Neutral molecules can carry information about nuclei
The 1922 silver-beam splitting and the 1933 proton-moment work asked different questions. Silver exposed discrete magnetic deflection patterns. Estermann, Frisch and Stern used a neutral hydrogen-molecule beam to reach a much smaller nuclear magnetic scale.
Neutrality is useful because the beam is not simply bent by the Lorentz force on a net charge. But a molecule is not an isolated proton. Connecting its response to a proton moment requires accounting for the molecule’s magnetic contributions and internal states, rather than relabeling its total deflection as one proton’s signal.
Read the evidence: O. Stern (1946), The Method of Molecular Rays — Nobel lecture, December 12, for the 1943 prizeI. Estermann, R. Frisch and O. Stern (1933), Magnetic Moment of the Proton, Nature 132, 169–170 — publication record
Use the nuclear scale before comparing sizes
The nuclear magneton below uses the positive elementary charge and proton mass. The early moment quoted by Stern was roughly two and a half of these units, rather than the single unit expected from the simple Dirac-particle comparison. It was a substantial discrepancy, not merely a tiny adjustment.
A moment expressed in Bohr magnetons would be numerically much smaller because the electron mass defines that unit. Changing units does not remove the discrepancy. One must distinguish the natural scale, the measured dimensionless ratio and the theoretical assumptions used to predict it.
Read the evidence: O. Stern (1946), The Method of Molecular Rays — Nobel lecture, December 12, for the 1943 prize
Longer interaction turns a small force into a displacement
In his lecture, Stern emphasizes the time a ground-state molecule can spend in a magnetic field, contrasting it with the brief lifetime of an optically excited atom. A weak magnetic response becomes accessible through a beam’s accumulated displacement. The field gradient, not a uniform field alone, supplies the deflecting force.
A measured profile still mixes velocities, apertures and internal-state populations. Slower molecules experience the field longer, so the same magnetic response does not give every molecule the same shift. Field-off reference profiles and a characterized field gradient are essential parts of an inference, not decorative apparatus details.
Read the evidence: O. Stern (1946), The Method of Molecular Rays — Nobel lecture, December 12, for the 1943 prize
The early anomaly does not already reveal quarks
The result challenged a simple point-particle expectation for the proton. It did not image a nuclear interior or determine a unique microscopic composition. Many proposed explanations can fail the same numerical comparison; rejecting one expectation does not automatically identify the replacement.
The electron anomaly and proton moment also differ greatly in scale and interpretation. A small radiative correction to an electron baseline should not be copied as an explanation of a proton moment several nuclear magnetons large. The modern composite-proton account belongs to additional theoretical and experimental evidence.
Read the evidence: O. Stern (1946), The Method of Molecular Rays — Nobel lecture, December 12, for the 1943 prizeX. Fan, T. G. Myers, B. A. D. Sukra and G. Gabrielse (2023), Measurement of the Electron Magnetic Moment, Phys. Rev. Lett. 130, 071801
A single trapped proton is a different measurement
DiSciacca and Gabrielse reported a direct single-proton Penning-trap measurement in 2012. Their abstract gives a moment near 2.793 nuclear magnetons and a parts-per-million uncertainty. This is a dated example of a later method, not the precision attained by the 1933 molecular beam.
A trap replaces a passing molecular ensemble with a stored charged particle and compares motional and spin-sensitive observables. That methodological change reduces one kind of inference while creating other calibration tasks. Improved agreement must be credited to the actual apparatus and correction model used.
Read the evidence: J. DiSciacca and G. Gabrielse (2012), Direct Measurement of the Proton Magnetic Moment — author abstract, arXiv:1201.3038v2
Keep the two Stern stories connected and distinct
The linked spin-filter Lab is useful for discussing outcome branches. It does not model hydrogen-molecule deflection, molecular rotation or extraction of the proton moment. The effective coupled-spin model likewise does not infer that moment from a historical beam profile.
Read the silver-beam article for directional quantization and this article for the nuclear scale. Together they explain why the 1943 award mentions both the molecular-ray method and the proton moment. They do not make every later nuclear magnetic measurement part of one already completed experiment.
Read the evidence: Otto Stern — official 1943 award factsJ. DiSciacca and G. Gabrielse (2012), Direct Measurement of the Proton Magnetic Moment — author abstract, arXiv:1201.3038v2