Hello fellows, this is not a real question since I found a workaround but I would like to know wether this is a bug in PyMOL or just the result of the Python selection algorithm.
Well, the problem was arising by playing around with Mr. DeLanos example script to produce the contact surface of two protein monomers. Apparently, as long as one creates the independent chain objects from an untouched dimer object all the surfaces will be fine. But when you have created something with the same file before, lets say make a dimer object first and second the two monomers so the finally surfaces will look scrambled. The way to solve the problem was to use 'create' from two different loaded pdb files. The copy command is not appropiate since you cannot do a selection. My conclusion is that after you picked up residues by 'create' from a file and want repeat it with residues selected before it will result in scrambled surfaces. Greetings, Bartholomeus