I ran a ADF test in gretl and in R.

(The regression was no constant, LNEG lagged once, d_LNEG lag1 and d_LNEG lag2)
In gretl and R, the results were for most purposes identical for all numbers for Z and d_LNEG lag 1

In gretl the results for d_LNEG lag2 were
                    coefficient   std. error   t-ratio   p-value
d_LNEG_2   -0.470623     0.126295     -3.726    0.2006
But in R the results were
z.diff.lag2      -0.47062      0.12629       -3.726    0.00053

Which is correct?

Charles W