If you have a large class, it may be useful to split the definition between multiple source files.
In addition, if you want to extend the class that is generated — say, from a T4 or ORM template — or in any situation where you cannot or should not modify the original source file, you can use the additional definition of the partial class to add new functions that will not be deleted when the template or ORM restores the original class.
Edit
Just an observation, but your quote:
when working on large projects, spreading the class to separate files allows you to work simultaneously with several programmers.
Not very good with me, even though I came from MSDN. If you use a decent version control system (for example, TFS or even svn), merging in the source files, where each developer works on individual methods, handles pretty well. Saying that using partial classes simplifies the work of several developers on the same class (and probably I will bet), encouraged someone there to separate each method and property of each class in their decision into separate source files. This is bad. Really bad. As in case of need - to open-forward-forward-when-negotiations-salaries-with-new-hires badly.
David Lively
source share