“…Indeed (see figure l), the node in the function uak,02(r12) occurs only for r12 = 48 au. In contrast, we find (see table 2 ) that the choice of a nodeless correlation function with k = 0, recommended by previous authors (Walsh andBorowitz 1959, Aubert-Frecon et a1 1981) leads to results that are less accurate compared either with experiment or (see table 2 ) the calculations of Jeung (1983Jeung ( , 1987. We should note however that when k is varied from 0 to 0.3, which is a wide range, the energies vary by less than lo%, the change in the binding energies of the upper states being less than 500 cm-I.…”