For my graduate classes, I have picked articles with a particular goal in mind. Before your presentation, we should briefly chat about the "spin" for each of the papers/topics.
Your audience will likely consist of two groups: those who have read the article and those who haven't. Try to include something for both. Be sure to cover the essential methods and results for those who haven't. For those who have, I recommend going somewhat beyond the article itself. Give some of the background. What was done on this topic before this paper (showing a bit of relevant data from previous papers can be good)? What key questions were unresolved prior to this paper? What is novel and interesting about this paper? What problems exist with the paper or what questions remain unresolved? What should the next step be? Can you spur a discussion among class participants?
Where possible, go low-tech. You do not need to "out-Spielberg" your classmates. Go with the simplest technology -- overheads are nice and simple, PowerPoint is easy to use (video projector available on request). I have a scanner you can use if you want to scan in figures. Demos and fancy slides can be good, but be careful. Make sure you put sufficient time into the content before you spend any time on bells & whistles. Most academics are turned off by particularly glitzy presentations. We think all those fancy PowerPoint things like snazzy backgrounds, slick transitions and excessive animations are for the business types who look for style over substance. Use these very sparingly.
On overheads or slides, keep text to a minimum. Use point form where possible.
Also, while I'm here, a pet peeve... Don't begin with a slide that says, "First I'm going to talk about the background, then I'll talk about the methods, then I'll talk about the results, then I'll discuss the implications." If your talk follows the standard recipe, you don't need to make that explicit. Although many suggest an overview slide, my own opinion is that it's only useful if your outline is unconventional, you have several themes within your talk, or you have your sections organized in such a way that the logic isn't obvious to the listener without an overview slide.
If you giving a formalish talk in a room that's unfamiliar to you, go in early and figure out how the equipment is set up. Does your computer hook up and work with the projection system? Do you know how the overhead projector works? Where are the room lights? How does the laser pointer work? The worst two talks I've ever given have occurred when I got utterly discombobulated because the equipment wasn't working the way I thought it would.
Practice. If the talk is a more formal and important one (e.g., Psychobiology), ask your colleagues, friends and advisor to sit in on a practice session. I find practice is particularly important for the first five minutes to get me through the brain-dead butterflies stage. I often script and rehearse the first few minutes and after that, I am comfortable speaking without a script.
If you are so nervous that your hand is shaking, you can compensate. If you point with a pen to an overhead, your jitters will appear exaggerated. Instead, point with a stick or your hand to the projection screen. Movements made with your large arm muscles are less jittery than those made with your hands. Similarly, if using a laser pointer, brace your upper arm against your torso to stabilize it. Have water handy in case your mouth gets dry from nervousness.
If you're using a laser pointer, use it very sparingly. Push the button only when you intend to point something out. Do not hold it down while you wave it around or your viewer's attention will be engaged there, like with an annoying fly buzzing around.
Don't leave preparation till the last minute. Read the article(s) ~a week before so you have some time to mull them over before you prepare your presentation. Get a good night's sleep.
Don't stress. Everyone has been nervous when speaking in public and even some highly experienced and respected professors still get nervous when speaking in front of a class or crowd.