On electrostatic screening of ions in astrophysical plasmas

There has been some controversy over the expression for the so-called `interaction energy' due to screening of charged particles in a plasma. Even in the relatively simple case of weak screening, first discussed in the context of astrophysical plasmas by Salpeter (1954), there is disagreement. In particular, Shaviv and Shaviv (1996) have claimed recently that by not considering explicitly in his calculation the complete screening cloud, Salpeter obtained a result for the interaction energy between two nuclei separated by a distance $r$ which in the limit $r \to 0$ is only 2/3 the correct value. It appears that this claim has arisen from a fundamental misconception concerning the dynamics of the interaction. We rectify this misconception, and show that Salpeter's formula is indeed correct.