[molpro-user] Possible bug in the DKH module in molpro 2006.1

[Lorenzo Lodi]

There seem to be a problem in molpro 2006.2 if the variable DKHO (order 
of the Douglas-Kroll-Hess hamiltonian) is set, in which cases a 
non-relativistic calculation is done instead.
For example consider the input
 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
***, Water calculation

DKROLL=1
DKHO=2

r1 =    1.8096, r2 =    1.8096, theta =  104.52
geometry={O;
            H1,O,r1;
            H2,O,r2,H1,theta}

RHF
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

As it is I obtain -76.02677991, which is the non-relativistic energy!
Removing the line DKHO=2 runs the 2nd order calculation.
I tried to place the line in other places, but it didn't help.
Anyone can suggest a workaround for this?

Incidentally, there seem to be an inconsistency in the manual (second 
line of pag. 89 of the PDF version). It is stated that "[molpro 2006.1] 
provide[s] the DKH Hamiltonian up to any arbitrary order or decoupling"
but in fact it is stated a few lines later that the maximum order is 14.

Lorenzo Lodi