How to Read the Periodic Table

A periodic table is a graph disguised as a chart. Position carries information, the numbers in each square carry more, and none of it is labelled.

This page is about extracting what is actually printed there — starting with the square in front of you, then the row and column it sits in, then the parts of the table that are conventions rather than facts and can therefore differ between one printed table and the next.

Which number is which

A typical cell carries four things: a whole number, a one- or two-letter symbol, a name, and a decimal number. Layouts vary — some tables put the whole number at the top, some at the bottom, some add the electron configuration or a list of shell occupancies — so position on the square is not a reliable guide.

What is reliable:

• The whole number is the atomic number, the count of protons, which defines the element.
• The decimal number is the standard atomic weight, the average mass of the element's atoms in unified atomic mass units.

If a table rounds the atomic weight and both numbers look like integers, there is a test that never fails: the atomic number increases by exactly one at every step along a row. Nothing else in the cell does that.

The atomic weight is a decimal because most elements are mixtures of isotopes, and the tabulated figure averages over them. It is not the mass of any individual atom, and it is not the number of particles in a nucleus.

Two further things are worth noticing when they appear. Some atomic weights are printed in square brackets — technetium's [98], for instance — which marks an element with no stable isotope, where the figure quoted belongs to whichever of its isotopes survives longest instead of averaging over several. And a few are printed as ranges, because isotope ratios vary with the sample's origin.

Eleven symbols that do not match their names

Most symbols are the first letter or two of the English name. Eleven are not, and each preserves whichever language chemistry happened to be written in before English took over.

Ten of those are Latin. Tungsten's W is the exception, coming from the German Wolfram — and the element is still called wolfram in much of Europe, which is why the symbol and the English name have no letters in common at all. Mercury's hydrargyrum is Greek by way of Latin and means "water-silver", a fair description of the only metal that pools at room temperature.

That these eleven are almost all metals known since antiquity is not a coincidence. They had names in older languages because people had been working with them for thousands of years before anyone thought to give them symbols.

Case matters absolutely. Co is cobalt; CO is carbon monoxide. A capital second letter turns one element into two.

Three group-numbering systems, two of which contradict each other

Columns are numbered 1 to 18 across the top on any modern table. Older tables use Roman numerals with A and B suffixes, and here the history is genuinely messy.

Two incompatible conventions were in simultaneous use for decades. In the American system, the main-group elements on the left were IA and IIA, the transition metals were the B groups, and the right-hand main groups were IIIA to VIIIA. In the European system the A and B labels were applied the other way round for much of the table, so that the same Roman numeral pointed at a different column depending on which side of the Atlantic the book was printed.

IUPAC introduced the plain 1-to-18 numbering in 1988 specifically to end that ambiguity, and it is the only system that is unambiguous without knowing the book's provenance.

One useful thing survives from the old labels: in the A-group notation, the Roman numeral is the number of valence electrons. Group VIA has six. That correspondence is the reason the modern scheme requires subtracting ten for groups 13 to 18, and it is why older textbooks make the valence count look more obvious than newer ones do.

The staircase, and why nobody agrees where it ends

A stepped line descends the right-hand side of the table, starting between boron and carbon. Metals lie to its left, non-metals to its right, and the elements sitting on the line itself are called metalloids.

The commonly listed metalloids are boron, silicon, germanium, arsenic, antimony and tellurium. Beyond that, agreement breaks down. Polonium and astatine are included by some sources and not others. Aluminum touches the line on many printed tables and is a straightforward metal. Selenium is treated as a metalloid in a fair number of references and as a non-metal in most.

The reason for the disagreement is simple and rarely stated: there is no IUPAC definition of a metalloid. It is a descriptive convenience for elements whose properties are intermediate, and where you draw the boundary depends on which property you weight. An element can conduct like a semiconductor and still be brittle and lustrous like a metal.

Hydrogen is the other place the geography misleads. It sits at the top left, in the metals' corner, and it is a non-metal gas.

Colours are the publisher's opinion

The colour blocks that group elements into alkali metals, transition metals, halogens and so on are editorial choices, not standards. Different tables use different category sets, and the number of categories ranges from about six to about twelve depending on the publisher.

That is not a criticism — the categories describe real chemical families — but two tables that disagree about whether an element is a "post-transition metal" or a "poor metal" are not disagreeing about chemistry. They are using different vocabularies. Where a colour and a measurable property conflict, trust the property.

Mendeleev's real achievement was the gaps

Dmitri Mendeleev is credited with the periodic table, and he was not the only person in the 1860s arranging elements by increasing atomic weight and noticing that properties recurred. Julius Lothar Meyer produced a very similar arrangement independently and at almost the same time.

What separated Mendeleev was confidence in the pattern over the data. Where no known element fit a position, he left the square empty and predicted what would eventually fill it.

For the gap below silicon — he called it eka-silicon — Mendeleev predicted in 1871 an atomic weight near 72, a density around 5.5 g/cm³, an oxide of formula EsO₂ with density near 4.7, and a chloride boiling below 100 °C. Germanium was isolated in 1886 with an atomic weight of 72.6, a density of 5.35, GeO₂ at 4.70, and GeCl₄ boiling at 86 °C.

Gallium and scandium filled two more of his gaps within the same decade. Predictions that specific are what turned a classification scheme into a law of nature.

Four places where mass order is wrong

Mendeleev ordered by atomic weight, and in four places he deliberately broke his own rule because the chemistry demanded it. He put tellurium before iodine even though tellurium is heavier.

He was right, and he could not have said why. The explanation came in 1913, when Henry Moseley measured how each element's characteristic X-ray emission shifted along the series and showed that one integer — the nuclear charge — fixed its position, not the atomic weight. Ordering by atomic number rather than mass fixes all the anomalies at once.

Four pairs still sit in "wrong" mass order on every modern table: argon before potassium, cobalt before nickel, tellurium before iodine, and thorium before protactinium. Each is a case where the lighter element has more protons, because the heavier one happens to have neutron-rich isotopes dominating its natural abundance.

Moseley was killed at Gallipoli in 1915, aged 27, two years after the work that gave the periodic table its physical foundation.