The reaction is facilitated by electron-releasing substituents (the inductive effect) and in general the observed order (with decreasing reactivity) is tert-butyl > isopropyl > ethyl > methyl.
In 1935 Baker and Nathan explained the observed difference in terms of a conjugation effect and in later years after the advent of hyperconjugation (1939) as its predecessor.
A fundamental problem with the effect is that differences in the observed order are relatively small and therefore difficult to measure accurately.
Numerous anomalous physical measurements, including bond lengths and dipole moments, have been examined through this concept.
The original formulation of the Baker-Nathan effect is no longer employed due to more logical reasons for rate accelerations in solutions and its historical context is discussed by Saltzman.