Dear Mr. Webb,

thank you very much for your quick and accurate answer.

>You seem to be confused between atom selections and restraints.

Actually, I was thinking of fixing the position of atoms in space. To me, this means adding constraints. Since I learned, that Modeller features no constraints, my next idea was to add a upper-bond harmonic potential, i.e. restraints to the r(1),r(2),r(3)-components in space. Studying this mailinglist, I found out that in Modeller, deselection would be the appropiate way.

> I don't know where you got that link from;

My apologize, I thought this was an official mirror, sorry for that.

> I can't give you an example that does "exactly what has been requested" because you didn't give the exact problem.

Anyway, your example tackled "exactly what has been requested" ;-) I was just confused by the first redefinition of the member actually *initializing* the selection_status parameter. However, I got it, now: The status is initializes *before* selecting the residues selected in selection_segment ;-) *That* was my confusion...

Sincerly yours