In 1978, physicists bombarded a target with a beam of ions and measured how quickly they slowed down in matter. Using the Doppler shift of gamma rays, they calculated the lifetimes of excited nuclear states. For iron-58, they obtained a value that did not agree with shell-model calculations.
Now, in a Coulomb excitation experiment, beams of iron-56 and iron-58 nuclei passed near a heavy target. The target's electric field gently jostled the nuclei, lifting them to higher energy states. By measuring the gamma radiation emitted as they returned to the ground state, physicists determined the transition probability between levels. For iron-56, the result matched previous data, but for iron-58 it was noticeably lower.
Re-evaluating the old data with modern stopping values resolves the discrepancy. The iron-58 nucleus, which seemed like a capricious exception, now follows the predictions of the nuclear shell model precisely. This is reminiscent of a situation where an old recipe had a mistake in the amount of salt, making the dish taste off — once corrected, everything falls into place.
The stopping power of matter for ions depends on their velocity somewhat like air resistance for a car: at low speeds the loss is linear, and at high speeds it's quadratic, but for ions the picture is more complex due to charge exchange.