By 1800 chemists knew from careful mass measurements that compounds always formed in fixed whole-number ratios — 8 g of oxygen combines with exactly 1 g of hydrogen to make water, never 7.5 g and never 8.5 g. John Dalton explained this with Dalton's atomic theory: matter is made of tiny indivisible particles called atoms; atoms of a given element are identical in mass and properties; atoms of different elements differ; and compounds form when atoms combine in simple whole-number ratios. His model of the atom itself was a solid, featureless sphere — like a billiard ball — with no internal structure. This was the first quantitative atomic model, and it successfully predicted the law of definite proportions and the law of multiple proportions. It could not, however, explain electricity, magnetism, or why atoms of the same element sometimes had different masses (isotopes, discovered much later).

J.J. Thomson worked with a cathode ray tube — a sealed glass tube with almost all the air pumped out and a high voltage applied between a cathode (−) and anode (+). A glowing beam, the cathode ray, streamed from cathode to anode. Thomson showed three critical things: (1) the beam bent toward a positively charged plate, so the ray carried negative charge; (2) the beam bent in a magnetic field the same way regardless of what gas was in the tube or what metal the cathode was made of, so these negative particles came from inside every atom; (3) by measuring the deflection, he calculated a charge-to-mass ratio about 1,800 times smaller than the lightest atom (hydrogen). Conclusion: atoms contain much smaller negative particles, later called electrons. Since atoms are neutral overall, Thomson proposed the plum pudding model — a diffuse ball of positive charge with tiny negative electrons embedded throughout, like raisins in a pudding.

To test the plum pudding model, Ernest Rutherford (with Geiger and Marsden) fired alpha particles — small, dense, positive helium nuclei — at a gold foil only a few atoms thick, and watched where they landed on a surrounding detector screen. If Thomson's model were correct, the diffuse positive 'pudding' should barely deflect the fast alpha particles at all — they should mostly pass straight through with tiny angle changes. What they actually saw: about 99% of alphas passed straight through, but roughly 1 in 8,000 bounced back at sharp angles, some almost straight backward. Rutherford famously said it was 'as if you fired a 15-inch shell at a piece of tissue paper and it came back and hit you.' The only explanation: nearly all the positive charge and mass of the atom must be concentrated in a very small, dense central nucleus, and the rest of the atom is mostly empty space with electrons somewhere out in that space. The nuclear atom was born.

Rutherford left a huge problem: if electrons just floated around a positive nucleus, classical physics predicted they should spiral inward and the atom should collapse in a fraction of a second. Also, when hydrogen gas is heated it emits only specific colors of light (a line spectrum: red 656 nm, blue-green 486 nm, violet 434 nm and 410 nm) — not a continuous rainbow. Niels Bohr proposed that electrons can only exist in specific allowed energy levels, or shells, labeled n = 1, 2, 3, … The electron does not radiate energy while in a shell (solving the collapse problem). When an electron drops from a higher shell to a lower one, it releases a photon whose energy exactly equals the gap between the levels — that's why hydrogen emits discrete colors instead of a smooth spectrum. The Bohr model of hydrogen — nucleus at center, one electron in n=1, can jump to n=2, n=3 by absorbing energy — quantitatively predicts all the visible hydrogen lines.

In the 1920s, de Broglie showed that electrons behave as waves as well as particles, and Heisenberg showed that you cannot simultaneously know an electron's exact position and momentum (uncertainty principle). Erwin Schrödinger combined these ideas into a wave equation whose solutions — wave functions — describe the probability of finding an electron at any point around the nucleus. In this quantum mechanical model, electrons do not travel in Bohr's neat circles. Instead they occupy three-dimensional regions called orbitals, and we visualize each orbital as an electron cloud — denser where the electron is more likely to be, fainter where it is less likely. The 1s orbital, for example, is a spherical cloud centered on the nucleus; there is no 'orbit path.' This model correctly predicts the spectra of multi-electron atoms and is the model chemists still use today.

Each model was the best explanation for the evidence available at that time, and every refinement came from a new experiment that the old model could not explain. Dalton's sphere explained mass ratios but not electricity. Thomson's cathode rays proved electrons exist, giving plum pudding. Rutherford's gold foil proved a dense nucleus, killing plum pudding. Bohr's fixed shells explained hydrogen's line spectrum but failed for multi-electron atoms. Schrödinger's wave equation gave probability clouds that work for every atom. Science is not a list of right and wrong answers — it is a chain of models, each replaced when new evidence demands it.