And keep in mind that changing values of a process template will only affect project areas that are subsequently instantiated from that process template, but will have have no effect on existing database objects, or on process areas that have already been instantiated from that process template.
In particular, to my knowledge, you can only archive existing iterations ... you cannot delete them from the database.   I assume this is to ensure that database objects that still refer to that iteration do not end up with a dangling pointer.  

Thanks for your fast reply. This could help. I have only some additional questions:

1. Do I have to delete the iterations in the process template always again if I update the template based on our template project?

2. Can I just deleted the lines for the iterations like shown below or are there any dependencies?

            <iteration has-deliverable="true" id="Release SW_00.01" name="Release SW_00.0X">
                <iteration has-deliverable="true" id="Release SW_00.01" name="Release SW_00.01"/>
                <iteration has-deliverable="true" id="Release SW_00.02" name="Release SW_00.02"/>
                <iteration has-deliverable="true" id="Release SW_00.03" name="Release SW_00.03"/>
                <iteration has-deliverable="true" id="Release SW_00.01 (1)" name="Release SW_00.01"/>
                <iteration has-deliverable="true" id="Release SW_00.02 (1)" name="Release SW_00.02"/>
                <iteration has-deliverable="true" id="Release SW_00.03 (1)" name="Release SW_00.03"/>

3. What dose the flag "has-deliverable" mean? Is it related to the releases?

 Hello Daniel,

Hello Eric,

I will try to explain Q1 more specific.
As I understood the process template is created based on a reference project. In our case the reference project it is called "Project_template". In RTC of this Project_template we created iterations by mistake that I do not want to have in new created projects. As you mentioned before this is not possible directly in the reference project. So I will delete it in the process template. The problem I see is that the iterations are still available in our reference project. So if we have to update the configuration of the reference project by any reason then we also have to update the process configuration to get the changes of the reference project. In this case I expect the iterations from the reference project will be again in the process template and I have to delete them again. Am I right or is my understanding of the process template wrong?

Daniel

The only connection between contents of the process template and the contents of the project area it was created from is that the process template is initialized from the contents of that project area.   After the process template is created, there is no longer any connection between the contents of the process template and the contents of that project area.  So in particular, adding and archiving iterations in that project area has no effect on the iterations in the process template, and conversely, adding or removing iterations in a process template has no effect on any existing project area (including the project area from which the template was created, and any project area that was initialized by that template).