A professor told me once that the chapter sizes (measured in number of pages) in a PhD dissertation should be coherent. That is, the number of pages in each chapter should be somehow close to each other (to the average). Note: This does not include the Introduction and Conclusion chapters.
Is this really an important factor when writing a dissertation?
It's just a general guideline, don't worry about it too much. Aside from the fact that there are many ways to lay out a thesis, there are many reasons this would be violated. I would venture that a better way to state this rule would be, "A chapter should be of medium length." If a given chapter is too short, it's likely an indicator that it can be combined with some other section. If a single chapter is too long, split it up into parts (experiment 1, exp 2, discussion, etc).
In any event, you should first check into how your advisor wants the thesis laid out. Consider asking your advisor to see some dissertations of other previous students to get a feel for how he likes to have things laid out; that'll be a good way to get a feel for how to lay out your own.
