Inconsistency in relation between sign of nonadiabatic coupling matrix and phase of wave function ?

[Takao Kobayashi]

Dear MOLPRO users,

I would like to calculate the nonadiabatic coupling matrix (NACM) element 
by the SA-MCSCF method. However, the sign of NACM element < (phi1) |
d(phi2)/dR >
does NOT seems to be consistent with the phase of SA-MC wave function 
(the sign of CI vectors of phi1 and phi2, which are printed out by NATORB,
CI. option). 
For example, in some cases the sign of NACM is CHANGED even if the wave
function phase 
is NOT CHANGED. How can I obtain the CONSISTENT information on the NACM
element and 
wave function phase?  

Thank you in advance.

Takao Kobayashi

************************************************************
                    Dr. Takao Kobayashi
             Mitsubishi Chemical Corporation,
       MCC-Group Science & Technology Research Center,
   1000 Kamoshida-cho, Aoba-ku, Yokohama 227-8502, Japan

           e-mail:  tkoba at rc.m-kagaku.co.jp
           Tel:     (+81) 45-963-3834
           Fax:     (+81) 45-963-3835
************************************************************