The Gaps That Proved the Pattern: Mendeleev and the Periodic Table

A card game with the elements

By the 1860s chemists knew of just over sixty elements and a great deal about each, but the collection felt like a junk drawer. There were hints of order. Some elements behaved like families, the reactive metals lithium, sodium, and potassium on one hand, the choking greenish gases fluorine and chlorine on another, but no one had found the rule that organized everything at once. Dmitri Mendeleev, a chemistry professor in Saint Petersburg, attacked the problem in 1869 in a famously hands-on way. He wrote the properties of each element on a card and shuffled them, looking for an arrangement that made sense, the way you might sort a hand of playing cards by suit and rank at the same time.

The arrangement he found was to lay the elements out in order of increasing atomic weight and then start a new row whenever the chemical properties began to repeat. Stacked this way, elements with similar behavior fell into the same vertical columns. The pattern recurred at regular intervals, which is why he called it periodic. Others were circling the same idea around then, notably the German chemist Lothar Meyer, but Mendeleev did something bolder than merely tabulate what was known.

Leaving holes on purpose

The boldness was in the gaps. When Mendeleev laid out his table, the periodic pattern did not line up neatly unless he left some squares empty. Rather than force the known elements together, he declared that the empty squares belonged to elements that existed but had not yet been discovered. This was a risky claim. A blank space in a table is easy to read as a mistake in the table. Mendeleev insisted instead that the blanks were predictions.

He went further still. Using the properties of the neighbors above, below, and beside each empty square, he described in detail what the missing elements should be like before anyone had laid eyes on them. He gave provisional names with the prefix eka, Sanskrit for one, so the undiscovered element below aluminium was eka-aluminium and the one below silicon was eka-silicon. For each he predicted atomic weight, density, the formula and properties of its oxide, how it would react. These were not vague guesses. They were numbers.

The predictions come true

The test came within Mendeleev’s lifetime, and it is the reason his table won out over rival schemes. In 1875 a French chemist isolated a new metal he named gallium. Its properties matched eka-aluminium closely, so closely that Mendeleev wrote to point out that the density the discoverer first reported was wrong and should be remeasured. It was remeasured, and Mendeleev’s predicted value was the correct one. He had known the density of an element better than the man holding the sample.

The most quoted case is eka-silicon, discovered in 1886 and named germanium. Set the 1871 prediction beside the measured reality:

Property	Predicted (eka-silicon, 1871)	Germanium (found 1886)
atomic weight	about 72	72.6
density (g/cm³)	about 5.5	5.35
oxide formula	EsO₂	GeO₂
oxide density	about 4.7	4.23
appearance	gray metal	gray-white metal

Scandium, filling the eka-boron slot, turned up in 1879. Three predicted elements, three discoveries, all matching. A table that merely organized the known elements would have been useful. A table that correctly described elements no one had seen was something else: evidence that the periodic pattern was real, a fact about nature rather than a filing convenience.

What the table did not yet know

Mendeleev had the pattern right but not its cause, and there were cracks he could not explain. A few elements sat in the wrong order if you went strictly by atomic weight; tellurium is heavier than iodine, yet their chemistry demands that tellurium come first. Mendeleev fudged these cases, trusting chemical behavior over the weights, and he turned out to be justified for a reason he could not have known. In 1913 Henry Moseley, using X-rays, showed that the true ordering principle is not atomic weight at all but atomic number, the count of protons in the nucleus. Reordered by atomic number, every anomaly vanished, tellurium and iodine included.

The deeper why came with quantum mechanics in the following decades. The periodic repetition of chemical properties reflects the way electrons fill up shells around the nucleus, with each row of the table corresponding to a shell being completed. Elements in the same column have similar arrangements of their outermost electrons, which is what makes them behave alike. Mendeleev’s families were quantum mechanics showing through, fifty years before quantum mechanics existed.

That is the part I find worth remembering. Mendeleev built a tool that worked, and predicted with it, long before anyone understood the machinery underneath. The empty squares he refused to fill with excuses, and chose instead to fill with predictions, are still there in the table on every classroom wall, now occupied by gallium, scandium, germanium, and the rest. The pattern was trustworthy before it was explained.