People love to talk about themselves and their work. Tell your interviewee know why you chose him. ("I heard you were an expert.") Let him know (or think) that you think he is the greatest thing and that his opinion in the paper will be vital. Ask for his credentials, it amounts to more flattery.
Tell him what you are researching; your point of view (unless you are interviewing a hostile person, which means she might be as nice as pie but believes on the other side of you. In that case, be quiet about your stance).
Have a list of questions to ask. Make them very specific and open (no yes or no). Say stuff like, "I've found in my research that 45 fishes have been discovered in the past year. How do you think this will impact the environmental balance?" ("Hows" and "Whys" are open-ended questions.)
However, let the interviewee lead if he or she wants to. Let her talk. Some good stuff you didn't think of might come out here; remember she is the expert and knows stuff you don't.
Take notes. It is okay. It is okay to record IF YOU GET PERMISSION. But jotting down notes is fine, he will expect that. Ask if you can follow-up if you have questions and get the best contact info to do that.
After leaving the interview, stop as soon as you can and write out a freewrite about the interview. The longer you wait to do this, the more you will forget.