Professor Carol Jones is in a fairly common situation for female faculty. Professor Jones, full plate or not, will probably decide to mentor postdoc Sarah.
Out of kinship, a sense of moral obligation, or simply because if-she-doesn't-do-it-who-will, I suspect Carol Jones will somehow find the time to advise Sarah. She will advise Sarah about what jobs to pursue among the various options she may have, which ones are best for her career, which ones are likely to be family-friendly, what to wear at interviews, and more importantly, what to say and what not to say. She will surely advise her to come across as a no-nonsense, enthusiastic, career-oriented scientist, who will plug away at the lab as hard as any male faculty member. And she may also advise Sarah to find out about "softer" issues from other female faculty members and postdocs.
These issues--such as maternity leave policies, the work-day culture in the prospective department (what hours do they keep?), how family-friendly the environment is--are best brought up at the second or third interview, if at all. It is unfortunate, but there are still insensitive departments with double standards for female scientists.
So what is a good mentor in science? A good mentor is one who will help a younger, less experienced scientist (in this case, a postdoc) get ahead. This is primarily done through interactions in the lab and in the classroom, and of course means helping steer the protégé through the science. Pointing a younger scientist toward a better project that is likely to result in publishable results, advice on how to master the techniques, discussion of the results, and help with writing the publications, are among some of the obvious examples of how a good mentor helps out. But there are other things a mentor can do: providing moral support when things aren't working out, letting the protégé witness or even participate in the writing of a grant, allowing a student or a postdoc to provide the first take on a review of a paper, introducing him or her to important people in the field, and yes, making some calls on his or her behalf when the protégé is looking for another postdoc or a job.
Every scientist needs many mentors at any given time. Some mentors are good for the science, the techniques, the papers, the grantsmanship, and others are good for other things, such as "where do you want to be in 5 years," what are some of the things you should be doing to position yourself better (at scientific meetings, in your lab, in your department, in your scientific society). Actually, a variety of mentors are good because you get several different opinions, and then you may be able to make a better decision or choice on your own.
Choose your variety of mentors on the basis of what you need. There should always be good chemistry between you, but good chemistry comes in a variety of flavors: Perhaps the hard-core taskmaster who will help you steer your science may not be the person you choose to have a beer with, but by gosh, your science will be first-rate. On the other hand, the nice person down the hall may not know a thing about your science, but she may be able to give you wonderful career advice.
Minority scientists are often in the same boat as Professor Carol Jones. When I was in the lab, all the Latino medical students who had to fulfill their research requirement came to me. Even the ones who were interested in something outside of my expertise came to my lab for coffee and comradeship. Many hours were spent advising and mentoring those students, and it was extremely fulfilling--although it took me away from the lab--because I felt I was really contributing to helping more minorities making it through medical school, and some through their beginnings in research. The Carol Joneses of the world do it because they want to help other female scientists "make it" in academic research.
This, of course, prompts the question: Are women and minority scientists the only ones that should be mentoring young women and minority scientists? I think not, and indeed I know not. But I think more nonminority male scientists should help--just think of how much they could contribute!
Despite what Charles Barkley, the NBA player, said once about not wanting to be a role model, we are role models and generally we take our self-imposed responsibility to help others like us very seriously.