## Abstract

In continuation of work reported earlier (Kuhn & Series 1950), the weak satellite $\alpha $-line in the 2P$_{\frac{3}{2}}$-3S$_{\frac{1}{2}}$ 6561 angstrom of the Balmer series of heavy hydrogen has been resolved. The light was produced as before in a U-tube cooled by liquid hydrogen, and analyzed by a double Fabry-Perot interferometer with spacings in the ratio 1:2. The satellite was found to be at a distance 0$\cdot $1326 $\pm $ 0$\cdot $001 cm.$^{-1}$ from the neighbouring strong component 2P$_{\frac{3}{2}}$-3D$_{\frac{5}{2}}$, in disagreement with the distance 0$\cdot $1409 cm.$^{-1}$ predicted by the Dirac relativistic theory of the hydrogen atom. The difference 0$\cdot $0083$_{-0\cdot 003}^{+0\cdot 002}$ cm.$^{-1}$ (where the limits also include possible errors in certain corrections) is interpreted as an upward shift of the 3S term, and is compared with the value 0$\cdot $0104 cm.$^{-1}$ predicted by the radiation theory of Bethe and others (see, for example, Bethe, Brown & Stehn 1950).