurement is wrong. Sometimes the pattern is. Either way you learn something.
For more on remembering the facts the table does not predict, see our post on learning inorganic chemistry without rote memory.
Try it now: cover one element in the p-block, predict its atomic mass and density from its four neighbours, then check the real values.
