MOPAC question

["Hr. Dr. S. Shapiro" <toukie # - at - # zui.unizh.ch>]

 Dear Colleagues;
 	I would like to do a MNDO optimisation for the molecule
 p-tolylacetylene; however, when I run MOPAC 6.0 using the Z-matrix
 for this molecule with the keywords "nointer ef precise polar", the
 calculation aborts because MOPAC rejects structures that appear to
 have too many atoms all in a straight line [e.g., the C(ar)-C#C-H atoms are
 aligned at 180 degrees].  I understand that MOPAC _is_ able to run this
 calculation if Cartesian coordinates are substituted for the MOPAC Z-matrix
 in the input.  However, I can't find any useful information about this
 option in the MOPAC 6.0 manual (though I confess that I _may_ have
 overlooked this information whilst going through my hard copy of the user's
 manual).
 	Therefore, I'd be grateful to anyone who can give me detailed advice about
 how to run a MOPAC calculation for p-tolylacetylene, with or without
 Cartesiabn coordinates as the input.  If Cartesian coordinates _must_ be
 used as the input, what special keywords do I need, and exactly how (for
 example) should this Cartesian coordinate input be written?
 	Thanks in advance to all responders.
 Sincerely,
 S. Shapiro
 toukie # - at - # zui.unizh.ch