This makes use of all the nonzero coefficients in the nth row starting with (~) and ending with
(nn). Now when n < 0, there are two sets of nonzero coefficients in the nth row, namely those (n)
with 0 < r < ~, and those with -~ < r < n. This suggests that for n < 0, the expression for
(11) should perhaps be interpreted as