Analytic value of the atomic three-electron correlation integral with Slater wave functions.

while in Eq. ~42! of Ref. @1#, line 8 should be (w12w2 1w3) instead of (w12w21w3) . With these two changes, both equations are now invariant under interchange of the nonlinear parameters w1 and w3 , as is required by the symmetry of the electron interaction operator. Harris @2# has shown that the formulas of Remiddi are equivalent to the older, more general formulas of Fromm and Hill @3# in their common region of applicability @Remiddi’s Eq. ~33! only#. However, he failed to point out the misprints. We provide further verification of the correctness of the modified Eq. ~42! of Ref. @1# for several values of the nonlinear parameters with a short table comparing 30 significant digit results using Remiddi’s formula ~with the above corrections! and the output of our recently developed triangle integral package ~details to be published!. We calculate, for various values of w1 , w2 , and w3 ,