I think an awful lot depends on your own motivation for doing the PhD. For me, it was a no-brainer: I wanted to be a theoretical physicist, and that's just not possible without a PhD. So I went straight from my bachelor's into a PhD program. Since I knew what I wanted to do, it didn't make sense to faff about doing something else for a few years.
Other people have different goals, or don't have the same kind of clear direction I did. I've known a number of people who've gone back to do a PhD later on, either because they wanted a career change or because they had personal ambitions to reach that level of education.
I agree with Bmore_2_UK that you should be sure that you want to do it first, but I would add that you need to clearly understand
why you want to. It's important to understand what you're getting into, not only the degree programme itself, but the job prospects and career progression you can expect afterward. One of my classmates spent 5 years on his PhD and ended up as a laser salesman, because he decided a research career wasn't a good fit for him.
I got married before I started, and although it was nightmarish at times, we're still happily married.

My husband even went back and did his master's while I was working on my PhD. No kids, though--not a chance. I've had male colleagues who've managed to have kids and get through a PhD at the same time, but it's much harder for women. (Disclaimer: I suspect that could differ among disciplines. There still aren't many women in physics.) I wish it were otherwise.
