A thin deposit on a glass plate
In 1922, Stern and Gerlach sent silver atoms from a heated source through narrow slits and an uneven magnetic field. The atoms landed on glass, building up an extremely thin silver deposit. The plate collected their landing positions over an exposure, so traces with the field on and off could be compared.
One beam became two separated traces. Even a long exposure left so little silver that development was needed to make it clear. Within the sensitivity of this method, the authors found no undeflected component between the traces. The marks on the glass became evidence that the atom’s magnetic response took discrete values.
Read the evidence: M. Bauer (2023), English translation of Gerlach and Stern’s 1922 paperW. Gerlach and O. Stern (1922), Der experimentelle Nachweis der Richtungsquantelung im Magnetfeld, Z. Phys. 9, 349–352
How a magnet separates neutral atoms
A silver atom has no net electric charge, but it has a magnetic moment: a measure of its magnetic strength and direction. Think of an atom carrying a tiny magnet. Where the field is stronger on one side, different internal magnetic states experience different pushes, and their paths separate. A uniform field does not produce this separating force in the same approximation.
If the atomic moments pointed in arbitrary directions with continuously varying projections, the beam would spread into a broad band. Two separated traces instead supported discrete projections along the field direction. The important feature was the gap between allowed responses.
Seeing the gap required a narrow incoming beam. The original report also discussed asymmetry caused by slit alignment and broadening caused by the atoms’ different speeds. Slower atoms spent longer in the field and bent farther. Both the beam preparation and its speed distribution helped determine the final deposit.
Read the evidence: M. Bauer (2023), English translation of Gerlach and Stern’s 1922 paper
The traces came before the spin explanation
The experiment tested space quantization in the old quantum theory: only certain projections of an atom’s magnetic moment were allowed. Electron spin had yet to be proposed. Stern and Gerlach designed their apparatus around the theory they had, then obtained a result that would acquire a different explanation.
In the modern account, ground-state silver has zero total electronic orbital angular momentum. Its familiar splitting comes mainly from the outer electron’s spin, an intrinsic form of angular momentum with two possible outcomes along a chosen measurement direction. Treating the atom as an ideal two-state system also neglects or leaves unresolved the nuclear hyperfine structure.
Read the evidence: M. Bauer (2023), English translation of Gerlach and Stern’s 1922 paperSchmidt-Böcking et al. (2016), The Stern–Gerlach Experiment Revisited — historical research
Later filters asked a further question
Keep one output of the first analyser and send it through a second analyser turned to another direction. What happens to the property selected by the first measurement? Modern sequences of spin filters ask this additional question. They require further selection and detection stages; the 1922 deposition experiment made the original beam-splitting observation.
Read the evidence: M. Bauer (2023), English translation of Gerlach and Stern’s 1922 paper
From silver atoms to the proton moment
Stern’s 1943 Nobel Prize recognized the molecular-beam method and the discovery of the proton’s magnetic moment. The method continued to develop: send particles through known fields, then infer internal properties from their landing positions or the number reaching a detector.
The silver experiment tested whether the magnetic response could vary continuously. Later hydrogen-molecule work addressed a much smaller nuclear magnetic response and required additional treatment of molecular rotation and speed. Rabi added an oscillating field that could change internal states, opening the way to magnetic-resonance measurements.
Read the evidence: Otto Stern — official 1943 award factsI. I. Rabi, J. R. Zacharias, S. Millman and P. Kusch (1938), A New Method of Measuring Nuclear Magnetic Moment, Phys. Rev. 53, 318 — original scan